An Assessment of Battery and Hydrogen Energy Storage Systems Integrated with Wind Energy Resources in California

Publication Number: CEC-500-2005-136
Publication Date: September 2005
PIER Program Area: Energy-Related Environmental Research

The executive summary, abstract and table of contents for this report are available below. This publication is available as an Adobe Acrobat Portable Document Format Files. In order to download, read and print PDF files, you will need a copy of the free Acrobat Reader software installed in and configured for your computer. The software can be downloaded from Adobe Systems Incorporated's website.

---

Executive Summary

California's renewable portfolio standard (RPS) requires the state's electricity generating companies to produce or purchase 20 percent of the electricity they sell from renewable technologies by 2017.^[1] In September 2003, the Energy Commission took that vision a step further and recommended that the goal be met by 2010.^[2] More recently, in support of a policy goal advocated by Governor Schwarzenegger, the Energy Commission suggested pursuing a goal of 33 percent renewable electricity by 2020 to maintain rather than reduce the rate of renewable energy development in California from 2010 to 2020.^[3]

Technologies such as wind turbines and solar photovoltaics are poised to contribute substantially in meeting this goal; however, the widespread acceptance and use of these technologies is hindered by their inability to provide power when the wind is not blowing or the sun is not shining. To help these intermittent renewable technologies become more competitive with fossil and hydroelectric power plants, their output can be stabilized with the use of energy storage systems, which would allow electricity to be produced at times of relatively low economic value and stored so that it can be dispatched at a later time. However, energy storage entails varying economic costs and environmental impacts depending on the specific location and type of generation involved, the energy storage technology used, and the other potential benefits that energy storage systems can provide (e.g., helping to optimize transmission and distribution systems, local power quality support, potential provision of spinning reserves and grid frequency regulation, etc.).

In order to investigate the potential benefits of various advanced energy storage systems in a future California context, the PIER-EA Exploratory Grant Program funded the University of California, Berkeley (UC Berkeley) to conduct a scooping study on the economics and environmental impacts of battery, hydrogen-based, and other advanced energy storage technologies. UC Berkeley researchers reviewed three types of battery technologies (conventional lead acid, advanced zinc bromine, and vanadium redox); hydrogen production, storage, and re-conversion to electricity; hydrogen production, storage, and sale to hydrogen-powered vehicles; compressed air energy storage, pumped hydro energy storage; mechanical flywheels; and superconducting magnetic energy storage. Of these, two battery and both hydrogen systems were analyzed in detail, based on the capabilities of the modeling platform used for the detailed analysis.

The researchers identified four sites in California that are likely to experience significant growth in renewable wind power generation under the statewide RPS and performed economic and environmental analyses of energy storage in the context of those sites. These sites were Altamont Pass and Solano County in Northern California, and Tehachapi and San Gorgonio passes in Southern California. Two time frames were considered: 2010, with 10 percent statewide wind power penetration, and 2020, with 20 percent statewide wind penetration. Based on present and future projected wind resources in the two halves of the state, wind penetration levels were assumed to be 1 percent in Northern California and 9 percent in Southern California (2010) and 2 percent in Northern California and 18 percent in Southern California (2020).

To perform these analyses, researchers used the HOMER model developed by the National Renewable Energy Laboratory (NREL). The model was modified to include hour-by-hour characterizations of the four California wind sites and additional input data to characterize hydrogen production and storage systems and lead acid battery and zinc-bromine flow battery storage systems.

Key Assumptions

Key assumptions guiding this analysis include the following:

• Wind power will expand in California under the statewide RPS program to a level of approximately 10 percent of total energy provided in 2010 and 20 percent by 2020, with most of this expansion in Southern California.


• Costs of flow battery systems are assumed to decline somewhat through 2020 and costs of hydrogen technologies (electrolyzers, fuel cell systems, and storage systems) are assumed to decline significantly through 2020.


• In the case where hydrogen is produced, stored, and then reconverted to electricity using fuel cell systems, we assume that the hydrogen can be safely stored in modified wind turbine towers at relatively low pressure at lower costs than more conventional and higher-pressure storage.


• In the case where hydrogen is produced and sold into transportation markets, we assume that there is demand for hydrogen for vehicles in 2010 and 2020, and that the hydrogen is produced at the refueling station using the electricity produced from wind farms (in other words, we assume that transmission capacity is available for this when needed).

Key Project Findings

Key findings from the HOMER model projections and analysis include the following:

• Energy storage systems deployed in the context of greater wind power development were not particularly well utilized (based on the availability of "excess" off-peak electricity from wind power), especially in the 2010 time frame (which assumed 10 percent wind penetration statewide), but were better utilized–up to 1,600 hours of operation per year in some cases–with the greater (20 percent) wind penetration levels assumed for 2020.


• The levelized costs of electricity from these energy storage systems ranged from a low of $0.41 per kWh - or near the marginal cost of generation during peak demand times - to many dollars per kWh (in cases where the storage was not well utilized). This suggests that in order for these systems to be economically attractive, it may be necessary to optimize their output to coincide with peak demand periods, and to identify additional value streams from their use (e.g., transmission and distribution system optimization, provision of power quality and grid ancillary services, etc.).


