NREL-公用事业规模电池储能成本预测-2023年更新（英文原版）.pdf

NREL is a national laboratory of the U.S. Department of Energy Office of Energy Efficiency & Renewable Energy Operated by the Alliance for Sustainable Energy, LLC This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications. Contract No. DE-AC36-08GO28308 Technical Report NREL/TP-6A40-85332 June 2023 Cost Projections for Utility-Scale Battery Storage: 2023 Update Wesley Cole and Akash Karmakar National Renewable Energy Laboratory NREL is a national laboratory of the U.S. Department of Energy Office of Energy Efficiency & Renewable Energy Operated by the Alliance for Sustainable Energy, LLC This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications. Contract No. DE-AC36-08GO28308 National Renewable Energy Laboratory 15013 Denver West Parkway Golden, CO 80401 303-275-3000 • www.nrel.gov Technical Report NREL/TP-6A40-85332 June 2023 Cost Projections for Utility-Scale Battery Storage: 2023 Update Wesley Cole and Akash Karmakar National Renewable Energy Laboratory Suggested Citation Cole, Wesley and Akash Karmakar. 2023. Cost Projections for Utility-Scale Battery Storage: 2023 Update. Golden, CO: National Renewable Energy Laboratory. NREL/TP-6A40-85332. https://www.nrel.gov/docs/fy23osti/85332.pdf. NOTICE This work was authored by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Strategic Programs, Policy and Analysis Office. The views expressed herein do not necessarily represent the views of the DOE or the U.S. Government. This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications. U.S. Department of Energy (DOE) reports produced after 1991 and a growing number of pre-1991 documents are available free via www.OSTI.gov. Cover Photos by Dennis Schroeder: (clockwise, left to right) NREL 51934, NREL 45897, NREL 42160, NREL 45891, NREL 48097, NREL 46526. NREL prints on paper that contains recycled content. iii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. Acknowledgments We are grateful to ReEDS modeling team for their input on this work. We also thank Bethany Frew, Vignesh Ramasamy, and Matt Rippe for providing feedback on this year’s report. This work was authored by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE- AC36-08GO28308. Funding provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Strategic Analysis team. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. All errors and omissions are the sole responsibility of the authors. iv This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. Executive Summary In this work we describe the development of cost and performance projections for utility-scale lithium-ion battery systems, with a focus on 4-hour duration systems. The projections are developed from an analysis of recent publications that include utility-scale storage costs. The suite of publications demonstrates wide variation in projected cost reductions for battery storage over time. Figure ES-1 shows the suite of projected cost reductions (on a normalized basis) collected from the literature (shown in gray) as well as the low, mid, and high cost projections developed in this work (shown in black). Figure ES-2 shows the overall capital cost for a 4-hour battery system based on those projections, with storage costs of $245/kWh, $326/kWh, and $403/kWh in 2030 and $159/kWh, $226/kWh, and $348/kWh in 2050. Battery variable operations and maintenance costs, lifetimes, and efficiencies are also discussed, with recommended values selected based on the publications surveyed. Figure ES-1. Battery cost projections for 4-hour lithium-ion systems, with values normalized relative to 2022. The high, mid, and low cost projections developed in this work are shown as bolded lines. Figure ES-2. Battery cost projections for 4-hour lithium-ion systems. 0 0.2 0.4 0.6 0.8 1 2020 2025 2030 2035 2040 2045 2050 4 - hr B a t t e ry C os t P roj e c t i ons ( r el at i v e t o 2022) High Mid Low Literature Values 0 100 200 300 400 500 600 2020 2025 2030 2035 2040 2045 2050 4 - hour B a t t e ry C a pi t a l C os t ( 2022$/ kW h ) High Mid Low v This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. Table of Contents 1 Background . 1 2 Methods . 1 3 Results and Discussion . 4 4 Summary . 9 References . 10 Appendix 12 List of Figures Figure ES-1. Battery cost projections for 4-hour lithium-ion systems, with values relative to 2022. . iv Figure ES-2. Battery cost projections for 4-hour lithium ion systems. iv Figure 1. Battery cost projections for 4-hour lithium-ion systems, with values relative to 2022. 4 Figure 2. Battery cost projections for 4-hour lithium ion systems. . 5 Figure 3. Current battery storage costs from recent studies. . 