GenCost 2020-21 Final report Paul Graham, Jenny Hayward, James Foster and Lisa Havas June 2021 Australia’s National Science Agency GenCost 2020-21 | i Citation Graham, P., Hayward, J., Foster J. and Havas, L. 2021, GenCost 2020-21: Final report, Australia. Copyright © Commonwealth Scientific and Industrial Research Organisation 2021. To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO. Important disclaimer CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. CSIRO is committed to providing web accessible content wherever possible. If you are having difficulties with accessing this document please contact csiroenquiries@csiro.au. ii | CSIRO Australia’s National Science Agency Contents Acknowledgements ......................................................................................................................... vi Executive summary ........................................................................................................................ vii 1 Introduction ...................................................................................................................... 10 1.1 Scope of the GenCost project and reporting....................................................... 10 1.2 CSIRO and AEMO roles ........................................................................................ 10 1.3 Incremental improvement and focus areas ........................................................ 11 1.4 Overview of consultation draft feedback and responses ................................... 11 2 Current technology costs .................................................................................................. 16 2.1 Current cost definition ........................................................................................ 16 2.2 Updates to current costs ..................................................................................... 16 2.3 Current generation technology capital costs ...................................................... 17 2.4 Current storage technology capital costs ............................................................ 17 3 Scenario narratives and data assumptions ....................................................................... 20 3.1 Scenario narratives .............................................................................................. 20 4 Projection results .............................................................................................................. 34 4.1 Global generation mix ......................................................................................... 34 4.2 Changes in capital cost projections ..................................................................... 35 4.3 Hydrogen electrolysers ........................................................................................ 51 5 Levelised cost of electricity analysis ................................................................................. 53 5.1 Overview of the new method .............................................................................. 54 5.2 LCOE estimates .................................................................................................... 55 Global and local learning model .......................................................................... 63 Data tables ........................................................................................................... 66 Shortened forms ........................................................................................................................... 78 References ............................................................................................................................. 81 GenCost 2020-21 | iii Figures Figure 2-1 Comparison of current cost estimates with previous work ........................................ 17 Figure 2-2 Capital costs of storage technologies in $/kWh (total cost basis) ............................... 18 Figure 2-3 Capital costs of storage technologies in $/kW (total cost basis) ................................. 19 Figure 3-1 Projected EV sales share under the Central scenario .................................................. 26 Figure 3-2 Projected EV adoption curve (vehicle sales share) under the High VRE scenario ....... 27 Figure 3-3 Projected EV sales share under the Diverse Technology scenario .............................. 27 Figure 3-4 Adoption curves for hydrogen technologies under the Central scenario ................... 28 Figure 3-5 Adoption curves for hydrogen technologies under the High VRE scenario ................ 29 Figure 3-6 Adoption curves for hydrogen technologies under the Diverse Technology scenario 29 Figure 4-1 Projected global electricity generation mix in 2030 and 2050 by scenario ................ 34 Figure 4-2 Projected capital costs for black coal supercritical by scenario compared to 2019-20 projections .................................................................................................................................... 36 Figure 4-3 Projected capital costs for black coal with CCS by scenario compared to 2019-20 projections .................................................................................................................................... 37 Figure 4-4 Projected capital costs for gas combined cycle by scenario compared to 2019-20 projections .................................................................................................................................... 38 Figure 4-5 Projected capital costs for gas with CCS by scenario compared to 2019-20 projections ....................................................................................................................................................... 39 Figure 4-6 Projected capital costs for gas open cycle (small) by scenario compared to 2019-20 projections .................................................................................................................................... 40 Figure 4-7 Projected capital costs for nuclear SMR by scenario compared to 2019-20 projections ....................................................................................................................................................... 41 Figure 4-8 Projected capital costs for solar thermal with 8 hours storage by scenario compared to 2019-20 projections .................................................................................................................. 42 Figure 4-9 Projected capital costs for large scale solar PV by scenario compared to 2019-20 projections .................................................................................................................................... 43 Figure 4-10 Projected capital costs for rooftop solar PV by scenario compared to 2019-20 projections .................................................................................................................................... 44 Figure 4-11 Projected capital costs for onshore wind by scenario compared to 2019-20 projections .................................................................................................................................... 45 Figure 4-12 Projected capital costs for offshore wind by scenario compared to 2019-20 projections .................................................................................................................................... 46 Figure 4-13 Projected total capital costs for 2 hour duration batteries by scenario (battery and balance of plant) ........................................................................................................................... 47 iv | CSIRO Australia’s National Science Agency Figure 4-14 Projected capital costs for pumped hydro energy storage (12 hours) by scenario .. 48 Figure 4-15 Projected technology capital costs under the Central scenario compared to 2019-20 projections .................................................................................................................................... 49 Figure 4-16 Projected technology capital costs under the High VRE scenario compared to 2019- 20 projections ............................................................................................................................... 50 Figure 4-17 Projected technology capital costs under the Diverse Technology scenario compared to 2019-20 projections ................................................................................................ 51 Figure 4-18 Projected technology capital costs for alkaline and PEM electrolysers by scenario . 52 Figure 5-1 Three types of electricity system models .................................................................... 55 Figure 5-2 Range of generation and storage capacity deployed in 2030 across the 9 weather year counterfactuals ..................................................................................................................... 57 Figure 5-3 Levelised costs of achieving 50%, 60%, 70%, 80% and 90% variable renewable energy shares in the NEM, NSW, VIC and QLD in 2030 ............................................................................ 58 Figure 5-4 Calculated LCOE by technology and category for 2020 ............................................... 61 Figure 5-5 Calculated LCOE by technology and category for 2030 ............................................... 61 Figure 5-6 Calculated LCOE by technology and category for 2040 ............................................... 62 Figure 5-7 Calculated LCOE by technology and category for 2050 ............................................... 62 Apx Figure A.1 Schematic of changes in the learning rate as a technology progresses through its development stages after commercialisation .............................................................................. 64 Tables Table 3-1 Scenarios and their key drivers ..................................................................................... 21 Table 3-2 Assumed technology learning rates under all scenarios .............................................. 23 Table 3-3 Assumed utility scale energy storage learning rates by scenario ................................. 25 Table 3-4 Hydrogen demand assumptions by scenario ................................................................ 30 Table 3-5 Renewable resource limits on generation in TWh in the year 2050. NA means the resource is greater than projected electricity demand. ............................................................... 31 Table 3-6 Assumed gas prices in $A/GJ ........................................................................................ 32 Table 3-7 Assumed black coal prices in $A/GJ .............................................................................. 33 Table 3-8 Assumed global oil price in $A/bbl ............................................................................... 33 Apx Table B.1 Current and projected generation technology capital costs under the Central scenario ......................................................................................................................................... 67 GenCost 2020-21 | v Apx Table B.2 Current and projected generation technology capital costs under the High VRE scenario ......................................................................................................................................... 68 Apx Table B.3 Current and projected generation technology capital costs under the Diverse Technology scenario ..................................................................................................................... 69 Apx Table B.4 One and two hour battery cost data by storage duration, component and total costs .............................................................................................................................................. 70 Apx Table B.5 Four and eight hour battery cost data by storage duration, component and total costs .............................................................................................................................................. 71 Apx Table B.6 Pumped hydro storage cost data by duration, all scenarios, total cost basis ....... 72 Apx Table B.7 Storage cost data by source, total cost basis ......................................................... 73 Apx Table B.8 Data assumptions for LCOE calculations ................................................................ 74 Apx Table B.9 Electricity generation technology LCOE projections data, 2020-21 $/MWh ......... 76 Apx Table B.10 Hydrogen electrolyser cost projections by scenario and technology, 2020-21 $/kW .............................................................................................................................................. 77 vi | CSIRO Australia’s National Science Agency Acknowledgements This final report consolidates input received from stakeholders on the GenCost 2020-21 Consultation Draft which, in partnership with AEMO, was released for public consultation in December 2020. Submissions on the consultation draft were provided to AEMO and CSIRO in the period up to 1 February 2021. We are also grateful for feedback received via a webinar with stakeholders on 3 March 2021 and in various personal communications to the GenCost team. GenCost 2020-21 | vii Executive summary GenCost is a collaboration between CSIRO and AEMO to deliver an annual process of updating electricity generation and storage costs with a strong emphasis on stakeholder engagement. This is the third update following the inaugural report in 2018 and a second report in 2019-20. The 2020-21 report incorporates updated current capital cost estimates commissioned by AEMO and delivered by Aurecon. Based on these updated current capital costs, the report provides projections of future changes in costs consistent with updated global electricity scenarios. Levelised costs of electricity (LCOEs) are also included and provide a summary of the relative competitiveness of generation technologies. Capital cost projections The projection methodology is grounded in a global electricity generation and capital cost projection model recognising that cost reductions experienced in Australia are largely a function of global technology deployment. Three scenarios are explored:  Central: Current stated global climate polices (as of late 2020), with the most likely assumptions for all other factors such as renewable resource constraints  High VRE: A world that is driving towards net zero emissions by 2050 and where technical, social and political support for variable renewable electricity generation is high  Diverse Technology: A world where most developed countries are striving for net zero emissions by 2050 but others are lagging such that global net zero emissions is reached by 2070. Furthermore, there is lack of social, technical and political support for variable renewable electricity generation and subsequently a greater role for other technologies. In the High VRE scenario, global non-hydro renewable generation reaches a share of 82% by 2050, with the majority sourced from solar photovoltaic (PV) and on- and off-shore wind. In the Diverse Technology and Central scenarios, global wind and solar PV shares are lower at around 50% in total. Access to wind and solar PV is assumed to be constrained in the Diverse Technology scenario. Consequently, generation from gas and coal with carbon capture and storage is deployed to meet the climate policy ambitions of that scenario. CCS is also used more commonly in hydrogen production. Nuclear small modular reactors could also play a role in the Diverse Technology scenario from 2030 so long as investors are willing to drive down costs through multiple deployments in the late 2020s. Battery costs fell the most in 2020-21 compared to any other generation or storage technology and are projected to continue to fall. We have also adjusted assumptions to recognise that batteries are achieving longer lives. Falling battery storage costs underpin the long-term competitiveness of variable renewables. Pumped hydro is also an important storage technology and is more competitive than batteries when longer duration storage is required. The technology cost projections have been extended to include hydrogen electrolysers reflecting strong interest in this technology that, combined with low cost renewable generation could potentially underpin a low emission hydrogen fuel industry for export or Australian domestic viii | CSIRO Australia’s National Science Agency consumption. The results indicate that substantial cost reductions are expected over the next few decades, with many demonstration projects underway worldwide. Levelised cost of electricity There have been concerns for many years that it is difficult to quantify the additional costs associated with variable renewable electricity generation. Traditional approaches to calculating the levelised cost of electricity fail to include these additional costs, underestimating the full costs to the electricity system. The GenCost team has been seeking to address this issue since the first report in 2018 where we outlined this problem and reviewed a number of alternative solutions. To calculate the additional costs CSIRO constructed an electricity system model that can calculate the required additional investment considering any existing resources in the system. The key additional investments required are in:  New transmission to access Renewable Energy Zones  Additional transmission to strengthen the grid so that dispersed renewable generation can reach key demand centres and expanded state interconnection so that connecting regions can provide more support for one another when renewable generation is low in one or more regions  Synchronous condensers to support system security  Battery and pumped hydro storage to meet demand during low renewable generation periods. ES Figure 0-1 Calculated LCOE by technology and category for 2030 The required amount of additional investment depends on the amount or share of variable renewable energy (VRE) generated. We calculated the additional costs of variable renewable generation for VRE shares from 50% to 90%1 for the National Electricity Market (NEM). We found 1 90% is about as high as variable renewable deployment is likely to need to go as increasing it further would result in the undesirable outcome of shutting down existing non-variable renewable generation from biomass and hydroelectric sources. 0 50 100 150 200 250 300 350 400 Gas turbine small Gas turbine large Gas reciprocating Black coal Brown coal Gas Black coal Brown coal Gas Black coal with CCS Brown coal with CCS Gas with CCS Solar thermal 8hrs Nuclear (SMR) Biomass (small scale) Wind Solar PV 50% VRE share 60% VRE share 70% VRE share 80% VRE share 90% VRE share Climate policy risk premium Standalone Wind & solar PV combined Peaking 20% load Flexible 40-80% load, high emission Flexible 40-80% load, low emission Variable Variable with integration costs 2020-21 A$/MWh GenCost 2020-21 | ix that the additional costs to support a combination of solar PV and wind generation in 2030 is estimated at between $6 to $19/MWh depending on the VRE share and region of the NEM. These represent a maximum of costs across nine weather years over which the costs were estimated. When added to variable renewable generation costs and compared to other technology options, these new estimates indicate that wind and solar PV remain the lowest cost new-build technology up to a 60% VRE share. The closest technology is the low range cost of a gas combined cycle generator which can match the costs of variable renewables with integration costs at a 70% or greater share. However, the low range 2030 gas combined cycle cost assumptions will be challenging to achieve. It requires no climate policy risk at the financing stage (despite the 25 year design life extending beyond the net zero emission targets of most states), a gas price just below $6/GJ throughout that period and a capacity factor of 80% in a system with 70% or greater share of energy from near-zero marginal cost renewables. 