The price of batteries in the quotation letter is usually clear. What is ambiguous are the assumptions surrounding it, and this is where the economics of BESS (Battery Energy Storage System) often fall short. Here are five assumptions that skew the results, based on benchmarks from the national laboratories of the United States. The figures are literature benchmarks, not market prices in Indonesia. Use them to check calculation methods, not to replace vendor quotes.
Price per kWh means nothing without duration
The U.S. national renewable energy laboratory (NREL) states battery system costs in US$/kWh, but not all components scale according to energy. Inverters, for example, scale according to power (kW). The formula is: total cost per kWh equals energy cost plus power cost divided by duration. An example from the report: a 4-hour battery priced at US$300/kWh means US$1,200/kW.
As a result, longer systems appear cheaper per kWh, while shorter ones are cheaper per kW. Two new quotes can only be compared if their durations are the same. NREL itself emphasizes that duration must be stated whenever costs are mentioned.
Round-trip efficiency measured at the connection point
The national lab PNNL defines round-trip efficiency (RTE) as the energy out divided by the energy in, measured at the system connection point. The typical value for lithium-ion in PNNL documentation is 83 percent. NREL chooses 85 percent and explains that this figure already accounts for cooling losses and control equipment.
The two-point difference may sound small. To output 1 kWh, a system with 85 percent efficiency needs to be charged with 1.176 kWh, while a system with 83 percent needs 1.205 kWh. That’s about 2.4 percent more electricity that must be purchased, per cycle, over the system's lifetime. Ask the vendor where their figures are measured: at the cell, on the rack, or at the connection point.
15-year lifespan, not 25
For large-scale lithium-ion batteries, NREL uses a lifespan of 15 years, which is the median value from the publications they reviewed. Spreadsheets that run batteries for 25 years without replacement have already exceeded that median. If a vendor promises a longer lifespan, write it down as a scenario, do not make it a basis.
Degradation: paid now or in installments
NREL places fixed operation and maintenance costs at 2.5 percent per year of the cost per kW. For the earlier example of US$1,200/kW, that amounts to US$30/kW per year. This figure is intentionally at the upper end of the literature because it includes additions or capacity replacements, so the system can continue to output its rated capacity throughout its lifespan. Lower figures generally only cover basic maintenance.
NREL also notes that this assumption is suitable for about one cycle per day. If the battery is used more frequently, that figure may not be sufficient. Check if your spreadsheet has a line for degradation, and how many cycles per day it assumes.
A single cycle figure does not represent your usage profile
Researchers at Sandia tested LFP, NCA, and NMC cells with varying discharge rates, depths of discharge (DoD), and temperatures. The results: even if the cells are used according to manufacturer specifications, cycling conditions have a significant impact. The time to drop to 80 percent capacity varies by thousands of hours, and the number of cycles also varies. Thus, "several thousand cycles" in a brochure is just one point under one test condition. Match it with the DoD, rate, and operating temperature you plan to use.
Summary for the spreadsheet
| Assumption | Questions to ask | NREL Benchmark |
|---|---|---|
| Duration | Are quotes compared at the same hours? | 4 hours; US$300/kWh = US$1,200/kW |
| RTE | Measured at the cell, rack, or connection point? | 85% (PNNL: typical 83%) |
| Lifespan | Calculated over how many years, with replacements? | 15 years |
| Degradation | Is there a line for capacity addition costs? | 2.5%/year of cost per kW |
| Cycles | Applicable for which DoD, rate, and temperature? | About 1 cycle per day |
A spreadsheet that passes all five criteria is not necessarily correct. However, the biggest errors have likely been eliminated before project figures are entered.
Sources
- Cole, W. and Karmakar, A. (NREL, June 2023). Cost Projections for Utility-Scale Battery Storage: 2023 Update, NREL/TP-6A40-85332. https://www.nrel.gov/docs/fy23osti/85332.pdf
- PNNL GridProjectIQ. Energy Storage System Efficiency. https://gridpiq.pnnl.gov/v2-beta/doc/technologies/es/es-efficiency/
- Preger, Y. et al. (Sandia National Laboratories, 2020). Degradation of Commercial Lithium-Ion Cells as a Function of Chemistry and Cycling Conditions, Journal of The Electrochemical Society 167(12). https://doi.org/10.1149/1945-7111/abae37
- PNNL. Energy Storage Cost and Performance Database. https://www.pnnl.gov/projects/esgc-cost-performance