5 Figure 4. Cost projections for power (left) and energy (right) components of lithium-ion systems. 6 Figure 5. Cost projections for 2-, 4-, and 6-hour duration batteries using the mid cost projection. . 7 Figure 7. Comparison of cost projections developed in this report (solid lines) against the values from the 2021 cost projection report (Cole, Frazier, and Augustine 2021) (dashed lines). 14 Figure 8. Comparison of cost projections developed in this report (solid lines) the values from the 2021 cost projection report (Cole, Frazier, and Augustine 2021) (dashed lines), with all values normalized to the “Mid” cost projection in the year 2020. 14 List of Tables Table 1. List of publications used in this study to determine battery cost and performance projections 2 Table 2. Values from Figure 1 and Figure 2, which show the normalized and absolute storage costs over time. Storage costs are overnight capital costs for a complete 4-hour battery system. 13 1 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. 1 Background Battery storage costs have changed rapidly over the past decade. In 2016, the National Renewable Energy Laboratory (NREL) published a set of cost projections for utility-scale lithium-ion batteries (Cole et al. 2016). Those 2016 projections relied heavily on electric vehicle battery projections because utility-scale battery projections were largely unavailable for durations longer than 30 minutes. In 2019, battery cost projections were updated based on publications that focused on utility-scale battery systems (Cole and Frazier 2019), with updates published in 2020 (Cole and Frazier 2020) and 2021 (Cole, Frazier, and Augustine 2021). There was no update published in 2022. This report updates those cost projections with data published in 2021, 2022, and early 2023. The projections in this work focus on utility-scale lithium-ion battery systems for use in capacity expansion models. These projections form the inputs for battery storage in the Annual Technology Baseline (NREL 2022). The projections are then utilized in NREL’s capacity expansion models, including the Regional Energy Deployment System (ReEDS) (Ho et al. 2021) and the Resource Planning Model (RPM) (Mai et al. 2013). 2 Methods The cost and performance projections developed in this work use a literature-based approach in which projections are generally based on the low, median, and highest values from the literature. Table 1 lists the publications that are presented in this work. Because of rapid price changes and deployment expectations for battery storage, only the publications released in 2022 and 2023 are used to create the projections. In addition to the publications in Table 1, we also include a 2020 report by the Electric Power Research Institute (EPRI 2020) for operations and maintenance (O&M) and performance assumptions, but we do not use their cost projection because it was published before 2022. There are a number of challenges inherent in developing cost and performance projections based on published values. First among those is that the definition of the published values is not always clear. For example, dollar year, online year, duration, depth-of-discharge, lifetime, and O&M are not always defined in the same way (or even defined at all) for a given set of values. As such, some of the values presented here required interpretation from the sources specified. Second, many of the published values compare their published projection against projections produced by others, and it is unclear how much the projections rely upon one-another. Thus, if one projection is used to inform another, that projection might artificially bias our results (toward that particular projection) more than others. Third, because of the relatively limited dataset for actual battery systems and the rapidly changing costs, it is not clear how different battery projections should be weighted. For example, should projections published in the latter half of 2022 be given higher weight than those published earlier? Or are some organizations better at making projections or capturing supply chain disruptions, and therefore should be given higher weight? In the interest of providing a neutral survey of the current literature, all cost projections included in this report are weighted equally. As we performed our review of published projections, we found that many of them cited either the previous updates of this report, or they cited the Annual Technology Baseline, which also relies on this cost projection report for its inputs. Thus, 2 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. including all of the latest published projections would create known redundancies (per the second challenge listed above) and