10 | CSIRO Australia’s National Science Agency 1 Introduction Current and projected electricity generation and storage technology costs are a necessary and highly impactful input into electricity market modelling studies. Modelling studies are conducted by the Australian Energy Market Operator (AEMO) for planning and forecasting purposes. They are also widely used by electricity market actors to support the case for investment in new projects or to manage future electricity costs. Governments and regulators require modelling studies to assess alternative policies and regulations. There are substantial coordination benefits if all parties are using similar cost data sets for these activities or at least have a common reference point for differences. The report provides an overview of updates to current costs in Section 2. This section draws significantly on updates to current costs provided in Aurecon (2020) and further information can be found in their report. The global scenario narratives and data assumptions for the projection modelling are outlined in Section 3. Capital cost projection results are reported in Section 4 and LCOE results in Section 5. CSIRO’s cost projection methodology is discussed in Appendix A. Appendix B provides data tables for those projections which can also be downloaded from CSIRO’s Data Access Portal2. 1.1 Scope of the GenCost project and reporting The GenCost project is a joint initiative of the CSIRO and AEMO to provide an annual process for updating electricity generation and storage cost data for Australia. The project is committed to a high degree of stakeholder engagement as a means of supporting the quality and relevancy of outputs. Each year a consultation draft (Graham et al., 2020) is released for feedback before the final report is completed. The project is flexible about including new technologies of interest or, in some cases, not updating information about some technologies where there is no reason to expect any change, or if their applicability is limited. GenCost does not seek to describe the set of electricity generation and storage technologies included in detail. 1.2 CSIRO and AEMO roles AEMO and CSIRO jointly fund the GenCost project by combining their own in-kind resources. AEMO commissioned Aurecon to provide an update of current electricity generation and storage cost and performance characteristics (Aurecon, 2021). Earlier drafts of Aurecon’s report were initially shared with of stakeholders during a webinar in September 2020 and as part of the December 2020 to February 2021 public consultation. 2 Search GenCost at https://data.csiro.au/collections GenCost 2020-21 | 11 Project management, workshops, capital cost projections (presented in Section 4) and development of this report are primarily the responsibility of CSIRO. 1.3 Incremental improvement and focus areas There are many assumptions, scope and methodological considerations underlying electricity generation and storage technology cost data. In any given year, we are readily able to change assumptions in response to stakeholder input. However, the scope and methods may take more time to change, and input of this nature may only be addressed incrementally over several years, depending on the priority. In this report, we have improved our approach to calculating Levelised Costs of Electricity (LCOE) for renewables by employing a new modelling approach which is able to calculate additional costs to the system associated with variable renewable generation. 1.4 Overview of consultation draft feedback and responses There was a strong level of interest and broad range of feedback provided on the December 2020 consultation draft. We summarise the feedback and how they have been addressed in the final report under the following themes. 1.4.1 Technology learning rates Technology learning rates are an important input into the cost projections. They are based on historical data or on learning rates achieved by technologies at a similar stage of development. They determine the cost reduction for each doubling of cumulative capacity deployed. Stakeholders proposed that there was evidence for a higher learning rate for offshore wind and a wider range of learning rates for batteries. We also found that there was evidence to support a stronger offshore wind learning rate and increased it to 15%. We also agreed that the range of battery projections across the scenarios was too narrow. We modified the scenarios so that instead of using a single rate we use a range of 7.5 to 15% across the scenarios. We also extended these learning rates to the balance of plant. There were also suggestions for a stronger learning rate for hydrogen electrolysers and to consider including assumptions about improvements in the round-trip efficiency of batteries over time. While these suggestions might be plausible, we found no strong basis upon which to base any assumptions and so the approach is unchanged. 1.4.2 Scope of technologies included It was requested that the GenCost project include two additional storage technologies: compressed air energy storage and non-lithium based batteries. At present, such projects have very limited deployment but are proposed. We are unable to accommodate new technologies in 2020-21 but will consider these for future inclusion. We will continue to trade off completeness against the cost to the project of updating a larger technology list. 12 | CSIRO Australia’s National Science Agency 1.4.3 Operating and maintenance costs Stakeholders advised that operating and maintenance costs were too high for batteries and nuclear SMR. For batteries, an additional issue was the relationship between battery operating costs and storage size. These have been revised in Aurecon (2021) to take account of these issues. For nuclear SMR, it was found that prior variable operating and maintenance cost assumptions could not be supported and have been revised downwards to $5.3/MWh. 1.4.4 Capacity factors There is a significant amount of confusion around capacity factors amongst stakeholders partly because they are used and reported in different ways by CSIRO and AEMO. In its various versions of the ISP input and assumptions workbooks, AEMO publishes medium and high average capacity factors as an indicator of the quality of wind and solar resources in Renewable Energy Zones. However, when conducting modelling they use a full half hourly production trace and vary the trace over future years by looping through “reference years”. For GenCost, the capacity factors we report are the range we use in our levelised cost of electricity (LCOE) calculations. The purpose of the LCOE calculations is to show the best and worst case for future new-build projects. Some stakeholders were disappointed that we had not included higher capacity factors for coal and nuclear technologies. However, historical evidence shows that coal generation is not achieving better than 80% capacity factor and most much lower. We therefore use 80% as a best case. 60% is the worst case because we assume a new plant, if built, will be more competitive than existing coal generators. Above 80% is not realistic in a system which is including more renewables (state renewable policies ensure this outcome will occur). For renewable capacity factors the approach is similar but the outcome different. We again look to the historical range of capacity factors achieved but we consider that newer plant can increase their capacity factor due to improvements in the technology. This is particularly the case of wind where larger turbines are better at capturing wind energy throughout a greater proportion of the day. We have therefore not made any changes to our capacity factor assumptions but hope this background is useful in understanding the basis of their selection. 1.4.5 New entrant versus existing coal generation costs Some stakeholders were concerned that our assumed coal costs were higher than what existing plant pay for coal. This is correct but consistent with our approach. GenCost is only concerned with the cost of new-build plant. New-build plant will have to compete with export markets to establish a coal supply and will therefore face higher prices. However, it should be noted that electricity modelling processes do include existing plant and lower coal costs for existing plant are considered in those processes. A breakdown of costs for existing plant are available in background documents to various AEMO planning processes and therefore there are no plans to duplicate that information in the GenCost project. GenCost 2020-21 | 13 1.4.6 Externalities There was a concern that generation technologies may impose external costs on the community that are not accounted for in the LCOE calculations or elsewhere. This is mostly correct by design. We wish to compare technologies on a common basis. Including each individual technology’s set of externalities would make the comparability low and somewhat arbitrary. We therefore try to keep to costs inside the plant gate. There are two exceptions we have included: the cost of climate policy risk on financing and the integration costs of variable renewables. We made these exceptions because they are too large and direct to ignore. We have no plans to extend to other issues but will continue to review this topic as issues emerge. 1.4.7 Design and technical life Design life is the period which tends to match the life of the initial financing and requires no additional capital costs or refurbishment (that is not already included in operating and maintenance costs). The operating life is longer and represents the full asset life inclusive of refurbishments. Stakeholders felt that the design life was too short for coal, pumped hydro and batteries. For coal, we acknowledge that a government might decide to finance a plant over a longer period, but we believe 30 years remains appropriate for design life for standard financing arrangements. The design life is used in LCOE calculations to determine the annualised cost of the capital. The operating life is of course longer (around 50 years if not retired for economic reasons), and operating lives are used in electricity system modelling. The two lifetime definitions are used for different purposes. For batteries and pumped hydro, we found the proposal for a longer design life was supported. Batteries are being more commonly provided with 20-year warranties and we have shifted to this assumption based on Aurecon (2021). However, this comes with an additional 20-year warranty cost3. Note, the 2019-20 assumptions were a 20-year project design life but with batteries completely replaced every 10 years. That battery replacement is no longer necessary. For pumped hydro, we found local experience and international studies do tend to use a longer design life and so this has been changed to 40 years. 1.4.8 Battery cost basis relative to usable capacity Stakeholders were understandably surprised that we had assumed a 100% depth of discharge and charge. Typically, batteries cannot be fully discharged or charged without degrading the battery. That remains true, however what we had failed to emphasise in the past is that Aurecon provides their current cost estimates for batteries on a usable-part-of-the-battery basis. To put it another way, the nameplate capacity of the plant is lower because it only represents the usable part of the battery. Therefore, when applying the battery costs in Aurecon (2021) and in GenCost, there is no need to apply further assumptions about maximum depth of discharge or charge. 3 Not reported here, see Aurecon (2021). 14 | CSIRO Australia’s National Science Agency 1.4.9 Implementation of global emissions reduction policies The global modelling requires the implementation of current and potential new policies to meet a range of global climate policy outcomes that are explored across the scenarios. Carbon prices are the most unbiased way in which to determine how meeting different emissions outcomes might impact global electricity generation technology choices. For this reason, we had tended to use stronger carbon prices than contemporary work which relies on a broader mix of policies. After further consideration we have decided to align our approaches more closely with the International Energy Agency, directly implementing their estimated carbon prices and using a mix of other policies (such as renewable energy targets) to match our electricity sector emissions outcomes to theirs. Some stakeholders also considered that the inclusion of carbon prices in the LCOE calculations meant that the technology comparisons were invalid or biased towards low emission technologies. To be clear, wherever we had imposed a carbon price on a high emission technology we always included another measure of the cost without a carbon price and any conclusions were always based on the non-carbon price data. However, given some confusion has arisen and that the carbon price assumptions have minor impact on the relative competitiveness of technologies, we have removed carbon prices from the LCOE analysis. 1.4.10 Current costs – high rate of deployment Tracking the costs for technologies that continue to be deployed while this report is in draft often leads to some stakeholders identifying differences in observed current costs. This is the case for solar and batteries. These two technologies have undergone minor revisions to reflect updated data. An additional issue raised around solar is that project sizes were being decreased closer to deployment to avoid extra costs associated with addressing system strength impacts. While we have not changed our standard size assumptions to reflect this it is a useful explanation for why some projects may not reflect the current costs reported here. The size fluctuation and its impact on costs will likely be an intermittent issue as renewable deployment sites both open up or become congested. 1.4.11 Current costs – low or no deployment The current costs for immature technologies are often fraught because there is often no local deployment to provide a basis for cost estimates. Stakeholders were able to provide additional data on biomass generation projects and this has supported a 16% reduction in capital cost compared to the consultation draft. We also have had a range of feedback into the assumed current costs for nuclear SMR over several years. Our current cost estimate is from GHD (2018). The basis of this estimate is the International Energy Agency and Nuclear Energy Association report Projected costs of electricity generation 2015. That report proposed that nuclear SMR typically costs 50% to 100% more than large scale nuclear. If we use the 100% value for Australia (because it has no experience in nuclear generation) and a 0.7 US$/A$ exchange rate the outcome matches the GHD (2018) estimate of $16,000/kW. If we update this number using more recent large-scale nuclear cost estimates and GenCost 2020-21 | 15 exchange rates, this estimate is not significantly different. It is also consistent with the higher end of more recent first of a kind estimate from EFWG (2019). However, a major source of discomfort for stakeholders is that this high cost estimate is of theoretical value only. A nuclear SMR plant is not planned to be built in Australia anytime soon. It is more likely that Australia would only take up nuclear SMR, if at all, from around 2030, after other countries have brought down the cost. It is this future cost level that stakeholders would prefer to focus on. GenCost’s projections of the after-learning future cost of nuclear SMR in the 2030s has not received significant feedback and is consistent with “nth of a kind” estimates such as EFWG (2019). On this basis, we will no longer be reporting nuclear SMR current costs before 2030. 1.4.12 Renewable integration costs Stakeholders had two broad items of feedback on the new method for calculating the integration costs of variable renewables. One is that they would like to see more information presented on the existing generation, storage and transmission resources assumed to be in the system before higher variable renewable shares are imposed. The other is that they would like to see more weather years included to account for variability in renewable supply. Both items have been addressed and were included in the revised results included in Section 5. 16 | CSIRO Australia’s National Science Agency 2 Current technology costs 2.1 Current cost definition Our preferred definition of current costs are the costs that have been demonstrated to have been incurred for projects completed in the current financial year (or within a reasonable period before). We do not wish to include in our definition of current costs, costs that represent quotes for delivery of projects in future financial years or project announcements. While all data is useful in its own context, our preference reflects the objective that the data must be suitable for input into electricity models. The way most electricity models work is that investment costs are incurred either before (depending on construction time assumptions) or in the same year as a project is available to be counted as a new addition to installed capacity4. Hence, current costs and costs in any given year must reflect the costs of projects completed in that year. Quotes received now for projects to be completed in future years are only relevant for future years. For technologies that are not frequently being constructed, the preference is to look overseas at the most recent projects constructed. This introduces several issues in terms of different construction standards and engineering labour costs which have been addressed by Aurecon (2021). Aurecon (2021) also provide more detail on specific definitions of the scope of cost categories included. 2.2 Updates to current costs AEMO commissioned Aurecon (2021) to provide an update of current cost and performance data for existing and selected new electricity generation and storage technologies. This data is used in this report as the starting point for projections of capital costs to 2050 and for calculations of the levelised cost of electricity. Compared to 2019-20, Aurecon has reviewed coal generation and included two gas open cycle unit sizes. CSIRO has updated costs for technologies which are more rarely deployed such as tidal/current and wave energy. Aurecon (2021) has included hydrogen electrolysers for the first time and these are separately reported. Pumped hydro has also not been updated by Aurecon (2021), but we have revised this data to be consistent with AEMO’s ISP 2020 which received further input from stakeholders on this technology. 4 This is not strictly true of all models but is most true of long-term investment models. In other models, investments costs are converted to an annuity (adjusted for different economic lifetimes) or additional capital costs may be added later in a project timeline for replacement of key components. GenCost 2020-21 | 17 2.3 Current generation technology capital costs Figure 2-1 provides a comparison of current (2020-21) cost estimates (drawing primarily on the Aurecon (2021) update) for electricity generation technologies with the four most recent previous reports: GenCost 2019-20, GenCost 2018, Hayward and Graham (2017) (also CSIRO) and CO2CRC (2015) which we refer to as APGT (short for Australian Power Generation Technology report). All costs are expressed in real 2020-21 Australian dollars and represent overnight costs since it would not be possible to build and financially close projects before July 2021. CSIRO’s estimate for 2020-21 rooftop solar PV cost is included in the “Aurecon/CSIRO” data as that technology was not part of Aurecon (2021). Rooftop solar PV costs are before subsidies from the Small-scale Renewable Energy Scheme. All data has been adjusted for inflation. Figure 2-1 Comparison of current cost estimates with previous work Coal generation capital costs have been revised upwards after not being significantly updated since the GHD (2018) analysis. The lack of Australian construction means there will always be a range of interpretations when converting overseas data to Australia. Solar thermal costs have increased on 2019-20 estimates reflecting inclusion of a first of a kind cost premium. Gas, wind and solar PV cost estimates have been relatively stable reflecting better data availability for Australian projects. 