• At low levels of wind penetration (1 percent–2 percent), the electrolyzer/fuel cell system was either inoperable or uneconomical (i.e., either no electricity was supplied by the energy storage system or the electricity provided carried a high cost per MWh).


• In the 2010 scenarios, the flow battery system delivered the lowest cost per energy stored and delivered.


• At higher levels of wind penetration, the hydrogen storage systems became more economical such that with the wind penetration levels in 2020 (18 percent from Southern California), the hydrogen systems delivered the least costly energy storage.


• Projected decreases in capital costs and maintenance requirements along with a more durable fuel cell allowed the electrolyzer/fuel cell to gain a significant cost advantage over the battery systems in 2020.


• Sizing the electrolyzer/fuel cell system to match the flow battery system's relatively high instantaneous power output was found to increase the competitiveness of this system in low energy storage scenarios (2010 and Northern California in 2020), but in scenarios with higher levels of energy storage (Southern California in 2020), the electrolyzer/fuel cell system sized to match the flow battery output became less competitive.


• In our scenarios, the hydrogen production case was more economical than the electrolyzer/fuel cell case with the same amount of electricity consumed (i.e., hydrogen production delivered greater revenue from hydrogen sales than the electrolyzer/fuel cell avoided the cost of electricity, once the process efficiencies are considered).


• Furthermore, the hydrogen production system with a higher-capacity power converter and electrolyzer (sized to match the flow battery converter) was more cost-effective than the lower-capacity system that was sized to match the output of the solid-state battery. This is due to economies of scale found to produce lower-cost hydrogen in all cases.


• In general, the energy storage systems themselves are fairly benign from an environmental perspective, with the exception of emissions from the manufacture of certain components (such as nickel, lead, cadmium, and vanadium for batteries). This is particularly true outside of the U.S., where battery plant emissions are less tightly controlled and potential contamination from improper disposal of these and other materials is more likely.

The overall value proposition for energy storage systems used in conjunction with intermittent renewable energy systems depends on diverse factors:

• The interaction of generation and storage system characteristics and grid and energy resource conditions at a particular location


• The potential use of energy storage for multiple purposes in addition to improving the dependability of intermittent renewables (e.g., peak/off-peak power price arbitrage, helping to optimize the transmission and distribution infrastructure, load-leveling the grid in general, helping to mitigate power quality issues, etc.)


• The degree of future progress in improving forecasting techniques and reducing prediction errors for intermittent renewable energy systems


• Electricity market design and rules for compensating renewable energy systems for their output

Conclusions

This study was intended to compare the characteristics of several technologies for providing energy storage for utility grids—in a general sense and also specifically for battery and hydrogen storage systems—in the context of greater wind power development in California. While more detailed site-specific studies will be required to draw firm conclusions, we believe that energy storage systems have relatively limited application potential at present but may become of greater interest over the next several years, particularly for California and other areas that are experiencing significant growth in wind power and other intermittent renewables.

Based on this study and others in the technical literature, we see a larger potential need for energy storage system services in the 2015–2020 time frame, when growth in renewables-produced electricity is expected to reach levels of 20 percent–30 percent of electrical energy supplied. Depending on the success in improved wind forecasting techniques and electricity market designs, the role for energy storage in the modern electricity grids of the future may be significant. We suggest further and more comprehensive assessments of multiple energy storage technologies for comparison purposes, and additional site- and technology-specific project assessments to gain a better sense of the actual value propositions for these technologies in the California energy system.

PIER Program Objectives and Potential Benefits for California

This project has helped to meet PIER program objectives and to benefit California in the following ways:

• Providing environmentally sound electricity. Energy storage systems have the potential to make environmentally attractive renewable energy systems more competitive by improving their performance and mitigating some of the technical issues associated with renewable energy/utility grid integration. This project has identified the potential costs associated with the use of various energy storage technologies as a step toward understanding the overall value proposition for energy storage as a means to help enable further development of wind power (and potentially other intermittent renewable resources as well).


• Providing reliable electricity. The integration of energy storage with renewable energy resources can help to maintain grid stability and adequate reserve margins, thereby contributing to the overall reliability of the electricity grid. This study identified the potential costs of integrating various types of energy storage with wind power, against which the value of greater reliability can be assessed along with other potential benefits.


• Providing affordable electricity. Upward pressure on natural gas prices, partly as a function of increased demand, has significantly contributed to higher electricity prices in California and other states. Diversification of electricity supplies with relatively low-cost sources, such as wind power, can provide a hedge against further natural gas price increases. Higher penetration of these other (non-natural-gas-based) electricity sources, potentially enabled by the use of energy storage, can reduce the risks of future electricity price increases.