were therefore excluded from this work. In some cases, our previous work was provided as a starting point for projections, and then adjustments were made to better capture analysts’ view of battery storage pricing. If that was the case, we considered the projection unique and included it in our survey. Table 1. List of publications used in this study to determine battery cost and performance projections. In several cases consultants were involved in creating the storage cost projections. In these instances we list the consulting firm first, followed by the organization they are supporting. Organization Source AES Indiana AES Indiana 2022 Integrated Resource Plan (AES Indiana 2022) BNEF Bullard (2023) Brattle Newell et al. (2022) Charles River Associates (CRA) / Duke Energy Duke Energy and CRA (2022) E3 / New York Department of Public Service (NYDPS) / New York State Energy Research and Development Authority (NYSERDA) New York’s 6 GW Energy Storage Roadmap (NYDPS and NYSERDA 2022) E Source Jaffe (2022) Energy Information Administration (EIA) Annual Energy Outlook 2023 (EIA 2023) Ascend Analytics / Grant Public Utility District (PUD) Grant PUD Integrated Resource Plan 2022 (Grant PUD 2022) Guidehouse Guidehouse (2021) International Energy Agency World Energy Outlook 2022 (IEA 2022) IHS / PJM Huntington and Wang (2022) Lazard Lazard (2021) Pacific Northwest National Laboratory (PNNL) Viswanathan et al. (2022) Siemens / Public Service Company of New Mexico (PNM) PNM and Siemens (2022) Tri-State Generation & Transmission Association All-Source Solicitation 30-Day Report (2022) Wood Mackenzie Wood Mackenzie (2022) All cost values were converted to 2022$ using the consumer pricing index. In cases where the dollar year was not specified, the dollar year was assumed to be the same as the publication year. When future costs were presented in nominal dollars, they were converted to real dollars using the inflation rate specified by the document. If no inflation rate was found in the document, we found used the inflation rate assumed in other recent documents produced by the same organization. 3 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications. We only used projections for 4-hour lithium-ion storage systems. We define the 4-hour duration as the output duration of the battery, such that a 4-hour device would be able to discharge at rated power capacity for 4-hours. In practice that would mean that the device would charge for more than 4 hours and would nominally hold more than its rated energy capacity in order to compensate for losses during charge and discharge. We report our price projections as a total system overnight capital cost expressed in units of $/kWh. However, not all components of the battery system cost scale directly with the energy capacity (i.e., kWh) of the system (Ramasamy et al. 2022). For example, the inverter costs scale according to the power capacity (i.e., kW) of the system, and some cost components such as the developer costs can scale with both power and energy. By expressing battery costs in $/kWh, we are deviating from other power generation technologies such as combustion turbines or solar photovoltaic plants where capital costs are usually expressed as $/kW. We use the units of $/kWh because that is the most common way that battery system costs have been expressed in published material to date. The $/kWh costs we report can be converted to $/kW costs simply by multiplying by the duration (e.g., a $300/kWh, 4-hour battery would have a power capacity cost of $1200/kW). To develop cost projections, storage costs were normalized to their 2022 value such that each projection started with a value of 1 in 2022. We chose to use normalized costs rather than absolute costs because systems were not always clearly defined in the publications. For example, it is not clear if a system is more expensive because it is more efficient and has a longer lifetime, or if the authors simply anticipate higher system costs. With the normalized method, many of the differences matter to a lesser degree. We defined our low, mid, and high projections as the minimum, median, and maximum point, respectively in 2023, 2024, 2025 and 2030. The minimum and median points were also defined in the same way because the minimum and median projections extended through 2050. The maximum projection in 2030 did not extend through 2050. One projection showed only a 5.8% cost decline from 2030 to 2050, so we used this as the basis for extending the highest cost 2030 projection through to 2050. In other words, the highest cost projection in 2030 was assumed to decline by 5.8% through 2050. Points in between 2025, 2030, and 2050 were set based