2.4 Current storage technology capital costs Updated and previous capital costs are provided on a total cost basis for various durations of battery and pumped hydro energy storage (PHES) in $/kW and $/kWh. Total cost basis means that 0 2000 4000 6000 8000 10000 12000 14000 16000 Brown coal, pf Brown coal with CCS Black coal, pf Black coal with CCS Gas combined cycle Gas with CCS Gas open cycle (large) Solar thermal (CSP) Rooftop PV Large scale PV Wind 2020-21 $/kW APGT 2015 CSIRO 2017 GHD/CSIRO 2018 Aurecon/CSIRO 2019-20 Aurecon/CSIRO 2020-21 18 | CSIRO Australia’s National Science Agency the costs are calculated by taking the total project costs divided by the capacity in kW or kWh5. As the storage duration of a project increases then more batteries or larger reservoirs need to be included in the project, but the power components of the storage technology remain constant. As a result, $/kWh costs tend to fall with increasing storage duration (Figure 2-2). The downward trend flattens somewhat with batteries since its power component, mostly inverters, is relatively small but adding more batteries is costly. However, the hydro turbine in a PHES project is a large capital expense while adding more reservoir is less costly. As a result, PHES costs fall steeply with more storage duration. Conversely, the costs in $/kW increase as storage duration increases because additional storage duration adds costs without adding any power rating to the project (Figure 2-3). Additional storage duration is most costly for batteries. These trends are one of the reasons why batteries tend to be more competitive in low storage duration applications, while PHES is more competitive in high duration applications. A combination of battery and pumped hydro with different durations may be required depending on the behaviour of other generation in the system, particularly the scale of variable renewable generation (see Section 5). Figure 2-2 Capital costs of storage technologies in $/kWh (total cost basis) Round trip efficiency, project design life and the potential for co-location also play a role in competitiveness of alternative storage technologies. Depth of discharge in batteries is also a relevant factor. However, all Aurecon battery costs are on a usable capacity basis such that depth 5 Component costs basis is when the power and storage components are separately costed and must be added together to calculate the total project cost. 0 200 400 600 800 1000 1200 Battery (1hr) Battery (2hrs) Battery (4hrs) PHES (6hrs) Battery (8hrs) PHES (12hrs) PHES (24hrs) PHES (48hrs) PHES (48hrs) Tasmania 2020-21 $/kWh Aurecon 2019-20 Aurecon 2020-21 GenCost 2019-20 AEMO ISP December 2020 GenCost 2020-21 | 19 of discharge is 100%. Aurecon (2021) also includes estimates of battery costs when they are integrated within an existing power plant and can share some of the power conversion technology. This results in a 9% lower battery cost for a 1-hour duration battery, scaling down to a 2.5% cost reduction for 8 hours duration. PHES is more difficult to co-locate. Battery current costs have declined in Aurecon (2021) compared to their previous work. These are based on projects deployed. In contrast, we have increased PHES costs by aligning with AEMO ISP December 2020 estimates. Feedback received in 2020 indicated that PHES was under-estimated in GenCost 2019-20. A new higher data point was included in the December 2020 ISP inputs and assumptions workbook based on submissions and discussion with proponents and reputable consultants with experience in PHES deployments. Some escalation in costs is consistent with major infrastructure projects where cost increases occur after initial estimates. However, we have also added a separate category for Tasmania PHES with 48hrs duration. This area of Australia has had the most detailed analysis undertaken of its PHES costs and, consistent with ISP regional cost adjustments, warrants greater certainty that it can achieve lower project cost estimates. Figure 2-3 Capital costs of storage technologies in $/kW (total cost basis) 0 1000 2000 3000 4000 5000 6000 Battery (1hr) Battery (2hrs) Battery (4hrs) PHES (6hrs) Battery (8hrs) PHES (12hrs) PHES (24hrs) PHES (48hrs) PHES (48hrs) Tasmania 2020-21 $/kW Aurecon 2019-20 Aurecon 2020-21 GenCost 2019-20 AEMO ISP December 2020 20 | CSIRO Australia’s National Science Agency 3 Scenario narratives and data assumptions 3.1 Scenario narratives The global climate policy ambitions for the Central, High VRE and Diverse Technology scenarios have been adopted from the International Energy Agency’s 2020 World Energy Outlook (IEA WEO 2020) scenarios matching to the Stated Policies scenario, Net Zero Emission by 2050 and Sustainable Development Scenario respectively. Other elements, such as the degree of vehicle electrification and hydrogen production, are also consistent with IEA WEO 2020. However, we also include other topics in our scenarios such as renewable resource constraints and the social and political acceptance and technical performance of renewables. 3.1.1 Central The Central scenario applies a 2.7 degrees consistent climate policy (using the carbon prices and other climate policies implemented by the IEA6). This represents 2020 climate and renewable energy policy commitments with no extension beyond current targets7. This implies that current 2030 Paris Nationally Determined Commitments are met but that the planned ramping up of ambition to prevent a greater than 2 degrees increase in temperature does not occur. There are moderate constraints applied with respect to global renewable energy resources (based on currently available information). Technical approaches for managing balancing of variable renewable electricity are based on current technology. Demand growth is moderate with moderate electrification of transport. 3.1.2 High VRE Under the High VRE scenario there is a strong climate policy consistent with maintaining temperature increases of 1.5 degrees and achieving net zero emissions by 2050 worldwide. Reflecting the low emission intensity of predominantly renewable electricity supply there is an emphasis on energy efficiency and high electrification across sectors such as transport, hydrogenbased industries and buildings leading to high electricity demand. Renewable energy resources are less constrained (both physically and socially) and balancing variable renewable electricity is less technically challenging. 6 Carbon trading markets exist in major greenhouse gas emitting regions overseas at present and are a favoured approach to global climate policy modelling because they do not introduce any technological bias. The IEA also includes a broad range of additional policies such as renewable energy targets and planned closure of fossil-based generation. We include these as well but cannot completely match the IEA implementation because of model structural differences. We align our own implementation of non-carbon price policies to ensure we match the emission outcomes in the relevant IEA scenario. 7 To be consistent with the IEA World Energy Outlook 2020, this does not include more recent announcements or changes of government since the IEA report was complete. For example, the WEO 2020 includes China’s 2060 net zero emissions pledge in its sustainable development scenario which we use for Diverse Technology but does not include recent announcements by Japan and South Korea, nor change of leadership in the United States. See Annex B of WEO 2020. GenCost 2020-21 | 21 3.1.3 Diverse Technology The Diverse Technology scenario assumes that physical and social constraints mean that access to variable renewable energy resources is more limited in most regions of the world. Governments subsequently limit their renewable targets below the threshold required for major deployment of balancing solutions. Consequently, there is a greater reliance on non-renewable technologies and a carbon price consistent with a 1.65 degrees climate policy ambition provides the investment signal necessary to deploy these technologies. Developed countries are still largely aiming for net zero emissions by 2050 but other countries are lagging such that worldwide net zero emissions are not achieved until 2070. Hydrogen trade (based mainly on gas with CCS and alkaline electrolysis) is relatively high allowing some regions with energy or CO2 storage resource limitations to access a low emission imported fuel. Table 3-1 Scenarios and their key drivers Key drivers High VRE Diverse Technology Central IEA WEO 2020 scenario alignment Net zero emission by 2050 Sustainable development scenario Stated policies scenario CO2 pricing / climate policy Consistent with 1.5 degrees world, not requiring negative abatement technologies Consistent with 1.65 degrees world (or 1.5 if negative abatement technologies deployed by 2070) Consistent with 2.7 degrees world Renewable energy targets and forced builds / accelerated retirement High (reflecting confidence in VRE) RE policies go to no more than 40% Current RE policies Demand / Electrification High Medium Medium Learning rates1 Higher for longer in solar and batteries Normal maturity path Higher for longer in solar and batteries Renewable resource & other renewable constraints Less constrained More constrained than existing assumptions Existing constraint assumptions2 Constraints around stability and reliability of variable renewables New low-cost solutions Conventional solutions but less demand for them Conventional solutions Decentralisation Less constrained rooftop solar photovoltaics (PV) More constrained rooftop solar PV constraints2 Existing rooftop solar PV constraints2 1 The learning rate is the potential change in costs for each doubling of cumulative deployment, not the rate of change in costs over time. In a normal maturity path, learning rates fall over time as per Apx Figure A.1. 2 Existing large-scale and rooftop solar PV renewable generation constraints are as shown in Table 3-5. 22 | CSIRO Australia’s National Science Agency 3.1.4 Scenario design considerations The GenCost scenarios are described in general in Table 3-1 and expanded on in the sub-sections below. The scenario drivers are based on the themes identified by stakeholders at a workshop in August 2019, together with insights from the modelling team on what would most likely deliver a broad range of technology cost outcomes. We acknowledge that there are potential wild card events that are not included in the scenarios such as completely new technologies and inter-regional high voltage interconnection. However, we chose to exclude wild cards. We also considered the possibility of aligning scenarios with other globally recognised scenarios. However, we found that drivers for other scenarios were not well targeted at producing changes in technology outcomes. In particular, experience has shown that climate change policy drivers alone do not result in major differences in technology adoption. 3.1.5 Technologies and learning rates As we explain further in Appendix A, we use two global and local learning models (GALLM). One is of the electricity sector (GALLME) and the other of the transport sector (GALLMT). GALLME projects the future cost and installed capacity of 31 different electricity generation and energy storage technologies. Where appropriate, these have been split into their components of which there are 48. Components have been shared between technologies; for example, there are two carbon capture and storage (CCS) components – CCS technology and CCS construction – which are shared among all CCS plant technologies. The technologies are listed in Table 3-2 and Table 3-3 showing the relationship between generation technologies and their components and the assumed learning rates under the central scenario (learning is on a global (G) basis, local (L) to the region, or no learning (-) is associated). The potential for local learning means that technology costs are different in different regions in the same time period. This has been of particular note for technology costs in China which can be substantially lower than other regions. GALLME uses current costs from Aurecon (2021) to calibrate 2020 Australian costs in GALLME. For technologies not commonly deployed in Australia these costs can be higher than other regions. However, the inclusion of local learning assumptions in GALLME means that they can quickly catch up to other regions if deployment occurs. However, they will not always fall to levels seen in China due to differences in production standards for some technologies. That is, to meet Australian standards, the technology product from China would increase in costs and align more with other regions. Regional labour construction and engineering costs also remain a source of differentiation. GenCost 2020-21 | 23 Table 3-2 Assumed technology learning rates under all scenarios Technology Component LR 1 (%) LR 2 (%) References Coal, pf - - - Coal, IGCC G - 2 (International Energy Agency, 2008; Neij, 2008) Coal/Gas/Biomass with CCS G 10 5 (EPRI Palo Alto CA & Commonwealth of Australia, 2010; Rubin et al., 2007) L 20 10 As above + (Grübler et al., 1999; Hayward & Graham, 2013; Schrattenholzer & McDonald, 2001) Gas peaking plant - - - Gas combined cycle - - - Nuclear G - 3 (International Energy Agency, 2008) SMR G 20 10 (Grübler et al., 1999; Hayward & Graham, 2013; Schrattenholzer & McDonald, 2001) Diesel/oil-based generation - - - Reciprocating engines - - - Hydro - - - Biomass G - 5 (International Energy Agency, 2008; Neij, 2008) Concentrating solar thermal (CST) G 14.6 7 (Hayward & Graham, 2013) Photovoltaics G 35 10 (Fraunhofer ISE, 2015; Hayward & Graham, 2013; Wilson, 2012) L - 17.5 As above Onshore wind G - 4.3 (Hayward & Graham, 2013) L - 11.3 As above Offshore wind G - 15 (Samadi, 2018) (van der Zwaan, Rivera- Tinoco, Lensink, & van den Oosterkamp, 2012) (Voormolen, Junginger, Sark, & M, 2016) Wave G - 9 (Hayward & Graham, 2013) CHP - - - Conventional geothermal G - 8 (Hayward & Graham, 2013) 24 | CSIRO Australia’s National Science Agency Technology Component LR 1 (%) LR 2 (%) References L 20 20 (Grübler et al., 1999; Hayward & Graham, 2013; Schrattenholzer & McDonald, 2001) Fuel cells G - 20 (Neij, 2008; Schoots, Kramer, & van der Zwaan, 2010) Pumped hydro G - L - 20 (Grübler et al., 1999; Schrattenholzer & McDonald, 2001) Electrolysis G 18 9 (Schmidt et al., 2017) L 18 9 Steam methane reforming with CCS G 10 5 (EPRI Palo Alto CA & Commonwealth of Australia, 2010; Rubin et al., 2007) L 20 10 As above + (Grübler et al., 1999; Hayward & Graham, 2013; Schrattenholzer & McDonald, 2001) Pf=pulverised fuel, IGCC=integrated gasification combined cycle, CHP=combined heat and power, SMR=small modular reactor Solar photovoltaics is listed as one technology with global and local components however there are three separate PV plant technologies in GALLME. Rooftop PV includes solar photovoltaic modules and the local learning component is the balance of plant (BOP). Large scale PV also include modules and BOP. However, a discount of 25% is given to the BOP to take into account economies of scale in building a large scale versus rooftop PV plant. PV with storage has all the components including batteries. Inverters are not given a learning rate instead they are given a constant cost reduction, which is based on historical data. Geothermal BOP includes the power generation. Shared technology components mean that when one of the technologies that uses that component is installed, the costs decrease not just for that technology but for all technologies that use that component. Up to two learning rates are assigned with LR1 representing the initial learning rate during the early phases of deployment and LR2, a lower learning rate, that occurs during the more mature phase of technology deployment. The learning rates are applied uniformly across all scenarios except in the case of batteries. Table 3-3 provides the learning rate by scenario for batteries. Given batteries are deployed in such large numbers across all scenarios due to vehicle electrification and to support increases in variable renewable generation, without these assumed differences in learning rates the battery cost would converge to a common end point. Allowing for differences in the learning rate is useful in reflecting greater uncertainty in the long-term battery costs. GenCost 2020-21 | 25 Table 3-3 Assumed utility scale energy storage learning rates by scenario Technology Scenario Component LR 1 (%) LR 2(%) Utility scale energy storage – Li-ion Central G - 10 L - 10 Utility scale energy storage – flow batteries Central G - 15 L - 10 Utility scale energy storage – Li-ion High VRE G - 15 L - 15 Utility scale energy storage – flow batteries High VRE G - 15 L - 15 Utility scale energy storage – Li-ion Diverse Tech G - 7.5 L - 7.5 Utility scale energy storage – flow batteries Diverse Tech G - 15 L - 7.5 Li-ion batteries are a component that is used in both PV with storage and utility scale Li-ion battery energy storage. Installation BOP is a component of utility scale battery storage that is shared between both types of utility scale battery storage. Source of High VRE learning rate and flow battery learning rate (Brinsmead, Graham, Hayward, Ratnam, & Reedman, 2015). Central and Diverse Technology li-ion learning rates based on CSIRO estimates. Offshore wind has a learning rate of 15% (IEA, 2020) which is significantly higher than that of onshore wind. In addition to capital cost reductions, offshore wind farms have seen significant increases in capacity factor as larger turbines are used, which reduce the LCOE (IRENA, 2019). We have included an exogenous increase up to the year 2050 of 6% in lower resource regions, and 7% in higher resource regions, up to a maximum of 55%, in capacity factor. Two types of reciprocating engines have been included in GALLME. The first type uses diesel as a fuel and the second, more expensive type uses hydrogen as fuel. They are considered to be mature technologies and therefore do not have a learning rate. They can be used as peaking or ‘baseload’ plant in the model. 