Footnotes

1. SB 1078, Sher, Chapter 516, Statutes of 2002.
2. California Energy Commission (2003c), Integrated Energy Policy Report, 2003 Update, Report 100-03-019F, December.
3. California Energy Commission (2004a), Integrated Energy Policy Report, 2004 Update, Report 100-04-006CM, November.

---

Abstract

This exploratory investigation examined energy storage technologies that can potentially enhance the operation of wind power and other intermittent renewable energy systems. We conducted economic and environmental analyses of four energy storage options: (1) lead acid batteries, (2) zinc bromine (flow) batteries, (3) a hydrogen electrolyzer and fuel cell storage system, and (4) a hydrogen storage option where the hydrogen was used for fueling hydrogen-powered vehicles. These were considered under two wind penetration scenarios (2010 and 2020) at four California sites that are likely to experience significant wind farm development.

Analysis with NREL's HOMER model showed that, in most cases, energy storage systems were not well utilized until higher levels of wind penetration were modeled (i.e., 18 percent penetration in Southern California in 2020). In our scenarios, hydrogen storage became more cost-effective than battery storage at higher levels of wind power production, and using the hydrogen to refuel vehicles was more economically attractive than reconverting the hydrogen to electricity. The overall value proposition for energy storage used in conjunction with intermittent renewable power sources depends on multiple factors. Our initial qualitative assessment found the various energy storage systems to be environmentally benign, except for emissions from the manufacture of some battery materials.

---

Table of Contents

Preface.. ii

Abstract vii

Executive Summary 1

1.0 Introduction and Project Overview 6

2.0 Background 6

3.0 Review of Energy Storage Technologies—Cost and Performance 6

4.0 Analysis of Energy Storage in Conjunction with Wind Power in California 6

5.0 Potential Environmental Impacts of Energy Storage Systems 6

6.0 Conclusions 6

7.0 References 6

8.0 Glossary of Abbreviations and Acronyms 6


List of Figures

Figure 1. Five-year California electricity supply forecast 6

Figure 2. Diurnal variation in total wind power output in California versus system load (MW)—January 6, 2005 6

Figure 3. Annual variation in wind power output in Denmark (normalized) 6

Figure 4. Energy storage systems and typical applications 6

Figure 5. Diagram of reversible electrolyzer/fuel cell with hydrogen storage 6

Figure 6. Hydrogen electrolyzer system being installed near wind farm in Palm Springs, California 6

Figure 7. Schematic of a zinc bromine flow battery 6

Figure 8. Major wind energy resources in California 6

Figure 9. HOMER output for electricity generated from wind in 2020 6

Figure 10. Estimates of backup capacity required relative to wind penetration level 6

Figure 11. Conceptual drawing of hydrogen storage in wind turbine tower 6

Figure 12. Cost breakdown for modifying wind turbine tower to include hydrogen storage 6

Figure 13. HOMER output for electricity provided by fuel cell system at the Tehachapi wind farm in 2020 6

Figure 14. Operation of electrolyzer powered by excess wind power from Tehachapi wind farm to produce hydrogen in 2020 6

Figure 15. Number of hours the energy storage system is providing electricity 6

Figure 16. Number of hours the energy storage system is charging or H2 production system is in operation 6

Figure 17. Energy storage system utilization relative to maximum capacity of storage system 6

Figure 18. 2010 scenario—annual energy provided by energy storage systems (MWh) 6

Figure 19. 2010 scenario—bid required to cover annual cost of energy storage system assuming a no-storage system bid of $50/MWh ($/MWh) 6

Figure 20. 2020 scenario—annual energy provided by energy storage systems (MWh) 6

Figure 21. 2020 scenario—bid required to cover annual cost of energy storage system assuming a no-storage system bid of $50/MWh ($/MWh) 6

Figure 22. NOX emissions from California power plants 6

Figure 23. Efficiency and NOX emission curves by utilization rate for combined-cycle natural gas power plants 6


List of Tables

Table 1. Summary of key characteristics of advanced storage systems for energy management applications 6

Table 2. Input data for Scenario 1—hydrogen electrolyzer/fuel cell energy storage with hydrogen storage in wind turbine towers 6

Table 3. Input data for Scenario 2—hydrogen from wind power for sale to hydrogen-powered vehicles 6

Table 4. Input data for Scenario 3—energy storage for wind power with conventional lead acid batteries 6

Table 5. Input data for Scenario 4—energy storage for wind power with zinc bromine flow batteries 6

Table 6. 2010 scenario—annual cost of stored energy ($/MWh) 6

Table 7. 2020 scenario—annual cost of stored energy ($/MWh) 6

Table 8. 2010 scenario—annual cost avoided from electricity production or annual revenue generated from H2 production 6

Table 9. 2020 scenario—annual cost avoided from electricity production or annual revenue generated from H2 production 6

Table 10. Levelized cost hydrogen production system ($/kg of hydrogen) 6

Table 11. 2020 scenario—comparison of the annual value of stored energy: peak/off-peak price structure vs. fixed price structure 6

Table 12. Qualitative environmental impacts of energy storage systems 6