3.1.6 Electricity demand and electrification In GenCost 2020-21 we have sought to deepen our approach to electrification and hydrogen production assumptions. Previously we had been reliant on existing published global demand scenarios to capture all demand effects. Our goal is to provide more explicit road vehicle electrification assumptions whilst still using existing sources to set underlying global electricity demand. Underlying electricity demand is sourced from the IEA’s latest version of the World 26 | CSIRO Australia’s National Science Agency Energy Outlook (IEA, 2020). Demand data is provided for the Sustainable Development Scenario (SDS), which is used in our Diverse Technology scenario. The demand data from the Stated Policies (STEPS) scenario is used in our Central scenario. Detailed demand data was not provided for the Net Zero Emissions scenario. However, the text indicates that it is higher than SDS and comparable with STEPS and thus we have applied the STEPS scenario demand assumptions to our Diverse Technology scenario. Added to this is the electric vehicle electricity consumption (net of existing electrification assumptions in the IEA scenarios). The IEA demand data also includes electricity used to make hydrogen by scenario. We have therefore assumed the same level of hydrogen demand per scenario as the IEA’s World Energy Outlook. Global vehicle electrification Global adoption of electric vehicles (EVs) by scenario is projected using an adoption curve calibrated to a different shape to correspond to the matching IEA World Energy Outlook scenario sales shares to ensure consistency across electricity and hydrogen demand. The rate of adoption is highest in the VRE scenario, medium in the Diverse Technology scenario and low in the Central scenario consistent with climate policy ambitions. The shape of the adoption curve varies by vehicle type and by region, where countries that have significant EV uptake already, such as China, Western Europe, India, Japan, North America and rest of OECD Pacific, are leaders and the remaining regions are followers. Cars and light commercial vehicles (LCV) have faster rates of adoption, followed by medium commercial vehicles (MCV) and buses. The EV adoption curves for the Central, High VRE, Diverse Technology scenarios are shown in Figure 3-1, Figure 3-2 and Figure 3-3 respectively. The adoption rate is applied to new vehicle sales shares. Figure 3-1 Projected EV sales share under the Central scenario 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2015 2020 2025 2030 2035 2040 2045 2050 Projected sales share Cars and LCVs Leaders Cars and LCVs Followers MCVs and Buses Leaders MCVs and Buses Followers GenCost 2020-21 | 27 Figure 3-2 Projected EV adoption curve (vehicle sales share) under the High VRE scenario Figure 3-3 Projected EV sales share under the Diverse Technology scenario 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2015 2020 2025 2030 2035 2040 2045 2050 Projected sales share Cars and LCVs Leaders Cars and LCVs Followers MCVs and Buses Leaders MCVs and Buses Followers 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2015 2020 2025 2030 2035 2040 2045 2050 Projected sales share Cars and LCVs Leaders Cars and LCVs Followers MCVs and Buses Leaders MCVs and Buses Followers 28 | CSIRO Australia’s National Science Agency 3.1.7 Hydrogen In previous GenCost projections, GALLME used an exogenous hydrogen price which varied by scenario. Given the large role hydrogen could potentially play in decarbonisation across the whole of the energy and industry sectors, hydrogen production technologies, namely electrolysis and steam methane reforming with CCS, now have learning rates applied and contribute to global electricity demand. Their capital costs have been projected based on deployment required to meet demand for hydrogen projected by the IEA and the technology contributions to meeting that demand have been based on adoption curves which vary by scenario. The learning rates used are shown in Table 3-2 and the adoption curves are shown in Figure 3-4 to Figure 3-6. The adoption curves have been designed to provide a range of future technology costs which match each scenario. In the High VRE scenario proton-exchange membrane electrolysis (PEM) is the dominant technology as this works best with VRE. In the Central scenario alkaline electrolysis (AE) is the dominant technology. In the Diverse Technology scenario steam methane reforming with CCS dominates. Figure 3-4 Adoption curves for hydrogen technologies under the Central scenario 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2020 2025 2030 2035 2040 2045 2050 Projected rate of adoption AE PEM GenCost 2020-21 | 29 Figure 3-5 Adoption curves for hydrogen technologies under the High VRE scenario Figure 3-6 Adoption curves for hydrogen technologies under the Diverse Technology scenario 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2020 2025 2030 2035 2040 2045 2050 Projected rate of adoption AE PEM 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2020 2025 2030 2035 2040 2045 2050 Projected rate of adoption AE PEM SMR with CCS 30 | CSIRO Australia’s National Science Agency There is currently a greater installed capacity of AE which has been commercially available since the 1950s, whereas PEM is a more recent technology. The current generation of AE are better suited to a steady and continuous supply of electricity whereas PEM can work with variable renewable supply. However, that balance has been changing with recent developments focussed on improving the performance of AE and reducing the cost of PEM. The IEA have included demand for electricity from electrolysis in their scenarios. Given that we are assuming the same rate of hydrogen demand per scenario as the IEA we have made no changes to electricity demand assumptions to take into account hydrogen production. The assumed hydrogen demand assumptions for the year 2040 are shown in Table 3-4 and include existing demand, the majority of which is met by steam methane reforming. The reason for including existing demand is that in order to achieve emissions reductions the existing demand for hydrogen will also need to be replaced with low emissions sources of hydrogen production. Table 3-4 Hydrogen demand assumptions by scenario Scenario 2040 total hydrogen demand (Mt) Central 80 High VRE 331 Diverse Technology 150 3.1.8 Government climate policies Carbon trading markets exist in major greenhouse gas emitting regions overseas at present and are a favoured approach to global climate policy modelling because they do not introduce any technological bias. We directly impose the IEA carbon prices which are reported in IEA (2020). IEA (2020) also includes a broad range of additional policies such as renewable energy targets and planned closure of fossil-based generation. The GALLME modelling includes these non-carbon price policies as well but cannot completely match the IEA implementation because of model structural differences. The IEA have greater regional and country granularity and are better able to include individual country emissions reduction policies. Some policies are difficult to recreate in GALLME due to its regional aggregation. Where we cannot match the policy implementation directly, we align our implementation of non-carbon price policies so that we match the emission outcomes in the relevant IEA scenario. The country policy commitments included are not completely up to date. To be consistent with the IEA World Energy Outlook 2020, the scenarios do not include more recent announcements or changes of government policy since the IEA report was complete. For example, the WEO 2020 includes China’s 2060 net zero emissions pledge in its sustainable development scenario which we use for Diverse Technology but does not include recent announcements by Japan and South Korea, nor change of leadership in the United States. See Annex B of WEO 2020. GenCost 2020-21 | 31 3.1.9 Resource constraints Constraints around the availability of suitable sites for renewable energy farms, available rooftop space for rooftop PV and sites for storage of CO2 generated from using CCS have been included in GALLME as a constraint on the amount of electricity that can be generated from these technologies (see Government of India, 2016, Edmonds, et al., 2013 and Hayward & Graham, 2017 for more information on sources). Constraints on key renewable technologies in the Central scenario are shown in Table 3-5. In the High VRE scenario, the resource constraint on renewables was removed. In the Diverse Technology scenario, variable renewables will be limited to 40% of generation below the year 2060. However, this will not limit all renewables i.e. all forms of biomass-fuelled and hydrogen-fuelled generation, hydro and geothermal are not limited. Table 3-5 Renewable resource limits on generation in TWh in the year 2050. NA means the resource is greater than projected electricity demand. Region Rooftop PV Large scale PV CSP Onshore wind AFR 565 NA NA NA AUS 113 NA NA NA CHI 1913 NA NA NA EUE 179 NA NA NA EUW 776 112 1155 2125 FSU 300 NA NA NA IND 416 1732 1465 550 JPN 165 17 174 247 LAM 587 NA NA NA MEA 531 NA NA NA NAM 1901 NA NA NA PAO 157 47 480 682 SEA 647 249 2566 974 The regions are Africa (AFR), Australia (AUS), China (CHI), Eastern Europe (EUE), Former Soviet Union (FSU), India (IND), Japan (JPN), Latin America (LAM), Middle East (MEA), North America (NAM), OECD Pacific (PAO), Rest of Asia (SEA), and Western Europe (EUW) 32 | CSIRO Australia’s National Science Agency 3.1.10 Other data assumptions GALLME international fossil fuel prices are based on (IEA, 2020) as shown in Table 3-6 for gas and Table 3-7 for black coal. Brown coal has a flat price of 0.6 $/GJ and there is one global oil price which is shown in Table 3-8. Table 3-6 Assumed gas prices in $A/GJ 2019 2025 2040 2050 AFR 12 13 16 18 AUS8 6 5 5 5 CHI 12 12 13 13 EUE 10 10 12 14 EUW 10 10 12 14 FSU 10 10 12 14 IND 15 13 13 13 JPN 15 13 13 13 LAM 12 12 13 13 MEA 4 5 6 7 NAM 4 5 6 7 PAO 15 13 13 13 SEA 10 10 12 14 8 It should be noted that the IEA’s Australian gas prices are more optimistic than we would normally expect. Given the strong climate policy ambition of the IEA scenarios, they are reflective of a system with low gas demand. Australian assumptions have very minor impact on the global modelling. GenCost 2020-21 | 33 Table 3-7 Assumed black coal prices in $A/GJ 2019 2025 2040 2050 AFR 5.1 4.7 4.7 4.7 AUS 2.9 2.7 2.7 2.7 CHI 5.6 5.0 4.8 4.6 EUE 3.7 4.0 4.2 4.3 EUW 3.7 4.0 4.2 4.3 FSU 2.8 3.2 3.0 2.9 IND 2.8 3.2 3.0 2.9 JPN 5.1 4.7 4.7 4.7 LAM 5.1 4.7 4.7 4.7 MEA 5.1 4.7 4.7 4.7 NAM 2.8 3.2 3.0 2.9 PAO 5.1 4.7 4.7 4.7 SEA 5.1 4.7 4.7 4.7 Table 3-8 Assumed global oil price in $A/bbl 2019 2025 2040 2050 Global price 91 103 123 139 Power plant technology operating and maintenance (O&M) costs, plant efficiencies and fossil fuel emission factors were obtained from (IEA, 2016b) (IEA, 2015), capacity factors from (IRENA, 2015) (IEA, 2015) (CO2CRC, 2015) and historical technology installed capacities from (IEA , 2008) (Gas Turbine World, 2009) (Gas Turbine World, 2010) (Gas Turbine World, 2011) (Gas Turbine World, 2012) (Gas Turbine World, 2013) (UN, 2015a) (UN, 2015b) (US Energy Information Administration, 2017) (US Energy Information Administration, 2017) (GWEC) (IEA) (IEA, 2016a) (World Nuclear Association, 2017) (Schmidt, Hawkes, Gambhir, & Staffell, 2017) (Cavanagh, et al., 2015). 34 | CSIRO Australia’s National Science Agency 4 Projection results 4.1 Global generation mix The rate of technology deployment is the key driver for the rate of reduction in technology costs for all non-mature technologies. However, the generation mix is determined by technology costs. Recognising this, the projection modelling approach simultaneously determines the global generation mix and the capital costs. The projected generation mix consistent with the capital cost projections described in the next section is shown in Figure 4-1. Figure 4-1 Projected global electricity generation mix in 2030 and 2050 by scenario The technology categories displayed are more aggregated than in the model to improve clarity. Solar includes solar thermal and solar photovoltaics. Central scenario has the lowest electrification because it has the least climate policy ambition. However, it has the least energy efficiency and industry transformation9. For this reason, it has similar overall electricity demand to High VRE which has the most climate policy ambition, high vehicle electrification and high hydrogen electrolysis but also high energy efficiency and industry transformation which offsets these sources of new electricity demand growth. Diverse Technology 9 Economies can reduce their emissions by reducing the activity of emission intensive sectors and increasing the activity of low emission sectors. This is not the same as improving the energy efficiency of an emission intensive sector. Industry transformation can also be driven by changes in consumer preferences away from emission intensive products. 0 10000 20000 30000 40000 50000 60000 Central scenario High VRE Diverse technology Central scenario High VRE Diverse technology 2030 2050 Generation (TWh) BECCS Coal Coal CCS Gas Gas CCS Hydro Nuclear Oil Solar Wind onshore Wind offshore Other renewables GenCost 2020-21 | 35 also has stronger climate policy ambition than Central, but its hydrogen production is dominated by gas with CCS. By design, Diverse Technology has a low renewable share for its level of climate policy ambition. Variable renewables such as wind and solar PV are limited to a 50% share and as a result total nonhydro renewable generation accounts for 59% of generation by 2050. Coal and gas with CCS are the main substitutes for lower renewables with gas being the most preferred CCS technology. A small amount of gas without CCS also remains in the mix. Nuclear has a proportionally higher role, 44% higher than in High VRE and 25% higher than Central. The Central scenario has the least climate policy ambition and as a result it has the highest amount of coal, gas and oil-based generation in 2030 and 2050. The non-hydro renewable share of generation is 50% by 2050 with a strong focus on solar and wind. The High VRE scenario is near zero emission by 2050 with a non-hydro renewable share of 82% by 2050. In 2030, it has the highest retirement of existing coal, gas and oil-based generation with earlier deployment of solar and wind generation. Offshore wind features strongly in this scenario, reflecting the strong need for renewable resources. Nuclear generation is the lowest in High VRE consistent with the dominance of lower cost renewables available at high volumes with high social, political and technical support. 4.2 Changes in capital cost projections This section discusses the changes in cost projections to 2050 compared to the 2019-20 projections. For mature technologies, where the current costs have not changed and the assumed improvement rate is very similar, their projection pathways often overlap. The assumed annual rate of cost reduction for mature technologies is 0.2% in this report. This is faster than the 0.01% calculated in GenCost 2019-20. The method for calculating the reduction rate for mature technologies is outlined in Appendix A. Data tables for the full range of technology projections are provided in Appendix B and from CSIRO’s Data Access Portal10. 10 Search GenCost at https://data.csiro.au/collections 36 | CSIRO Australia’s National Science Agency 4.2.1 Black coal supercritical The 2019-20 black coal generation capital costs were based on GHD (2018). For the 2020-21 projections, Aurecon (2021) has increased the current cost by around $1000/kW. However, the assumed rate of improvement in mature technologies is faster which leads to a modest amount of convergence in the projections over time. Figure 4-2 Projected capital costs for black coal supercritical by scenario compared to 2019-20 projections 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 37 4.2.2 Coal with CCS The 2019-20 black coal with CCS current capital costs were based on GHD (2018) and have been updated by Aurecon (2020) for the 2020-21 projections. Consequently, these projections begin from a higher starting point of just over $9000/kW. CCS is not deployed in Central and high VRE and so it does not achieve any cost reduction. This is more negative than the 2019-20 results for those scenarios. The update to higher current costs is likely responsible. Given assumed lower confidence in the deployment of variable renewables, the Diverse Technology scenario has the earliest and highest deployment of CCS in both the generation sector and in gas-based hydrogen production. Substantial deployment commences from around 2027 which is a four years later than in the 2019-20 projections. While the global generation share of coal with CCS is low at 5%, CCS cost reductions include co-learning from deployment of gas with CCS (global CCS generation share is 13%). Brown coal with CCS is included in the Appendix B data tables. It experiences a similar cost trajectory to black coal with CCS due to co-learning. Figure 4-3 Projected capital costs for black coal with CCS by scenario compared to 2019-20 projections 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central 38 | CSIRO Australia’s National Science Agency 4.2.3 Gas combined cycle Gas combined cycle is classed as a mature technology for projection purposes and as a result its change in capital cost is governed by our assumed cost improvement rate for mature technologies. Consequently, the rate of improvement is constant across the Central, High VRE and Diverse Technology scenarios. The current capital cost for gas combined cycle was updated by Aurecon (2021) and is only slightly higher than in 2019-20. The faster assumed reduction in mature technology costs in the 2020-21 projections results in a convergence with 2019-20 projections by 2025 and low costs in the long run. Figure 4-4 Projected capital costs for gas combined cycle by scenario compared to 2019-20 projections 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 39 4.2.4 Gas with CCS The current cost for gas with CCS has been revised slightly upwards for the 2020-21 projections based on Aurecon (2021). Given assumed lower confidence in the deployment of variable renewables, the Diverse Technology scenario has the earliest and highest deployment of gas with CCS in both the generation sector and in gas-based hydrogen production. Coal with CCS, to a lesser extent, also contributes to co-learning between these three CCS technologies. Substantial deployment commences from around 2027 which is around four years later than in the 2019-20 projections. The cost reduction by 2050 is not as great as that which was projected in the 2019-20 reflecting a lower scale of deployment. Deployment does not occur in any significant amount in Central and High VRE reflecting a preference for building lower cost renewables (plus the use of existing coal and gas without CCS in Central). As such there is no cost reduction achieved. Figure 4-5 Projected capital costs for gas with CCS by scenario compared to 2019-20 projections 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central 40 | CSIRO Australia’s National Science Agency 4.2.5 Gas open cycle (small) The 2020-21 projections include results for both large- and small-scale gas open cycle generation. However, only small scale is shown in Figure 4-6 because only small was included in the 2019-20 projections. Both projections are provided in Appendix B with large open cycle starting at around $900/kW. Open cycle gas is classed as a mature technology for projection purposes and as a result its change in capital costs is governed by our assumed cost improvement rate for mature technologies. Consequently, the rate of improvement is constant across the scenarios. The faster rate of cost reduction for mature technologies assumed in the 2020-21 projections means that the projection exceeds cost reductions projected by 2050 in 2019-20. Aurecon (2021) reports that current gas open cycle costs are impacted by global over supply and so there is some risk that costs will be adjusted upward if future conditions allow. Figure 4-6 Projected capital costs for gas open cycle (small) by scenario compared to 2019-20 projections 0 200 400 600 800 1000 1200 1400 1600 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 41 4.2.6 Nuclear SMR The 2020-21 projections start from 2030 consistent with feedback that if nuclear SMR is deployed in Australia, it will not be before 2030. Like the 2019-20 projections, the scenarios present a divergent set of possibilities for nuclear SMR. On the one hand, nuclear SMR does not make any significant cost reduction in the High VRE scenario because deployment of SMR does not proceed. The model has chosen instead to invest in reducing costs of renewables as the most efficient solution reflecting the already low cost of renewables and the scenario context of abundant renewable resources. Alternatively, nuclear SMR is deployed in Central and Diverse Technology. In both scenarios, significant capital cost reductions occur in the period leading up to 2030 (this could be because of an industry development program in one of the leading nuclear energy nations). This resulting cost projection is consistent with the sort of building program for a modular technology which manufacturers are hoping to undertake. Modular plants reduce the number of unique inputs that need to be manufactured. In Central and Diverse Technology, capital costs are around $7000/kW. These results are consistent with the 2019-20 projections except that nuclear SMR development is slightly delayed in the Central scenario. Figure 4-7 Projected capital costs for nuclear SMR by scenario compared to 2019-20 projections 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central 42 | CSIRO Australia’s National Science Agency 4.2.7 Solar thermal with 8 hours storage The current capital cost of solar thermal generation was revised upwards for the 2020-21 projections reflecting escalation in Australian project cost estimates. Cost reductions across the scenarios proceed at a faster and steady pace across the scenarios compared to the 2019-20 projections reflecting stronger alignment of the scenarios with the climate ambitions in the IEA 2020 World Energy Outlook. The overall scale of capital cost reduction, around $2500/kW, is similar in Diverse Technology and High VRE. However, the cost reduction in Central is lower. Central does not need to deploy as large a volume of renewables due to weaker climate policy ambitions and so concentrates on lower cost solar PV and wind. Figure 4-8 Projected capital costs for solar thermal with 8 hours storage by scenario compared to 2019-20 projections 0 1000 2000 3000 4000 5000 6000 7000 8000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 43 4.2.8 Large scale solar PV Large-scale solar PV was expected to continue to track their historical learning rate, however cost reductions have slowed in the 2020-21 current cost update. This reflects local challenges in the Australia industry whereby several solar developers went out of business. These developments have reduced competition and led to more conservative outcomes. The modelling has not built any further industry disruptions into the projection and so cost reductions resume in forward years. For future years, the capital cost projections are reasonably aligned with the 2019-20 projections. Under Diverse Technology, variable renewables are limited so that solar PV deployment is lower and as a result less learning occurs, and capital costs are at a higher level. Central and High VRE have greater deployment and subsequent learning. Higher levels of deployment result in High VRE having the lowest costs. Figure 4-9 Projected capital costs for large scale solar PV by scenario compared to 2019-20 projections 0 200 400 600 800 1000 1200 1400 1600 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central 44 | CSIRO Australia’s National Science Agency 4.2.9 Rooftop solar PV Rooftop solar PV capital costs have been adjusted to align with a 6.6kW system size given the increasing popularity of this system size. The 2019-20 assumption was 5kW. This change to larger systems which have economies of scale in installation costs together with general cost reductions across all system sizes means that the projection starts with a significant reduction in capital costs. The current costs for rooftop solar PV systems are sourced from historical data published by Solar Choice. However, they note that there are significantly discounted rooftop solar PV system prices available at any time and so their data is best interpreted as a mean and may not align with the lowest cost systems available. Rooftop solar PV benefits from co-learning with the components in common with large scale PV generation and is also impacted by the same drivers for variable renewable generation deployment across scenarios. As a result, we can observe similar trends in the rate of capital cost reduction in each scenario as for large-scale solar PV. Figure 4-10 Projected capital costs for rooftop solar PV by scenario compared to 2019-20 projections 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 45 4.2.10 Onshore wind The current capital cost for onshore wind remains like 2019-20 projections and this is consistent with observations that the capital cost learning rate of wind is slowing, at around 4% for each doubling of cumulative global capacity. However, while capital costs are falling slower for wind than solar PV, it is making improvements in its capacity factor which continue to make this technology one of the lowest cost available. Capital costs fall the slowest in Central reflecting lower climate change policy ambition. Diverse Technology and High VRE achieve similar reductions in wind capital costs over time with High VRE just slightly lower. Figure 4-11 Projected capital costs for onshore wind by scenario compared to 2019-20 projections 0 500 1000 1500 2000 2500 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central 46 | CSIRO Australia’s National Science Agency 4.2.11 Offshore wind Offshore wind plays an important role globally in countries with good wind resources, relatively shallow coastal depths and strong competition for land use onshore. The current capital cost of offshore wind has been revised downwards based on Aurecon (2021). The learning rate has also been increased based on more evidence of improvement. Consistent with these changes the projected capital cost reductions are lower than all the 2019-20 projections. In addition to its capital cost, offshore wind has a high potential to improve its capacity factor since very large turbines can be built without impinging on the amenity of neighbouring land uses. These high capacity factors ensure offshore wind is a competitive technology globally, contributing just under 18% of electricity generation by 2050 in High VRE. Capital cost reductions are highest in High VRE which has the greatest deployment as expected. Cost reduction are lowest in Diverse Technology where it is assumed variable renewable generation technologies are more limited in their deployment. Costs reduction are slow to develop in Central due to more limited climate policy ambition. Figure 4-12 Projected capital costs for offshore wind by scenario compared to 2019-20 projections 4.2.12 Battery storage Batteries have been able to sustain high cost reduction rates over time and the use of different learning rates by scenario has meant the projections can reflect some uncertainty as to how well they will be able to continue to achieve historical cost reduction trends. Historical cost reductions have mainly been achieved through deployment in industries other than electricity such as in consumer electronics and electric vehicles. However, small- and large-scale stationary electricity system applications are growing globally from a small base. Under the three global scenarios, 0 1000 2000 3000 4000 5000 6000 7000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 47 batteries have a large future role to play supporting variable renewables alongside other storage and flexible generation options and in growing electric vehicle deployment. Aurecon (2021) has revised the current capital cost of 2-hour duration batteries (including balance of plant) downwards from $622/kWh to around $529/kWh11. Based on this updated current cost, the projected future change in battery pack costs is shown in Figure 4-13 (total costs are in Appendix B). Figure 4-13 Projected total capital costs for 2 hour duration batteries by scenario (battery and balance of plant) Battery deployment is strongest in the High VRE scenario reflecting stronger deployment of variable renewables increasing electricity sector storage requirements and stronger uptake of electric vehicles to support achieving near zero emission by 2050. Together with an assumed high learning rate this leads to the fastest cost reduction which is most consistent with recent trends. Diverse Technology and Central have slower rates of cost reduction reflecting slower uptake of electric vehicles and stationary storage and assumed lower learning rates. Consistent with the scenario narrative of a lack of confidence in variable renewable energy generation, the battery learning rate is assumed to be lowest in Diverse Technology. While Diverse Technology has strong climate policy ambition and strong deployment of electric vehicles to support transport sector greenhouse gas abatement, it has the slowest reduction in battery total project costs. The Central scenario has low climate policy ambition and the slowest uptake of electric vehicles. However, its assumed higher learning rate of batteries means that battery costs fall faster than in Diverse Technology, particularly from the late 2020s. 11 These are large scale batteries. Small scale batteries for home use with 2-hour duration cost around $1250/kWh (SunWiz, 2021) 0 100 200 300 400 500 600 700 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kWh 2019-20 Central 2019-20 Diverse technology 2019-20 High VRE 2020-21 Central 2020-21 Diverse technology 2020-21 High VRE 48 | CSIRO Australia’s National Science Agency A breakdown of battery pack and balance of plant costs for various storage durations are provided in Appendix B. 4.2.13 Pumped hydro energy storage Pumped hydro energy storage is assumed to be a mostly mature technology with only a small proportion of site drilling/piping having the potential to improve with deployment12. Given the strong deployment of variable renewables in all scenarios and subsequent need for storage, this component of learning is maximised in all scenarios so that their cost trajectory is identical over time. The increase in costs compared to the 2019-20 projections is due to the change in sources to AEMO’s December 2020 ISP assumptions. Appendix B includes the costs of pumped hydro energy storage at different durations. We also assume that Tasmania 48 hour pumped hydro storage is 46% the cost of the mainland owing to greater confidence in Tasmanian project cost estimates (and consistent with the AEMO ISP). Figure 4-14 Projected capital costs for pumped hydro energy storage (12 hours) by scenario 12 This improvement occurs generically for the capital cost of pumped hydro energy storage. However, any capital cost estimate is a mean of projects that may have a wide distribution of costs due to site conditions. It is possible that poorer site conditions may offset cost savings from improved drilling productivity. 0 500 1000 1500 2000 2500 3000 2015 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW 2020-21 High VRE 2020-21 Diverse technology 2020-21 Central 2019-20 High VRE 2019-20 Diverse technology 2019-20 Central GenCost 2020-21 | 49 4.2.14 Other technologies There are several technologies that are not commonly deployed in Australia but may be important from a global energy resources perspective or as emerging technologies. These additional technologies are included in the projections for completeness and discussed below. They are each influenced by revisions to current costs. While the estimate for wave electricity generation has not changed all other technologies have. Biomass with CCS has been revised upwards to be consistent with the proportional costs of CCS in coal generation (which increased due to Aurecon (2021) updates). Tidal/current technology updates to capital costs have been sourced from AUSTEn (2020) and reflect more in-depth analysis. Fuel cell updates were included in Aurecon (2021) and mainly reflect a smaller assumed average plant size. Central scenario Biomass with CCS is not adopted in the Central scenario because the climate policy ambition is not strong enough to incentivise deployment. Cost reductions reflect co-learning from other CCS technologies which are deployed. Fuel cell cost improvements are mainly a function of deployment and co-learning in the vehicle sector rather than electricity generation. There are modest cost reductions in tidal/current mainly reflecting a limited number of quality sites in various regions of world. Wave generation achieves the greatest cost reduction reflecting a higher assumed learning rate due to its relative immaturity. Earlier deployment compared to the 2019-20 projection reflects higher climate policy ambition in the 2020-21 Central scenario assumptions. Figure 4-15 Projected technology capital costs under the Central scenario compared to 2019-20 projections 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 2018 2022 2026 2030 2034 2038 2042 2046 2050 2020-21 $/kW Tidal/Ocean current Fuel cell Wave Biomass with CCS 2019-20 Tidal/Ocean current 2019-20 Fuel cell 2019-20 Wave 2019-20 Biomass with CCS 50 | CSIRO Australia’s National Science Agency High VRE scenario Biomass with CCS is not adopted in the High VRE scenario. Although this scenario has the highest climate policy ambition, it can reach near zero emissions earlier using various renewables so that a higher cost negative abatement technology is not required or competitive. Cost reductions reflect co-learning from other CCS technologies which are deployed. Fuel cell generation does not achieve a significant share due to its high costs, but cost reductions are achieved through co-learning with fuel cell vehicles. Tidal/current generation is deployed the fastest in this scenario reflecting the greater need for alternative energy sources to reach net zero emissions. Wave generation deploys earlier than in 2019-20 also reflecting a stronger climate policy. Figure 4-16 Projected technology capital costs under the High VRE scenario compared to 2019-20 projections Diverse Technology scenario Biomass with CCS is deployed early in the Diverse Technology scenario reflecting the assumed limitations on variable renewable energy generation. This result is also consistent with the climate policy ambition of this scenario. Biomass with CCS benefits from co-learning from the significant deployment of gas and coal with CCS generation and from hydrogen production from gas with CCS in this scenario. Wave generation is relatively delayed in this scenario reflecting assumed limitations on variable renewables which are slightly tighter than that assumed in the 2019-20 assumptions. Both tidal/current and fuel cell generation remain niche. 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 2018 2022 2026 2030 2034 2038 2042 2046 2050 2020-21 $/kW Tidal/Ocean current Fuel cell Wave Biomass with CCS 2019-20 Tidal/Ocean current 2019-20 Fuel cell 2019-20 Wave 2019-20 Biomass with CCS GenCost 2020-21 | 51 Figure 4-17 Projected technology capital costs under the Diverse Technology scenario compared to 2019-20 projections 4.3 Hydrogen electrolysers Alkaline electrolysers are currently lower cost than proton-exchange membrane (PEM) electrolysers and they have a common learning rate applied in the modelling. However, we assume that PEM electrolysers are more suited to varying their daily output which makes them more suited to matching their production to low cost variable renewable energy generation. As the costs of both technologies fall, energy input costs increase in proportion making it increasingly more efficient to sacrifice electrolyser capacity utilisation for lower energy costs. Hence PEM electrolysers are projected to be lower cost over the long term. Electrolyser deployment is being supported by a substantial number of hydrogen supply and enduse trials globally and in Australia. Experience with other emerging technologies indicates that this type of globally coincident technology deployment activity can lead to a scale-up in manufacturing which supports cost reductions through economies of scale. Very low costs, at the bottom end of the projections here, have been reported in China. However, differences in engineering and operating and maintenance costs mean these are not able to be immediately replicated in other regions. They do indicate, however, a likely achievable level for other regions over the longer term. Cost reductions are projected to be greatest in the High VRE scenario where global hydrogen production is assumed to be the largest. There is also substantial hydrogen production in Diverse Technology but gas with CCS takes a greater share of hydrogen production leading to lower deployment of electrolysers. The Central scenario achieves the least reduction in costs owing to lower global demand for hydrogen consistent with less climate policy ambition in this scenario. 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 2018 2022 2026 2030 2034 2038 2042 2046 2050 2020-21 $/kW Tidal/Ocean current Fuel cell Wave Biomass with CCS 2019-20 Tidal/Ocean current 2019-20 Fuel cell 2019-20 Wave 2019-20 Biomass with CCS 52 | CSIRO Australia’s National Science Agency Figure 4-18 Projected technology capital costs for alkaline and PEM electrolysers by scenario 0 500 1000 1500 2000 2500 3000 3500 4000 2020 2025 2030 2035 2040 2045 2050 2020-21 $/kW Central PEM Diverse technology PEM High VRE PEM Central Alkaline Diverse technology Alkaline High VRE Alkaline GenCost 2020-21 | 53 5 Levelised cost of electricity analysis Levelised cost of electricity (LCOE) data is an electricity generation technology comparison metric. It is the total unit costs a generator must recover to meet all its costs including a return to investment. Modelling studies such as AEMO’s Integrated System Plan do not require or use LCOE data13. LCOE is a simple screening tool for quickly determining the relative competitiveness of electricity generation technologies. It is not a substitute for detailed project cashflow analysis or electricity system modelling which both provide more realistic representations of electricity generation project operational costs and performance. Furthermore, in the GenCost 2018 report and a supplementary report on methods for calculating the additional costs of renewables (Graham, 2018), we described several issues and concerns in calculating and interpreting levelised cost of electricity. These include:  LCOE does not take account of the additional costs associated with each technology and in particular the integration costs of variable renewable electricity generation technologies  LCOE applies the same discount rate across all technologies even though fossil fuel technologies face a greater risk of being impacted by the introduction of current or new state or commonwealth climate change policies.  LCOE does not recognise that electricity generation technologies have different roles in the system. Some technologies are operated less frequently, increasing their costs, but are valued for their ability to quickly make their capacity available at peak times. In Graham (2018), after reviewing several alternatives from the global literature, we proposed a new method for addressing the first dot point – inclusion of balancing and other costs unique to variable renewables costs. That new method has been implemented and we include those results in the projected LCOEs. To address other issues not associated with additional cost of renewables, when we present LCOE information we:  Separate and group together peaking technologies, flexible technologies and variable technologies  Include additional LCOE data on fossil fuel technologies which includes an additional risk premium of 5% based on Jacobs (2017). 13 LCOE is a measure of the long run marginal cost of generation which could partly inform generator bidding behaviour in a model of the electricity dispatch system. However, in such cases, it would be expected that the LCOE calculation would be internal to the modelling framework to ensure consistency with other model inputs rather than drawn from separate source material. 54 | CSIRO Australia’s National Science Agency 5.1 Overview of the new method 5.1.1 Options considered Graham (2018) reviewed the methods of seven published studies which were relevant in developing a method for taking account of additional integration costs of variable renewables. Some of the reviews included the International Energy Agency’s value adjusted levelised cost, the Energy Information Agency’s levelised avoided cost of electricity and outputs from the MEGS model that has been applied in the Australian context. In evaluating the different approaches, we developed an ideal set of criteria to compare methods. They are that the method should:  Include the full breadth of renewable balancing solutions,  Include the capacity to recognise the context in which the renewables are being deployed,  Include the ability to draw conclusions about separate technologies as opposed to combinations, and  Be transparent and repeatable. No existing method was able to meet all these criteria and we concluded that it was unlikely that any new method would. Instead we must choose between simpler Excel implementable tools and complex system models. Simple Excel based tools can examine each technology separately and are highly transparent but can only focus on one balancing cost and are not able to say when these additional costs will be required. Complex system models can simultaneously examine the broadest range of additional costs of variable renewables and provide context on when these costs will need to be incurred but are only transparent and repeatable to the model or licence owner, not the audience. It was concluded that the system modelling approach is preferred because, while transparency is lost, a greater weight is placed on the ability to study the broadest range of balancing solutions, at the right scale to meet a variety of relevant contexts. None of the system modelling approaches reviewed included all the major relevant balancing solutions for variable renewables which include transmission, storage, other flexible generation and technologies for maintaining inertia and system strength. They also tended to emphasise reliable solutions or least cost investment but not both at the same time. Two commonly applied system modelling frameworks are generation expansion models (intertemporal optimisation models) and dispatch models (optimisation models at half-hour time-scales and below) (Figure 5-1). Dispatch models provide the highest confidence that the balancing solutions will be reliable in the context of the Australian National Electricity Market (NEM). However, on their own, dispatch models require a high number of iterations to optimise investment in the portfolio of solutions. Generation expansion models do optimise investment in solutions but their oversimplified time slicing (representing a year through a small number of representative time periods) means those solutions are not reliable. A third option which we have concluded is the best compromise is an intermediate horizon model (which intertemporally optimises investments over a shorter horizon while also optimising operation of the assets during each day). Intermediate horizon models can automatically co GenCost 2020-21 | 55 optimise investment in all balancing solutions while also simulating their operation to meet demand with a reasonable degree of reliability. They do this by simultaneously optimising most or all hours in a one to five-year timeframe. Figure 5-1 Three types of electricity system models 5.1.2 Development of STABLE CSIRO has developed an intermediate horizon model called STABLE: Spatial Temporal Analysis of Balancing Levelised-cost of Energy. STABLE has drawn on the open source DIETER model for its basic design and been modified substantially to incorporate the details of the National Electricity Market. Time is represented hourly for one year. Most underlying data is based on the July 2020 AEMO ISP inputs and assumptions workbook and various other data (such as renewable energy production traces) published as part of the Integrated System Plan process. Demand is solved at the transmission zone level and Renewable Energy Zones are the smallest spatial supply subregions associated with each transmission zone. 5.2 LCOE estimates 5.2.1 Calculating additional costs of variable renewables We implement STABLE by selecting a future year of interest, 203014, imposing a required variable renewable energy (VRE) share and running the model to determine the optimal investment to 14 This year makes the most sense within the framework applied because there is enough time to plausibly reach high VRE shares but in the counterfactual or business as usual variable renewable shares are still expected to be at or below 50% in the larger states. In the 2040s and 2050s, much of the existing flexible capacity in the system will retire due to end of asset life and be replaced with variable renewables (see AEMO ISP and other long-term modelling). As such, most of the additional costs will already be incurred in the counterfactual. Intermediate horizon model Dispatch model Generation expansion model STABLE Co-optimised generation, storage and transmission investment (annual steps to 2050 and beyond with limited subannual timeslices) Co-optimised investment and operation for reliability and security (one to several hours time steps for one to five years) Operation of existing capacities to the reliability standard is optimised each five minutes to half hour. Partial investment optimisation can be achieved through model iteration. 56 | CSIRO Australia’s National Science Agency support the VRE share. In practice, although wave, current, solar thermal and offshore wind are available as variable renewable technologies, onshore wind and large-scale solar PV are the only variable renewables deployed in the modelling due to their cost competitiveness15. We also implement a business as usual (BAU) optimisation of the same future year and use this as a counterfactual to determine which investments were additional to support the variable renewable shares imposed. STABLE’s BAU is like AEMO’s Central scenario with the following exceptions:  GenCost cost data was used for all generation and storage costs  Demand side participation is currently excluded from STABLE  Rooftop solar PV adoption, electric vehicle adoption and battery adoption with Virtual Power Plant (VPP) participation are consistent with the ESOO 2020 Central projections rather than current ISP assumptions. Customer non-VPP battery and electric charging patterns are also consistent with ESOO 2020.  Nine weather years, 2011 to 2019, are applied to the variable renewable supply traces and demand profiles  No emissions constraint is applied  The New South Wales Electricity Infrastructure Roadmap (which was not included in ISP 2020 modelling) is included. We also apply 50% renewables targets in Victoria and Queensland and 100% in Tasmania. With South Australia already over 50% this means the NEM as a whole is just under 50% VRE share in 2030 before we impose 50% or higher VRE targets. The exclusion of demand side participation (typically around 5% of peak demand) means that the model must deploy other resources to manage system balancing. This makes the result slightly more conservative in terms of investment required to meet demand. New South Wales, Queensland and Victoria are the main states of interest because Tasmania and South Australia are already dominated by renewables such that the BAU already includes all necessary investment to support very high VRE shares. However, the NEM is an interconnected system so we are also interested in how the states support each other and the overall costs for the NEM. The VRE share is applied in each state at the same time but individual states can exceed the share if it is economic to do so. The BAU includes similar retirements of existing coal plants to the ISP. As we implement higher variable renewable energy shares, we must further forcibly retire coal plant as meeting the variable renewable share and the minimum load requirements on coal plant would otherwise eventually become infeasible16. Snowy 2.0 is assumed to be constructed before 2030 in the BAU as well as various transmission expansion projects already flagged by the ISP process to be necessary before 2030. 15 This does not preclude other types of projects proceeding in reality but is a reflection of modelling inputs. 16 The model would be unable to simultaneously meet the minimum VRE share and the minimum run requirements of coal plant which are around 25 to 40% of rated capacity. GenCost 2020-21 | 57 Variable renewable energy shares (VREs) are explored in the range 50% to 90%. Below 50% is not of interest because the BAU already achieves 47% to 48% across the NEM (depending on the weather year). Above 90% VRE share is also not of interest because it would mean forcibly retiring other non-variable renewables such as hydro and biomass which would not be optimal for the system. In the nine weather year counterfactuals, the model does not choose to build any new fossil fuelbased generation capacity (Figure 5-2). However, it also chooses the same level of battery and pumped hydro storage. The main investment response to the different weather is to build more or less of the wind and solar PV capacity with each varying by around 1 GW. The capacities shown are reasonably consistent with 2020 AEMO ISP 2030 capacity projections. This capacity mix is higher in renewables and storage mainly because the 2020 ISP modelling did not include the New South Wales Electricity Infrastructure Roadmap. Figure 5-2 Range of generation and storage capacity deployed in 2030 across the 9 weather year counterfactuals The costs of VRE share scenarios were compared against the same counterfactual weather year to determine the additional integration costs of achieving higher VRE shares. We use the maximum cost across all weather years as the resulting integration cost on the basis that the maximum cost represents a system that has been planned to be reliable across the worst outcomes from weather variation. The results, shown in Figure 5-3, include storage transmission and synchronous condenser costs. Synchronous condensers are required to replace lost inertia from mainly fossil fuel-based generation when it retires to make way for the higher VRE shares. As expected, the results, indicate that additional costs increase with higher VRE shares. Previous analysis (see for example Campey et al. (2017)) has indicated that storage requirements increase 0 2 4 6 8 10 12 14 16 18 20 Wind Solar PV CCGT Peaking gas & liquids Black coal Brown coal Hydro Pumped hydro Battery GW 58 | CSIRO Australia’s National Science Agency non-linearly with VRE share, starting with little or no requirement at 50% VRE, and the results conform to that expectation. Transmission costs include the transmission costs to connect Renewable Energy Zones (REZs) to the grid and other transmission which includes state interconnectors and general expansion of existing lines that connect transmission zones within states. REZ expansion costs appear to be required at similar levels for each additional 10% increase. Other transmission expenditure, not already in the BAU, is mainly required in significant levels in NSW and Victoria. The transmission investment in NSW reflects its central position in the NEM and both NSW and Victoria need strengthening of transmission between zones to get energy from the new REZs to demand centres. In Queensland the REZ transmission investment itself is sufficient. Storage requirements are highest in NSW and Queensland. This reflects existing flexible resources and the quality of the variable renewable resources. Queensland storage requirements are significant but not as high as NSW because it has a wind resource which tends to be stronger at night which is therefore well suited to filling in the gaps left by solar PV. Queensland has 15% less peaking gas and liquids capacity than Victoria. NSW has 40% less peaking gas and liquids capacity than Victoria and more coincident wind and solar resource. The greater peaking gas and liquids capacity in Victoria (around 3 GW) explains part of why it does not require significant additional storage. The other reason is that it has good proximity to hydro and pumped hydro capacity in Tasmania and New South Wales. Figure 5-3 Levelised costs of achieving 50%, 60%, 70%, 80% and 90% variable renewable energy shares in the NEM, NSW, VIC and QLD in 2030 0 10 20 30 40 50 60 70 80 90 50% VRE 60% VRE 70% VRE 80% VRE 90% VRE $/MWh Generation REZ transmission Other transmission Synchronous condensers Storage NEM 0 10 20 30 40 50 60 70 80 90 50% VRE 60% VRE 70% VRE 80% VRE 90% VRE $/MWh Generation REZ transmission Other transmission Synchronous condensers Storage NSW 0 10 20 30 40 50 60 70 80 90 50% VRE 60% VRE 70% VRE 80% VRE 90% VRE $/MWh Generation REZ transmission Other transmission Synchronous condensers Storage VIC 0 10 20 30 40 50 60 70 80 90 50% VRE 60% VRE 70% VRE 80% VRE 90% VRE $/MWh Generation REZ transmission Other transmission Synchronous condensers Storage QLD GenCost 2020-21 | 59 Additional expenditure on synchronous condenser capacity is required in all states and increasing moderately with variable renewable share. Higher VRE share leads to the retirement of fossil fuelbased capacity that otherwise supplies most of system inertia. The larger role of hydro generation in New South Wales means that it has a lower requirement for additional synchronous condensers. The variation in state requirements is due to different resources and the system exploits the interconnectivity between state to reduce overall costs. As such higher or lower costs in different states are averaged out at the NEM level. At a 50% VRE share additional NEM level integration costs are generally very low, because they only include REZ transmission costs – the existing flexibility and inertia in the system is adequate to manage this VRE share without additional investment in storage or synchronous condensers. The cost of REZ transmission expansions adds around $5/MWh to $6/MWh, hardly changing at all as the VRE share increases. Synchronous condensers costs are low at between $1.0/MWh to $1.5/MWh increasing moderately with VRE share. Other transmission adds between $0 to $3.4/MWh at the NEM level with costs accelerating with VRE share. Storage adds between $0.1 to $9.6/MWh, also increasing non-linearly with VRE share. 5.2.2 Variable renewables with and without integration costs The results for the additional costs for increasing variable renewable shares are used to update and extend our LCOE estimates. We expand the results for 2030 to include a combined wind and solar PV category for different VRE shares. We have also removed the wind plus storage and solar PV plus storage categories that were included in GenCost 2018 and GenCost 2019-20. These were always designed to be temporary estimates until a better approach was available. In GenCost 2019-20, for 2030, the simple approach of adding 2 or 6 hours storage added $19 to $106/MWh to the cost of variable renewables for an unspecified share of generation. With the new approach the additional costs to support renewables are estimated at $6 to $19/MWh depending on the VRE share (Figure 5-5). As such, the previous approach was too conservative. While it did not consider transmission and synchronous condensers, which are important additional costs, it over-estimated the need for storage17 and, in total, over-estimated the additional integration costs that might be associated with variable renewable generation. Variable renewables (wind and solar PV) without transmission or storage costs are the lowest cost generation technology by a significant margin. From 2030, the new estimates on additional costs associated with increasing variable renewable generation confirms that they are also competitive when transmission, synchronous condenser and storage costs are included. The closest technology is the low range cost of a gas combined cycle generator which can match the costs of variable renewables with integration costs at a 70% or greater share. The low range 2030 gas combined cycle cost assumptions require no climate policy risk at the financing stage (despite the 25 year design life extending beyond the net zero emission targets of most states), a gas price just below 17 The previous approach assigned an equal capacity storage plant to every solar PV and wind project. In practice, the system modelling shows that it is not necessary to deploy storage at a one to one ratio. The previous approach fails to take account of existing flexible capacity in the system and non-coincident renewable supply (such that renewable generation would never be zero across the NEM). 60 | CSIRO Australia’s National Science Agency $6/GJ throughout that period and a capacity factor of 80% in a system with 70% or greater share of energy from near zero marginal cost renewables. 5.2.3 Peaking technologies The peaking technology category includes two sizes for gas turbines and a gas reciprocating engine. Fuel comprises the majority of costs, but the lower capital costs of the larger gas turbine make it the most competitive. Reciprocating engines have higher efficiency and consequently, for applications with relatively higher capacity factors and where a smaller unit size is required, they can be the lower cost choice. 5.2.4 Flexible technologies Evaluated purely on their energy costs, black coal, brown coal and gas-based generation technologies that are designed to deliver energy for 40 to 80% of the year are the next most competitive generation technologies after variable renewables (with or without transmission and storage). It is difficult to say which fossil fuel is more competitive as it depends very much on whether gas generation can secure gas supply at the lower end of the fuel cost range (just under $6/GJ). New fossil fuel generation faces the risk of higher financing costs over time because all states in Australia have either legislated or aspirational net zero emission targets. There is also bipartisan commitment to the Paris agreement which is aiming for net zero emissions in the second half of the century. We address these risks in the cost estimations by including a separate estimate which assumes a 5% risk premium on borrowing costs. Natural gas-based generation is less impacted by the risk premium because of its lower emission fuel, higher thermal efficiency (in combined cycle configuration only) and lower capital cost. We do not include a risk premium for low emission flexible technologies. Gas with CCS is the most competitive of this group however the lower end of the range is only achievable if it can source low cost gas. Solar thermal and small modular reactor (SMR) nuclear are the next most competitive. Achieving the lower end of the SMR range requires that SMR is deployed globally in large enough numbers to bring down costs available to Australia. GenCost 2020-21 | 61 Figure 5-4 Calculated LCOE by technology and category for 2020 Figure 5-5 Calculated LCOE by technology and category for 2030 0 50 100 150 200 250 300 350 400 Gas turbine small Gas turbine large Gas reciprocating Black coal Brown coal Gas Black coal Brown coal Gas Black coal with CCS Brown coal with CCS Gas with CCS Solar thermal 8hrs Biomass (small scale) Wind Solar PV Climate policy risk premium Standalone Peaking 20% load Flexible 40-80% load, high emission Flexible 40-80% load, low emission Variable 2020-21 A$/MWh 0 50 100 150 200 250 300 350 400 Gas turbine small Gas turbine large Gas reciprocating Black coal Brown coal Gas Black coal Brown coal Gas Black coal with CCS Brown coal with CCS Gas with CCS Solar thermal 8hrs Nuclear (SMR) Biomass (small scale) Wind Solar PV 50% VRE share 60% VRE share 70% VRE share 80% VRE share 90% VRE share Climate policy risk premium Standalone Wind & solar PV combined Peaking 20% load Flexible 40-80% load, high emission Flexible 40-80% load, low emission Variable Variable with integration costs 2020-21 A$/MWh 62 | CSIRO Australia’s National Science Agency Figure 5-6 Calculated LCOE by technology and category for 2040 Figure 5-7 Calculated LCOE by technology and category for 2050 0 50 100 150 200 250 300 350 400 Gas turbine small Gas turbine large Gas reciprocating Black coal Brown coal Gas Black coal Brown coal Gas Black coal with CCS Brown coal with CCS Gas with CCS Solar thermal 8hrs Nuclear (SMR) Biomass (small scale) Wind Solar PV Climate policy risk premium Standalone Peaking 20% load Flexible 40-80% load, high emission Flexible 40-80% load, low emission Variable 2020-21 A$/MWh 0 50 100 150 200 250 300 350 400 450 500 Gas turbine small Gas turbine large Gas reciprocating Black coal Brown coal Gas Black coal Brown coal Gas Black coal with CCS Brown coal with CCS Gas with CCS Solar thermal 8hrs Nuclear (SMR) Biomass (small scale) Wind Solar PV Climate policy risk premium Carbon price Standalone Peaking 20% load Flexible 40-80% load, high emission Flexible 40-80% load, low emission Variable 2020-21 A$/MWh GenCost 2020-21 | 63 Global and local learning model A.1 GALLM The Global and Local Learning Models (GALLMs) for electricity (GALLME) and transport (GALLMT) are described briefly here. More detail can be found in several existing publications (Hayward & Graham, 2017) (Hayward & Graham, 2013) (Hayward, Foster, Graham, & Reedman, 2017). A.1.1 Endogenous technology learning Technology cost reductions due to ‘learning-by-doing’ were first observed in the 1930s for aeroplane construction (Wright, 1936) and have since been observed and measured for a wide range of technologies and processes (McDonald & Schrattenholzer, 2001). Cost reductions due to this phenomenon are normally shown via the equation: where IC is the unit investment cost at CC cumulative capacity and IC0 is the cost of the first unit at CC0 cumulative capacity. The learning index b satisfies 0 < b < 1 and it determines the learning rate which is calculated as: (typically quoted as a percentage ranging from 0 to 50%) and the progress ratio is given by PR=100-LR. All three quantities express a measure of the decline in unit cost with learning or experience. This relationship says that for each doubling in cumulative capacity of a technology, its investment cost will fall by the learning rate (Hayward & Graham, 2013). Learning rates can be measured by examining the change in unit cost with cumulative capacity of a technology over time. Typically, emerging technologies have a higher learning rate (15–20%), which reduces once the technology has at least a 5% market share and is considered to be at the intermediate stage (to approximately 10%). Once a technology is considered mature, the learning rate tends to be 0–5%. The transition between learning rates based on technology uptake is illustrated in Apx Figure A.1. 𝐼𝐶 = 𝐼𝐶0 × 􁉀 𝐶𝐶 𝐶𝐶0 􁉁 −𝑏 , or equivalently log(𝐼𝐶) = log( 𝐼𝐶0 ) − 𝑏(log(𝐶𝐶) − log(𝐶𝐶0)) 𝐿𝑅 = 100 × (1 − 2−𝑏 ) 64 | CSIRO Australia’s National Science Agency Apx Figure A.1 Schematic of changes in the learning rate as a technology progresses through its development stages after commercialisation However, technologies that do not have a standard unit size and can be used in a variety of applications tend to have a higher learning rate for longer (Wilson, 2012). This is the case for solar photovoltaics and historically for gas turbines. Technologies are made up of components and different components can be at different levels of maturity and thus have different learning rates. Different parts of a technology can be developed and sold in different markets (global vs. regional/local) which can impact on the cost reductions as each region will have a different level of demand for a technology and this will affect its uptake. 5.2.5 The modelling framework To project the future cost of a technology using experience curves, the future level of cumulative capacity/uptake needs to be known. However, this is dependent on the costs. The GALLM models solve this problem by simultaneously projecting both the cost and uptake of the technologies. The optimisation problem includes constraints such as government policies, demand for electricity or transport, capacity of existing technologies, exogenous costs such as for fossil fuels and limits on resources (e.g. rooftops for solar photovoltaics). The models have been divided into 13 regions and each region has unique assumptions and data for the above listed constraints. The regions have been based on Organisation for Economic Co-operation Development (OECD) regions (with some variation to look more closely at some countries of interest) and are: Africa, Australia, China, Eastern Europe, Western Europe, Former Soviet Union, India, Japan, Latin America, Middle East, North America, OECD Pacific, Rest of Asia and Pacific. The objective of the model is to minimise the total system costs while meeting demand and all constraints. The model is solved as a mixed integer linear program. The experience curves are segmented into step functions and the location on the experience curves (i.e. cost vs. cumulative GenCost 2020-21 | 65 capacity) is determined at each time step. See (Hayward & Graham, 2013) and (Hayward, Foster, Graham, & Reedman, 2017) for more information. Both models run from the year 2006 to 2100. However, results are only reported from the present year to 2050. A.1.2 Mature technologies and the “basket of costs” There are three main drivers of mature technology costs: imported materials and equipment, domestic materials and equipment, and labour. The indices of these drivers over the last 20 years (ABS data) combined with the split in capital cost of mature technologies between imported equipment, domestic equipment and labour (Bureau of Resource and Energy Economics (BREE), 2012) was used to calculate an average rate of change in technology costs: - 0.2%. This value has been applied as an annual capital cost reduction factor to mature technologies and to operating and maintenance costs. 66 | CSIRO Australia’s National Science Agency Data tables The following tables provide data behind the figures presented in this document. GenCost 2020-21 | 67 Apx Table B.1 Current and projected generation technology capital costs under the Central scenario Black coal Black coal with CCS Brown coal Brown coal with CCS Gas combined cycle Gas open cycle (small) Gas open cycle (large) Gas with CCS Gas reciprocating Hydrogen reciprocating Biomass (small scale) Biomass with CCS (large scale) Large scale solar PV Rooftop solar panels Solar thermal (8 hrs) Wind Offshore wind Wave Nuclear (SMR) Tidal/ocean current Fuel cell Integrated solar and battery (2 hrs) $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW 2020 4450 9311 6868 14293 1743 1465 873 4396 1471 1750 7265 18438 1505 1439 7411 1951 5771 9384 - 6971 9333 2139 2021 4441 9297 6854 14269 1740 1462 871 4391 1469 1746 7260 18406 1358 1304 7281 1942 5632 9384 - 6954 9144 1974 2022 4432 9283 6841 14245 1736 1459 869 4387 1466 1743 7256 18373 1258 1211 7204 1934 5522 9384 - 6798 8951 1852 2023 4423 9269 6827 14221 1733 1456 868 4382 1463 1739 7254 18341 1205 1162 7140 1929 5460 9384 - 6652 8766 1780 2024 4414 9255 6813 14197 1729 1453 866 4378 1460 1736 7254 18309 1158 1117 7083 1926 5423 9384 - 6506 8600 1713 2025 4406 9241 6800 14173 1726 1450 864 4374 1457 1732 7254 18277 1113 1076 6983 1923 5395 9384 - 6506 8460 1652 2026 4397 9227 6786 14149 1722 1447 862 4369 1454 1729 7254 18245 1070 1036 6881 1920 5371 9384 - 6506 8342 1593 2027 4388 9213 6773 14125 1719 1444 861 4365 1451 1725 7254 18213 1030 998 6762 1918 5356 9384 - 6506 8248 1539 2028 4379 9199 6759 14102 1715 1441 859 4361 1448 1722 7254 18181 995 963 6657 1915 5356 9028 - 6506 8179 1490 2029 4370 9185 6746 14078 1712 1438 857 4356 1445 1718 7254 18149 963 933 6570 1912 5354 8495 - 6506 8110 1416 2030 4362 9171 6732 14054 1709 1436 856 4352 1442 1715 7254 18117 933 904 6496 1910 5351 7704 7375 6506 8064 1345 2031 4353 9158 6719 14031 1705 1433 854 4348 1439 1711 7253 18086 910 882 6432 1908 5348 7078 7346 6506 8032 1279 2032 4344 9144 6705 14007 1702 1430 852 4343 1437 1708 7253 18054 892 863 6377 1905 5347 6519 7345 6506 8030 1246 2033 4336 9130 6692 13984 1698 1427 850 4339 1434 1705 7253 18022 878 849 6328 1900 5347 6144 7345 6506 8028 1218 2034 4327 9116 6678 13960 1695 1424 849 4335 1431 1701 7253 17991 865 835 6286 1894 5347 5911 7345 6506 8026 1190 2035 4318 9103 6665 13937 1691 1421 847 4331 1428 1698 7253 17959 844 815 6248 1887 5347 5560 7345 6506 8024 1157 2036 4310 9089 6652 13913 1688 1418 845 4326 1425 1694 7253 17928 826 797 6214 1882 5347 5282 7345 6495 8023 1126 2037 4301 9075 6638 13890 1685 1416 844 4322 1422 1691 7253 17896 808 780 6183 1877 5292 5032 7345 6482 8021 1099 2038 4292 9061 6625 13867 1681 1413 842 4318 1419 1688 7253 17865 800 772 6156 1873 5198 5011 7345 6469 7891 1081 2039 4284 9048 6612 13844 1678 1410 840 4314 1417 1684 7253 17834 791 763 6131 1868 5077 4869 7345 6467 7663 1064 2040 4275 9034 6599 13820 1675 1407 839 4309 1414 1681 7253 17803 778 751 6087 1863 5010 4674 7345 6467 7406 1046 2041 4267 9021 6585 13797 1671 1404 837 4305 1411 1678 7253 17772 760 733 6026 1858 4980 4434 7345 6467 7255 1025 2042 4258 9007 6572 13774 1668 1401 835 4301 1408 1674 7253 17741 738 713 5948 1854 4972 4314 7345 6435 7199 1001 2043 4250 8994 6559 13751 1665 1399 834 4297 1405 1671 7253 17710 716 693 5877 1850 4966 4256 7345 6395 7167 977 2044 4241 8980 6546 13728 1661 1396 832 4292 1402 1667 7253 17679 696 674 5811 1847 4962 4243 7345 6225 7155 953 2045 4233 8967 6533 13705 1658 1393 830 4288 1400 1664 7253 17648 676 655 5751 1844 4953 4226 7345 6080 7143 932 2046 4224 8953 6520 13682 1655 1390 829 4284 1397 1661 7253 17617 661 641 5695 1842 4938 4218 7345 5943 7132 915 2047 4216 8940 6507 13659 1651 1387 827 4280 1394 1657 7253 17586 649 629 5644 1839 4909 4206 7345 5906 7122 901 2048 4207 8926 6494 13637 1648 1385 825 4276 1391 1654 7253 17556 638 619 5596 1836 4889 4194 7345 5875 7114 889 2049 4199 8913 6481 13614 1645 1382 824 4272 1388 1651 7253 17525 630 611 5552 1832 4861 4183 7345 5778 7105 879 2050 4195 8906 6474 13602 1643 1381 822 4270 1387 1649 7253 17510 624 606 5530 1830 4853 4178 7345 5745 7101 873 68 | CSIRO Australia’s National Science Agency Apx Table B.2 Current and projected generation technology capital costs under the High VRE scenario Black coal Black coal with CCS Brown coal Brown coal with CCS Gas combined cycle Gas open cycle (small) Gas open cycle (large) Gas with CCS Gas reciprocating Hydrogen reciprocating Biomass (small scale) Biomass with CCS (large scale) Large scale solar PV Rooftop solar panels Solar thermal (8 hrs) Wind Offshore wind Wave Nuclear (SMR) Tidal/ocean current Fuel cell Integrated solar and battery (2 hrs) $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW 2020 4450 9311 6868 14293 1743 1465 873 4396 1471 1750 7265 18438 1505 1439 7411 1951 5771 9384 - 6971 9333 2139 2021 4441 9297 6854 14269 1740 1462 871 4391 1469 1746 7260 18406 1210 1170 7386 1942 5706 9384 - 6954 9314 1779 2022 4432 9283 6841 14245 1736 1459 869 4387 1466 1743 7256 18373 1020 997 7331 1936 5632 9384 - 6945 9294 1526 2023 4423 9269 6827 14221 1733 1456 868 4382 1463 1739 7254 18341 949 931 7198 1931 5552 9384 - 6945 9275 1401 2024 4414 9255 6813 14197 1729 1453 866 4378 1460 1736 7254 18309 903 888 6984 1921 5483 9384 - 6945 9257 1308 2025 4406 9241 6800 14173 1726 1450 864 4374 1457 1732 7254 18277 874 861 6746 1908 5424 9384 - 6945 9242 1243 2026 4397 9227 6786 14149 1722 1447 862 4369 1454 1729 7254 18245 849 837 6555 1894 5373 9384 - 6945 9230 1189 2027 4388 9213 6773 14125 1719 1444 861 4365 1451 1725 7254 18213 828 815 6427 1882 5329 9384 - 6945 9222 1144 2028 4379 9199 6759 14102 1715 1441 859 4361 1448 1722 7254 18181 806 793 6299 1874 5289 9384 - 6945 9216 1103 2029 4370 9185 6746 14078 1712 1438 857 4356 1445 1718 7254 18149 786 774 6136 1867 5254 9384 - 6945 8864 1059 2030 4362 9171 6732 14054 1709 1436 856 4352 1442 1715 7254 18117 768 755 5968 1863 5223 9384 16487 6945 8490 1021 2031 4353 9158 6719 14031 1705 1433 854 4348 1439 1711 7254 18086 757 744 5827 1858 5195 9384 16487 6945 8098 995 2032 4344 9144 6705 14007 1702 1430 852 4343 1437 1708 7254 18054 742 728 5740 1854 5169 9384 16487 6945 8037 977 2033 4336 9130 6692 13984 1698 1427 850 4339 1434 1705 7254 18022 730 716 5658 1850 5146 9384 16487 6798 7987 961 2034 4327 9116 6678 13960 1695 1424 849 4335 1431 1701 7254 17991 702 688 5597 1846 5125 9384 16487 6652 7945 930 2035 4318 9103 6665 13937 1691 1421 847 4331 1428 1698 7254 17959 680 667 5531 1842 5106 9384 16487 6467 7912 906 2036 4310 9089 6652 13913 1688 1418 845 4326 1425 1694 7254 17928 646 635 5483 1837 5089 9384 16487 6428 7885 872 2037 4301 9075 6638 13890 1685 1416 844 4322 1422 1691 7254 17896 616 607 5432 1833 5072 8349 16487 6389 7861 841 2038 4292 9061 6625 13867 1681 1413 842 4318 1419 1688 7254 17865 588 580 5386 1830 5045 7015 16487 6389 7839 812 2039 4284 9048 6612 13844 1678 1410 840 4314 1417 1684 7254 17834 571 565 5311 1827 5007 5584 16487 6389 7820 794 2040 4275 9034 6599 13820 1675 1407 839 4309 1414 1681 7254 17803 569 562 5234 1822 4959 4986 16487 6389 7805 789 2041 4267 9021 6585 13797 1671 1404 837 4305 1411 1678 7254 17772 566 558 5138 1817 4915 4579 16487 6389 7794 784 2042 4258 9007 6572 13774 1668 1401 835 4301 1408 1674 7254 17741 563 555 5063 1812 4873 4205 16487 6389 7788 779 2043 4250 8994 6559 13751 1665 1399 834 4297 1405 1671 7254 17710 560 552 4979 1803 4835 3995 16487 6377 7785 774 2044 4241 8980 6546 13728 1661 1396 832 4292 1402 1667 7254 17679 558 549 4923 1794 4800 3735 16487 6323 7784 769 2045 4233 8967 6533 13705 1658 1393 830 4288 1400 1664 7254 17648 555 546 4871 1785 4766 3472 16487 6270 7783 764 2046 4224 8953 6520 13682 1655 1390 829 4284 1397 1661 7254 17617 553 544 4838 1783 4735 3243 16487 6228 7782 761 2047 4216 8940 6507 13659 1651 1387 827 4280 1394 1657 7254 17586 549 539 4808 1780 4706 3158 16487 6228 7781 755 2048 4207 8926 6494 13637 1648 1385 825 4276 1391 1654 7254 17556 545 535 4783 1778 4679 3128 16487 6228 7780 749 2049 4199 8913 6481 13614 1645 1382 824 4272 1388 1651 7254 17525 536 527 4760 1775 4653 3081 16487 6118 7779 739 2050 4195 8906 6474 13602 1643 1381 822 4270 1387 1649 7254 17510 532 523 4748 1774 4640 3060 16487 6063 7779 735 GenCost 2020-21 | 69 Apx Table B.3 Current and projected generation technology capital costs under the Diverse Technology scenario Black coal Black coal with CCS Brown coal Brown coal with CCS Gas combined cycle Gas open cycle (small) Gas open cycle (large) Gas with CCS Gas reciprocating Hydrogen reciprocating Biomass (small scale) Biomass with CCS (large scale) Large scale solar PV Rooftop solar panels Solar thermal (8 hrs) Wind Offshore wind Wave Nuclear (SMR) Tidal/ocean current Fuel cell Integrated solar and battery (2 hrs) $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW $/kW 2020 4450 9311 6868 14293 1743 1465 873 4396 1471 1750 7265 18438 1505 1439 7411 1951 5771 9384 - 6971 9333 2139 2021 4441 9297 6854 14269 1740 1462 871 4391 1469 1746 7260 18406 1449 1386 7394 1937 5728 9384 - 6932 9235 2076 2022 4432 9283 6841 14245 1736 1459 869 4387 1466 1743 7256 18373 1402 1341 7354 1926 5669 9384 - 6913 9136 2020 2023 4423 9269 6827 14221 1733 1456 868 4382 1463 1739 7254 18341 1364 1305 7247 1918 5598 9384 - 6913 9046 1976 2024 4414 9255 6813 14197 1729 1453 866 4378 1460 1736 7254 18309 1332 1274 7073 1909 5515 9384 - 6913 8970 1937 2025 4406 9241 6800 14173 1726 1450 864 4374 1457 1732 7254 18277 1301 1246 6843 1901 5446 9384 - 6913 8912 1894 2026 4397 9227 6786 14149 1722 1447 862 4369 1454 1729 7254 18245 1273 1218 6632 1894 5386 9384 - 6913 8865 1845 2027 4388 9139 6773 14051 1719 1444 861 4292 1451 1725 7254 18138 1247 1193 6484 1888 5335 9384 - 6913 8832 1792 2028 4379 9015 6759 13917 1715 1441 859 4180 1448 1722 7254 17996 1225 1171 6396 1882 5290 9384 - 6913 8809 1745 2029 4370 8863 6746 13754 1712 1438 857 4041 1445 1718 7254 17825 1206 1152 6333 1877 5251 9384 - 6913 8790 1693 2030 4362 8674 6732 13556 1709 1436 856 3865 1442 1715 7251 17618 1189 1135 6235 1872 5217 9226 7239 6913 8778 1646 2031 4353 8477 6719 13348 1705 1433 854 3682 1439 1711 7248 17403 1177 1122 6104 1867 5187 8207 7229 6913 8770 1605 2032 4344 8308 6705 13170 1702 1430 852 3526 1437 1708 7245 17215 1168 1113 5955 1863 5160 7015 7218 6913 8770 1581 2033 4336 8248 6692 13099 1698 1427 850 3476 1434 1705 7245 17137 1162 1105 5829 1861 5135 5981 7217 6913 8769 1560 2034 4327 8227 6678 13069 1695 1424 849 3465 1431 1701 7245 17099 1155 1098 5730 1859 5112 5808 7217 6789 8730 1538 2035 4318 8182 6665 13014 1691 1421 847 3430 1428 1698 7245 17035 1148 1090 5658 1857 5095 5808 7213 6577 8645 1517 2036 4310 8136 6652 12959 1688 1418 845 3394 1425 1694 7221 16972 1141 1083 5615 1855 5087 5738 7210 6361 8511 1498 2037 4301 8092 6638 12905 1685 1416 844 3360 1422 1691 7193 16910 1126 1068 5578 1854 5080 5562 7207 6269 8378 1472 2038 4292 8070 6625 12873 1681 1413 842 3348 1419 1688 7137 16871 1108 1051 5553 1852 5076 5323 7207 6265 8245 1445 2039 4284 8050 6612 12843 1678 1410 840 3337 1417 1684 7098 16832 1086 1030 5522 1849 5069 5155 7207 6264 8127 1417 2040 4275 8030 6599 12813 1675 1407 839 3327 1414 1681 7043 16795 1071 1016 5505 1846 5066 5092 7207 6264 8041 1396 2041 4267 8011 6585 12785 1671 1404 837 3317 1411 1678 7005 16759 1057 1003 5475 1836 5063 5092 7207 6264 7994 1379 2042 4258 7994 6572 12759 1668 1401 835 3310 1408 1674 6973 16724 1044 990 5442 1828 5056 5092 7207 6253 7973 1363 2043 4250 7979 6559 12735 1665 1399 834 3305 1405 1671 6955 16692 1032 979 5403 1818 5048 4917 7207 6240 7950 1347 2044 4241 7966 6546 12711 1661 1396 832 3300 1402 1667 6948 16661 1020 967 5370 1815 5038 4742 7207 6226 7928 1331 2045 4233 7952 6533 12688 1658 1393 830 3296 1400 1664 6934 16630 1009 957 5346 1812 5033 4551 7207 6218 7908 1317 2046 4224 7938 6520 12665 1655 1390 829 3291 1397 1661 6925 16599 1000 948 5305 1808 5024 4535 7207 6214 7892 1305 2047 4216 7925 6507 12642 1651 1387 827 3287 1394 1657 6913 16568 992 940 5251 1803 5010 4518 7207 6209 7878 1294 2048 4207 7911 6494 12619 1648 1385 825 3283 1391 1654 6905 16537 987 935 5175 1798 4990 4518 7207 6209 7865 1287 2049 4199 7898 6481 12596 1645 1382 824 3278 1388 1651 6879 16506 980 928 5102 1794 4968 4518 7207 6209 7853 1278 2050 4195 7891 6474 12585 1643 1381 822 3276 1387 1649 6867 16491 977 925 5063 1793 4956 4518 7207 6209 7847 1274 70 | CSIRO Australia’s National Science Agency Apx Table B.4 One and two hour battery cost data by storage duration, component and total costs Battery storage (1 hr) Battery storage (2 hrs) Total Battery BOP Total Battery BOP Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh 2020 789 789 789 340 340 340 449 449 449 529 529 529 300 300 300 229 229 229 2021 763 777 720 323 335 282 440 442 438 509 521 472 285 296 249 224 225 223 2022 741 766 673 310 331 246 431 435 427 493 514 435 273 292 217 220 222 217 2023 718 754 624 295 327 209 422 428 415 476 506 396 260 288 185 215 218 211 2024 701 744 588 287 323 184 413 420 404 464 500 368 254 285 163 211 214 206 2025 680 733 557 275 320 165 404 413 392 449 493 345 243 283 145 206 210 200 2026 662 713 530 266 307 149 395 406 381 436 477 326 235 271 132 201 207 194 2027 645 689 506 258 290 137 386 399 369 425 459 309 228 256 121 197 203 188 2028 619 670 486 242 278 128 377 391 358 406 445 295 213 245 113 192 199 182 2029 593 645 462 225 260 116 368 384 346 386 426 278 198 230 102 188 196 176 2030 567 621 441 208 245 106 359 377 335 367 408 264 184 216 93 183 192 171 2031 542 599 422 191 230 99 350 370 323 347 391 252 169 203 87 178 188 165 2032 516 585 410 174 222 98 341 362 312 328 381 245 154 196 86 174 185 159 2033 500 571 397 167 215 97 332 355 300 317 371 239 148 190 86 169 181 153 2034 484 557 385 161 209 96 323 348 289 306 361 232 142 184 85 165 177 147 2035 467 542 373 153 201 96 314 341 277 295 351 226 135 178 84 160 174 141 2036 453 529 361 148 196 95 305 333 266 286 343 219 130 173 84 156 170 135 2037 438 517 349 142 191 94 296 326 254 276 334 213 125 168 83 151 166 130 2038 425 505 337 138 187 94 287 319 243 268 327 207 122 165 83 146 162 124 2039 412 494 325 134 182 93 278 312 231 260 319 200 118 161 82 142 159 118 2040 400 483 313 131 179 93 269 304 220 253 313 194 116 158 82 137 155 112 2041 391 475 301 130 178 93 260 297 209 248 309 188 115 157 82 133 151 106 2042 381 467 290 130 178 93 251 290 197 242 304 182 114 157 82 128 148 100 2043 370 458 278 127 175 93 242 283 186 236 299 176 112 155 82 123 144 95 2044 359 449 266 126 173 92 233 275 174 230 293 170 111 153 81 119 140 89 2045 349 440 255 125 172 92 224 268 163 224 288 164 110 152 81 114 137 83 2046 339 432 243 124 171 92 215 261 151 219 284 158 110 151 81 110 133 77 2047 330 424 231 124 170 92 206 253 140 214 279 152 109 150 81 105 129 71 2048 320 416 220 123 170 92 197 246 128 209 275 146 109 150 81 100 125 65 2049 311 408 208 123 169 92 188 239 117 204 271 140 108 149 81 96 122 59 2050 301 400 197 122 169 92 179 232 105 199 267 134 108 149 81 91 118 54 GenCost 2020-21 | 71 Apx Table B.5 Four and eight hour battery cost data by storage duration, component and total costs Battery storage (4 hrs) Battery storage (8 hrs) Total Battery BOP Total Battery BOP Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE Central Diverse Technolo gy High VRE $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh $/kWh 2020 421 421 421 300 300 300 121 121 121 371 371 371 300 300 300 71 71 71 2021 403 414 366 285 296 249 118 119 117 354 366 318 285 296 249 70 70 69 2022 389 409 332 273 292 217 116 117 114 341 361 285 273 292 217 68 69 67 2023 374 403 296 260 288 185 113 115 111 327 356 250 260 288 185 67 68 66 2024 365 398 271 254 285 163 111 113 108 319 352 226 254 285 163 65 66 64 2025 351 393 250 243 283 145 108 111 105 307 348 207 243 283 145 64 65 62 2026 341 379 234 235 271 132 106 109 102 297 335 192 235 271 132 62 64 60 2027 331 363 220 228 256 121 104 107 99 289 319 179 228 256 121 61 63 58 2028 315 350 209 213 245 113 101 105 96 273 307 170 213 245 113 60 62 56 2029 297 333 195 198 230 102 99 103 93 256 290 157 198 230 102 58 61 55 2030 280 317 183 184 216 93 96 101 90 240 275 146 184 216 93 57 59 53 2031 263 302 174 169 203 87 94 99 87 224 261 138 169 203 87 55 58 51 2032 245 293 170 154 196 86 92 97 84 208 253 135 154 196 86 54 57 49 2033 237 285 166 148 190 86 89 95 81 200 246 133 148 190 86 52 56 47 2034 228 278 163 142 184 85 87 93 77 193 239 131 142 184 85 51 55 46 2035 219 269 159 135 178 84 84 91 74 185 232 128 135 178 84 50 54 44 2036 212 262 155 130 173 84 82 89 71 179 226 126 130 173 84 48 53 42 2037 205 256 152 125 168 83 79 87 68 172 220 124 125 168 83 47 51 40 2038 199 250 148 122 165 83 77 85 65 167 215 121 122 165 83 45 50 38 2039 193 244 145 118 161 82 75 84 62 162 210 119 118 161 82 44 49 37 2040 188 240 141 116 158 82 72 82 59 158 206 117 116 158 82 43 48 35 2041 185 237 138 115 157 82 70 80 56 156 204 115 115 157 82 41 47 33 2042 182 234 135 114 157 82 67 78 53 154 202 113 114 157 82 40 46 31 2043 177 230 131 112 155 82 65 76 50 151 199 111 112 155 82 38 45 29 2044 174 227 128 111 153 81 63 74 47 148 197 109 111 153 81 37 43 27 2045 170 224 125 110 152 81 60 72 44 146 194 107 110 152 81 35 42 26 2046 167 221 122 110 151 81 58 70 41 143 192 105 110 151 81 34 41 24 2047 164 218 118 109 150 81 55 68 37 142 190 103 109 150 81 33 40 22 2048 161 216 115 109 150 81 53 66 34 140 189 101 109 150 81 31 39 20 2049 159 213 112 108 149 81 50 64 31 138 187 99 108 149 81 30 38 18 2050 156 211 109 108 149 81 48 62 28 136 185 97 108 149 81 28 37 17 72 | CSIRO Australia’s National Science Agency Apx Table B.6 Pumped hydro storage cost data by duration, all scenarios, total cost basis $/kW $/kWh 4hrs 6hrs 8hrs 12hrs 24hrs 48hrs 48hrs Tas 4hrs 6hrs 8hrs 12hrs 24hrs 48hrs 48hrs Tas 2020 1865 2323 2425 2658 3417 5133 2384 466 387 303 221 142 107 50 2021 1861 2319 2421 2653 3411 5124 2380 465 387 303 221 142 107 50 2022 1858 2315 2417 2648 3405 5115 2376 465 386 302 221 142 107 50 2023 1855 2312 2413 2644 3399 5106 2372 464 385 302 220 142 106 49 2024 1852 2308 2409 2639 3393 5098 2368 463 385 301 220 141 106 49 2025 1849 2304 2405 2635 3388 5089 2364 462 384 301 220 141 106 49 2026 1846 2300 2400 2630 3382 5080 2360 461 383 300 219 141 106 49 2027 1842 2296 2396 2626 3376 5072 2356 461 383 300 219 141 106 49 2028 1839 2292 2392 2621 3370 5063 2352 460 382 299 218 140 105 49 2029 1836 2288 2388 2617 3365 5054 2348 459 381 299 218 140 105 49 2030 1833 2284 2384 2612 3359 5046 2344 458 381 298 218 140 105 49 2031 1830 2280 2380 2608 3353 5037 2340 457 380 298 217 140 105 49 2032 1827 2276 2376 2603 3347 5028 2336 457 379 297 217 139 105 49 2033 1824 2272 2372 2599 3342 5020 2332 456 379 296 217 139 105 49 2034 1820 2268 2368 2595 3336 5011 2328 455 378 296 216 139 104 48 2035 1817 2265 2364 2590 3330 5003 2324 454 377 295 216 139 104 48 2036 1814 2261 2360 2586 3325 4994 2320 454 377 295 215 139 104 48 2037 1811 2257 2356 2581 3319 4986 2316 453 376 294 215 138 104 48 2038 1808 2253 2352 2577 3313 4977 2312 452 375 294 215 138 104 48 2039 1805 2249 2348 2573 3308 4969 2308 451 375 293 214 138 104 48 2040 1802 2245 2344 2568 3302 4960 2304 450 374 293 214 138 103 48 2041 1799 2241 2340 2564 3296 4952 2300 450 374 292 214 137 103 48 2042 1796 2238 2336 2559 3291 4943 2296 449 373 292 213 137 103 48 2043 1793 2234 2332 2555 3285 4935 2292 448 372 291 213 137 103 48 2044 1790 2230 2328 2551 3279 4927 2288 447 372 291 213 137 103 48 2045 1787 2226 2324 2546 3274 4918 2284 447 371 290 212 136 102 48 2046 1784 2222 2320 2542 3268 4910 2281 446 370 290 212 136 102 48 2047 1781 2219 2316 2538 3263 4901 2277 445 370 289 211 136 102 47 2048 1778 2215 2312 2533 3257 4893 2273 444 369 289 211 136 102 47 2049 1774 2211 2308 2529 3252 4885 2269 444 369 289 211 135 102 47 2050 1771 2207 2304 2525 3246 4876 2265 443 368 288 210 135 102 47 GenCost 2020-21 | 73 Apx Table B.7 Storage cost data by source, total cost basis $/kWh $/kW Aurecon 2019-20 Aurecon 2020-21 GenCost 2019-20 AEMO ISP December 2020 Aurecon 2019-20 Aurecon 2020-21 GenCost 2019-20 AEMO ISP December 2020 Battery (1hr) 988 789 - - 988 789 - - Battery (2hrs) 622 529 - - 1244 1058 - - Battery (4hrs) 491 421 - - 1964 1682 - - PHES (6hrs) - - 308 387 - - 1850 2323 Battery (8hrs) 446 371 - - 3564 2968 - - PHES (12hrs) - - 177 221 - - 2118 2658 PHES (24hrs) - - 131 142 - - 3139 3417 PHES (48hrs) - - 73 107 - - 3517 5133 PHES (48hrs) Tasmania - - - 50 - - - 2384 Notes: Batteries are large scale. Small scale batteries for home use with 2-hour duration cost around $1250/kWh (SunWiz, 2021). 74 | CSIRO Australia’s National Science Agency Apx Table B.8 Data assumptions for LCOE calculations Constant Low assumption High assumption Economic life Construction time Efficiency O&M fixed O&M variable CO2 storage Capital Fuel Capacity factor Capital Fuel Capacity factor Years Years $/kW $/MWh $/MWh $/kW $/GJ $/kW $/GJ 2020 Gas with CCS 25 1.5 44% 16.4 7.2 1.9 4396 5.8 80% 4396 11.3 60% Gas combined cycle 25 1.5 51% 10.9 3.7 0.0 1743 5.8 80% 1743 11.3 60% Gas open cycle (small) 25 1.3 36% 12.6 4.1 0.0 1465 5.8 20% 1465 11.3 20% Gas open cycle (large) 25 1.1 33% 10.2 2.4 0.0 873 5.8 20% 873 11.3 20% Gas reciprocating 25 1.0 41% 24.1 7.6 0.0 1471 5.8 20% 1471 11.3 20% Black coal with CCS 30 2.0 30% 77.8 8.0 4.1 9311 2.8 80% 9311 4.1 60% Black coal 30 2.0 40% 53.2 4.2 0.0 4450 2.8 80% 4450 4.1 60% Brown coal with CCS 30 4.0 21% 101.6 11.6 4.7 14293 0.6 80% 14293 0.7 60% Brown coal 30 4.0 32% 69.0 5.3 0.0 6868 0.6 80% 6868 0.7 60% Biomass (small scale) 30 2.0 23% 131.6 8.4 0.0 7265 0.5 60% 7265 2.0 40% Large scale solar PV 25 0.5 100% 17.0 0.0 0.0 1505 0.0 32% 1505 0.0 22% Solar thermal (8hrs) 25 1.8 100% 142.5 0.0 0.0 7411 0.0 52% 7411 0.0 42% Wind 25 1.0 100% 25.0 0.0 0.0 1951 0.0 44% 1951 0.0 35% 2030 Gas with CCS 25 1.5 44% 16.4 7.2 1.9 3865 5.8 80% 4352 11.8 60% Gas combined cycle 25 1.5 51% 10.9 3.7 0.0 1709 5.8 80% 1709 11.8 60% Gas open cycle (small) 25 1.3 36% 12.6 4.1 0.0 1436 5.8 20% 1436 11.8 20% Gas open cycle (large) 25 1.1 33% 10.2 2.4 0.0 856 5.8 20% 856 11.8 20% Gas reciprocating 25 1.0 41% 24.1 7.6 0.0 1442 5.8 20% 1442 11.8 20% Black coal with CCS 30 2.0 30% 77.8 8.0 4.1 8674 2.9 80% 9171 3.8 60% Black coal 30 2.0 40% 53.2 4.2 0.0 4362 2.9 80% 4362 3.8 60% Brown coal with CCS 30 4.0 21% 101.6 11.6 4.7 13556 0.7 80% 14054 0.7 60% Brown coal 30 4.0 32% 69.0 5.3 0.0 6732 0.7 80% 6732 0.7 60% Biomass (small scale) 30 2.0 23% 131.6 8.4 0.0 7254 0.5 60% 7254 2.0 40% Nuclear (SMR) 30 3.0 35% 200.0 5.3 0.0 7239 0.5 80% 16487 0.7 60% Large scale solar PV 25 0.5 100% 17.0 0.0 0.0 768 0.0 32% 933 0.0 19% Solar thermal (8hrs) 25 1.8 100% 142.5 0.0 0.0 5968 0.0 52% 6496 0.0 42% Wind 25 1.0 100% 25.0 0.0 0.0 1863 0.0 46% 1910 0.0 35% GenCost 2020-21 | 75 2040 Gas with CCS 25 1.5 44% 16.4 7.2 1.9 3327 5.8 80% 4309 11.8 60% Gas combined cycle 25 1.5 51% 10.9 3.7 0.0 1675 5.8 80% 1675 11.8 60% Gas open cycle (small) 25 1.3 36% 12.6 4.1 0.0 1407 5.8 20% 1407 11.8 20% Gas open cycle (large) 25 1.1 33% 10.2 2.4 0.0 839 5.8 20% 839 11.8 20% Gas reciprocating 25 1.0 41% 24.1 7.6 0.0 1414 5.8 20% 1414 11.8 20% Black coal with CCS 30 2.0 30% 77.8 8.0 4.1 8030 2.9 80% 9034 3.8 60% Black coal 30 2.0 40% 53.2 4.2 0.0 4275 2.9 80% 4275 3.8 60% Brown coal with CCS 30 4.0 21% 101.6 11.6 4.7 12813 0.7 80% 13820 0.7 60% Brown coal 30 4.0 32% 69.0 5.3 0.0 6599 0.7 80% 6599 0.7 60% Biomass (small scale) 30 2.0 23% 131.6 8.4 0.0 7254 0.5 60% 7254 2.0 40% Nuclear (SMR) 30 3.0 40% 200.0 5.3 0.0 7207 0.5 80% 16487 0.7 60% Large scale solar PV 25 0.5 100% 17.0 0.0 0.0 569 0.0 32% 778 0.0 19% Solar thermal (8hrs) 25 1.8 100% 142.5 0.0 0.0 5234 0.0 52% 6087 0.0 42% Wind 25 1.0 100% 25.0 0.0 0.0 1822 0.0 48% 1863 0.0 35% 2050 Gas with CCS 25 1.5 44% 16.4 7.2 1.9 3276 5.8 80% 4270 11.8 60% Gas combined cycle 25 1.5 51% 10.9 3.7 0.0 1643 5.8 80% 1643 11.8 60% Gas open cycle (small) 25 1.3 36% 12.6 4.1 0.0 1381 5.8 20% 1381 11.8 20% Gas open cycle (large) 25 1.1 33% 10.2 2.4 0.0 822 5.8 20% 822 11.8 20% Gas reciprocating 25 1.0 41% 24.1 7.6 0.0 1387 5.8 20% 1387 11.8 20% Black coal with CCS 30 2.0 30% 77.8 8.0 4.1 7891 2.9 80% 8906 3.8 60% Black coal 30 2.0 40% 53.2 4.2 0.0 4195 2.9 80% 4195 3.8 60% Brown coal with CCS 30 4.0 21% 101.6 11.6 4.7 12585 0.7 80% 13602 0.7 60% Brown coal 30 4.0 32% 69.0 5.3 0.0 6474 0.7 80% 6474 0.7 60% Biomass (small scale) 30 2.0 23% 131.6 8.4 0.0 7254 0.5 60% 7253 2.0 40% Nuclear (SMR) 30 3.0 45% 200.0 5.3 0.0 7207 0.5 80% 16487 0.7 60% Large scale solar PV 25 0.5 100% 17.0 0.0 0.0 532 0.0 32% 624 0.0 19% Solar thermal (8hrs) 25 1.8 100% 142.5 0.0 0.0 4748 0.0 52% 5530 0.0 42% Wind 25 1.0 100% 25.0 0.0 0.0 1774 0.0 50% 1830 0.0 35% Notes: Wind is onshore. Large-scale solar PV is single axis tracking. The discount rate used for all technologies is 5.99% unless a risk premium of 5% is added. 76 | CSIRO Australia’s National Science Agency Apx Table B.9 Electricity generation technology LCOE projections data, 2020-21 $/MWh Category Assumption Technology 2020 2030 2040 2050 Low High Low High Low High Low High Peaking 20% load Gas turbine small 139 195 138 199 136 197 135 196 Gas turbine large 112 172 111 177 110 176 110 175 Gas reciprocating 142 190 140 194 139 192 138 191 Flexible 40-80% load, high emission Black coal 89 120 88 116 87 114 86 113 Brown coal 112 145 111 143 110 142 108 139 Gas 67 114 67 117 66 117 66 116 Climate policy risk premium Black coal 127 171 126 166 124 163 122 161 Brown coal 193 254 190 249 188 246 184 242 Gas 80 132 80 134 79 134 78 133 Flexible 40-80% load, low emission Black coal with CCS 165 220 158 214 151 212 149 210 Brown coal with CCS 229 296 220 292 211 289 208 285 Gas with CCS 112 175 105 179 99 178 98 178 Solar thermal 8hrs 172 213 145 192 131 182 121 169 Nuclear (SMR) 128 322 127 321 127 320 Biomass (small scale) 154 246 154 246 154 246 154 246 Variable Standalone Wind 48 61 44 60 42 58 39 58 Solar PV 49 72 28 55 22 48 21 40 Variable with integration costs Wind & solar PV combined 50% share 46 66 60% share 47 67 70% share 50 72 80% share 52 76 90% share 55 80 GenCost 2020-21 | 77 Apx Table B.10 Hydrogen electrolyser cost projections by scenario and technology, 2020-21 $/kW Central High VRE Diverse Technology Alkaline PEM Alkaline PEM Alkaline PEM 2020 2516 3510 2516 3510 2516 3510 2021 2066 2760 1847 2716 2062 2743 2022 1773 2239 1631 2164 1771 2212 2023 1580 1868 1491 1771 1583 1836 2024 1458 1607 1409 1493 1467 1571 2025 1390 1432 1374 1305 1406 1393 2026 1346 1303 1350 1164 1370 1260 2027 1305 1188 1315 1041 1337 1143 2028 1264 1086 1280 932 1307 1039 2029 1225 998 1246 840 1280 950 2030 1185 919 1208 758 1253 870 2031 1147 852 1169 689 1224 796 2032 1108 792 1127 628 1194 731 2033 1068 738 1080 574 1161 674 2034 1028 691 1031 527 1125 622 2035 989 649 982 485 1088 577 2036 950 612 931 449 1048 538 2037 914 581 883 417 1009 504 2038 882 554 838 390 972 475 2039 853 531 797 367 937 450 2040 828 513 760 347 905 429 2041 808 497 728 330 868 405 2042 791 484 699 315 834 385 2043 777 474 674 302 804 367 2044 766 466 652 291 776 352 2045 757 459 632 282 750 338 2046 750 454 614 273 727 326 2047 744 449 598 265 705 315 2048 739 446 582 258 684 305 2049 735 443 568 251 665 295 2050 732 441 555 245 647 287 78 | CSIRO Australia’s National Science Agency Shortened forms Abbreviation Meaning AE Alkaline electrolysis AEMO Australian Energy Market Operator BAU Business as usual bbl Barrel BOP Balance of plant CCS Carbon capture and storage CHP Combined heat and power CO2 Carbon dioxide CO2e Carbon dioxide equivalent CSIRO Commonwealth Scientific and Industrial Research Organisation CSP Concentrated solar power DIETER Dispatch and investment evaluation tool with endogenous renewables EV Electric vehicle GALLME Global and Local Learning Model Electricity GALLMs Global and Local Learning Models GALLMT Global and Local Learning Model Transport GJ Gigajoule GW Gigawatt hrs Hours IEA International Energy Agency IGCC Integrated gasification combined cycle GenCost 2020-21 | 79 Abbreviation Meaning ISP Integrated System plan kW Kilowatt kWh Kilowatt hour LCOE Levelised Cost of Electricity LCV Light commercial vehicle MCV Medium commercial vehicle Li-ion Lithium-ion LR Learning Rate Mt Million tonnes MWh Megawatt hour NEM National Electricity Market O&M Operations and Maintenance OECD Organisation for Economic Cooperation and Development PEM Proton-exchange membrane electrolysis pf Pulverised fuel PHES Pumped hydro energy storage PV Photovoltaic REZ Renewable Energy Zone SDS Sustainable Development Scenario SMR Small modular reactor STABLE Spatial Temporal Analysis of Balancing Levelised-cost of Energy STEPS Stated Policies t tonne 80 | CSIRO Australia’s National Science Agency Abbreviation Meaning TWh Terawatt hour VPP Virtual Power Plant VRE Variable Renewable Energy WEO World Energy Outlook GenCost 2020-21 | 81 References Aurecon 2021, 2020 costs and technical parameter review, AEMO. AUSTEn 2020 Tidal Energy in Australia - Assessing resource and feasibility for Australia's future energy mix. Milestone 4 report to ARENA Australian Electricity Market Operator (AEMO) 2020, Input and assumptions workbook v1 5 Jul 20, AEMO. Bureau of Resource and Energy Economics (BREE) 2012, Australian Energy Technology Assessment, BREE, Canberra. Brinsmead T.S., Graham, P., Hayward, J., Ratnam, E.L., and Reedman, L. 2015, Future Energy Storage Trends: An Assessment of the Economic Viability, Potential Uptake and Impacts of Electrical Energy Storage on the NEM 2015-2035, AEMC, Australia. Campey, T., Bruce, S., Yankos, T., Hayward, J., Graham, P., Reedman, L., Brinsmead, T., Deverell, J. 2017, Low Emissions Technology Roadmap, CSIRO, Australia. Cavanagh, K., Ward, J. K., Behrens, S., Bhatt, A. R., E, O., & J, H. 2015, Electrical energy storage: technology overview and applications. CSIRO for AEMC. CO2CRC 2015, Australian Power Generation Technology Report, CO2CCRC, Canberra. Economic and Finance Working Group (EFWG) 2019, SMR roadmap, Canadian Nuclear Association. Edmonds, J., Lucknow, P., Calvin, K., Wise, M., Dooley, J., Kyle, P., . . . Clarke, L. 2013, Can radiative forcing be limited to 2.6Wm(-2) without negative emissions from bioenergy and CO2 capture and storage? Climatic Change, 118(1), 29-43 SI. doi:10.1007/s10584-012-0678.2 Electric Power Research Institute (EPRI) 2010, Australian Electricity Generation Technology Costs – Reference Case 2010. Department of Resources, Energy and Tourism, Canberra. Fraunhofer ISE, 2015. Current and Future Cost of Photovoltaics. Long-term Scenarios for Market Development, System Prices and LCOE of Utility-Scale PV Systems, s.l.: Study on behalf of Agora Energiewende Gas Turbine World 2009, Gas Turbine World Handbook. Gas Turbine World 2010, Gas Turbine World Handbook. Gas Turbine World 2011, Gas Turbine World Handbook. Gas Turbine World 2012, Gas Turbine World Handbook. Gas Turbine World 2013, Gas Turbine World Handbook. GHD 2018, AEMO costs and technical parameter review: Report final Rev 4 9110715, AEMO, Australia. Government of India. 2016, A new dawn in renewable energy. India: Ministry of new and renewable energy. Retrieved from http://pib.nic.in/newsite/PrintRelease.aspx?relid=155612 82 | CSIRO Australia’s National Science Agency Graham, P., Hayward, J., Foster, J. and Havas, L. 2020, GenCost 2019-20, Australia. Graham, P.W., Hayward, J, Foster, J., Story, O. and Havas, L. 2018, GenCost 2018, CSIRO, Australia. Graham, P. 2018, Review of alternative methods for extending LCOE calculations to include balancing costs. CSIRO, Australia. Grübler, A., Nakicenovic, N., & Victor, D. G. 1999, Dynamics of energy technologies and global change. Energy Policy, 27(5), 247-280. GWEC. (n.d.). Global Wind Report Series 2006 to 2016. Global Wind Energy Council. Hayward, J. A., Foster, J. D., Graham, P. W., & Reedman, L. J. 2017. A Global and Local Learning Model of Transport (GALLM-T). In Syme, G., Hatton MacDonald, D., Fulton, B. and Piantadosi, J. (eds) MODSIM2017, 22nd International Congress on Modelling and Simulation. Modelling and Simulation Society of Australia and New Zealand, December 2017, pp. 818- 824. ISBN: 978-0-9872143-7-9. https://www.mssanz.org.au/modsim2017/F1/hayward.pdf [accessed 18 December 2018] Hayward, J.A. and Graham, P.W. 2017, Electricity generation technology cost projections: 2017- 2050, CSIRO, Australia, https://publications.csiro.au/rpr/download?pid=csiro:EP178771&dsid=DS2 Hayward, J. and Graham, P. 2013, A global and local endogenous experience curve model for projecting future uptake and cost of electricity generation technologies, Energy Economics, 40, 537-548. International Energy Agency (IEA) 2008, Electricity Information 2008 Edition. Paris, France: IEA. International Energy Agency (IEA) 2008, World Energy Outlook. Paris, France: IEA. International Energy Agency (IEA) 2015,. Projected costs of Generating Electricity . Paris, France: OECD. International Energy Agency (IEA) 2016a, CSP Projects Around the World. Retrieved from SolarPACES: http://www.solarpaces.org/csp-technology/csp-projects-around-the-world International Energy Agency (IEA) 2016b, World Energy Outlook . Paris, France: OECD. International Energy Agency (IEA) 2020, World Energy Outlook . Paris: OECD. International Energy Agency (IEA) n.d. Trends in Photovoltaic Applications Series from 2007 to 2016. Paris, France: IEA. IPCC, 2014, Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. IRENA 2019, Renewable Power Generation Costs in 2018. IRENA. Jacobs 2017, Report to the Independent Review into the Future Security of the National Electricity Market: Emission mitigation policies and security of supply, Department of Energy and Environment, Canberra. McDonald, A., & Schrattenholzer, L. 2001, Learning rates for energy technologies. Energy Policy, 29, 255-261. GenCost 2020-21 | 83 Neij, L. 2008, Cost development of future technologies for power generation-A study based on experience curves and complementary bottom-up assessments. Energy Policy, 36(6), 2200- 2211. Rubin, E. S., Yeh, S., Antes, M., Berkenpas, M., & Davison, J. (2007). Use of experience curves to estimate the future cost of power plants with CO2 capture. International Journal of Greenhouse Gas Control, 1(2), 188-197. Samadi, S. 2018, The experience curve theory and its application in the field of electricity generation technologies, Renewable & Sustainable Energy Reviews, Vol. 82 , pp 2346-2364 Schmidt, O., Hawkes, A., Gambhir, A., & Staffell, I. 2017, The future cost of electrical energy storage based on experience rates. Nature Energy, 2, 17110. Schoots, K., Kramer, G.J. and Zwaan, B. 2010, Technology Learning for Fuel Cells: An Assessment of Past and Potential Cost Reductions, Energy Policy, vol. 38, pp. 2887-2897 Schrattenholzer, L., and McDonald, A. 2001 Learning rates for energy technologies. Energy Policy, 29, 255-261. SunWiz 2021, Australian battery market report – 2021, SunWiz. UN 2015a, Energy Statistics Yearbook 2012. New York, USA: United Nations. UN 2015b, Energy Statistics Yearbook 2013. New York, USA: United Nations. US Energy Information Administration 2017, World installed liquids-fired generating capacity by region and country. Retrieved from https://www.eia.gov/outlooks/aeo/data/browser/#/?id=17-IEO2016®ion=0- 0&cases=Reference&start=2010&end=2016&f=A&linechart=Reference-d021916a.2-17- IEO2016&sourcekey=0 US Energy Information Administration 2017, World installed natural-gas-fired generating capacity by region and country. Retrieved from https://www.eia.gov/outlooks/aeo/data/browser/#/?id=18-IEO2016®ion=0- 0&cases=Reference&start=2010&end=2016&f=A&sourcekey=0 Voormolen, J.A., Junginger, H.M. and van Sark, W.G.J.H.M., 2016. Unravelling historical cost developments of offshore wind energy in Europe, Energy Policy, Elsevier, vol. 88(C), pp 435- 444. Wilson, C. (2012). Up-scaling, formative phases, and learning in the historical diffusion of energy technologies. Energy Policy, 50(0), 81-94, http://dx.doi.org/10.1016/j.enpol.2012.04.077 World Nuclear Association 2017, World Nuclear Power Reactors and Uranium Requirements. Retrieved from www.world-nuclear.org: http://www.world-nuclear.org/informationlibrary/ facts-and-figures/world-nuclear-power-reactors-and-uranium-requireme.aspx Wright, T. P. 1936, Factors Affecting the Cost of Airplanes. Journal of the Aeronautical Sciences, 3, 122-128. Zwaan, B., Rivera Tinoco, R., Lensink, S. and Van den Oosterkamp, P. 2012, Cost reductions for offshore wind power: Exploring the balance between scaling, learning and R&D, Renewable Energy, pp389-393 84 | CSIRO Australia’s National Science Agency The GenCost project is a partnership of CSIRO and AEMO. As Australia’s national science agency and innovation catalyst, CSIRO is solving the greatest challenges through innovative science and technology. CSIRO. Unlocking a better future for everyone. Contact us 1300 363 400 +61 3 9545 2176 csiroenquiries@csiro.au csiro.au For further information Energy Paul Graham +61 2 4960 6061 paul.graham@csiro.au csiro.au/energy