Energy storage |
SOC errors in LFP batteries: the $1 million per GWh challenge for BESS
Inaccuracy in state-of-charge (SOC) estimation that can reach up to 15% is a major challenge for LFP-based BESS projects, causing operators to miss out on more than 1$ million dollars per GWh of installed capacity, equating to around 5% of annual trading revenue for BESS in dynamic markets like ERCOT. The traditional BMS (battery management system) is no longer sufficient, but cloud-based predictive analytics can now bring SOC errors to below 2% deviation. This means operators can dispatch with confidence and capture the revenue their assets were built to earn
Darya Rüwald Director of Battery Intelligence at ACCURE
Grid-scale battery energy storage system (BESS) projects are now dominated by lithium iron phosphate (LFP) batteries, where they have overtaken nickel manganese cobalt (NMC) technology. This is due to their cost-effectiveness, durability, reduced safety concerns, and supply chain advantages. But LFP technology comes with a persistent challenge: accurately estimating the battery’s state-of-charge.
Traditional battery management systems (BMS) use two methods to estimate SOC: Coulomb counting; and the voltage method. Coulomb counting tracks current flow, in and out, but is prone to error accumulation. Voltage-based calibration can correct this persistent drift, yet LFP cells have a flat open-circuit voltage (OCV) curve that makes voltage a poor indicator of charge. The result is widespread inaccuracy. For operators, unreliable SOC estimation is more than a nuisance. Their dispatch and trading decisions depend on these values. Overestimation risks overselling power and triggering penalties, causing operators to implement wide safety margins in their trading decisions, directly cutting into usable capacity and potential revenue. Underestimation, on the other hand, leaves capacity idle and unsold.
SOC errors can reach up to ±15% To gauge the true impact of the problem, we investigated SOC field data as part of the 2025 ACCURE Energy Storage System Health & Performance Report. This brought together findings from over 100 commercially operating BESS sites around the globe, sourced from our monitoring database of more than 20 GWh of battery capacity, the world’s largest independent dataset of its kind. We found SOC errors up to ±15%, with some outliers deviating more than 40% from true value. A typical example is shown in Figure 1, which plots SOC estimate (%) vs day. This suggests a daily difference of over 12% between the BMS-estimated SOC and what’s provided by the capability of ACCURE’s SOC Correction feature. This level of misalignment in fast-moving electricity markets, such as ERCOT (Electric Reliability Council of Texas), can reduce annual revenue for a typical BESS by more than $1 million
per gigawatt- hour of installed capacity. For a BESS operating in ERCOT, that’s roughly 5% of annual trading revenue, indicating the scale of impact that might be rectified with reliably accurate SOC estimation. Outside ERCOT, the financial impact can be similar, or even greater. Taking the UK as an
Time (days) SOC Errors: >12% inaccurate daily
Figure 1. Example of SOC estimation errors for a 50 MWh+ LFP BESS. Source: ACCURE.
example, a BESS operator recently lost 7-10% of total revenue over two months to frequency service penalties directly caused by inaccurate energy estimation.
Exploring the hidden cost of SOC inaccuracy and how to correct it To understand the full practical impact of SOC misalignment, ACCURE worked with Gore Street Capital, the international renewable energy and private equity investment manager. Together, we analysed operating data over three months at the 75 MW Dogfish BESS in Pecos, Texas. This 1-hour facility is designed to support the ERCOT power grid and began commercial operation in April 2025. Our analysis uncovered an example that illustrates the issue perfectly. During a two-hour price peak, when the BESS index reached $267/MWh, 3.8 MWh of tradable energy was left idle because the reported SOC did not reflect the true available capacity. By the time the energy management system (EMS) recalibrated, the price window had already closed, and the estimated revenue loss exceeded $1000 in just that single discharge cycle. This was not a one-off anomaly. It reflected a structural mismatch between what the system reported as being available and what the battery could realistically and reliably deliver to market.
32 | May/June 2026 |
www.modernpowersystems.com
To address the issue, we implemented SOC Correction directly through the site’s EMS, using the existing infrastructure with no additional hardware or laboratory testing required. This has allowed Gore Street Capital to improve confidence in available energy and dispatch precision, without placing additional burden on its operations and maintenance teams.
Based on historical analysis, the estimated financial benefit at Dogfish exceeds $110 000 annually, with further optimisation potential already identified.
On a market-wide basis, this corresponds to roughly a 5% uplift relative to average BESS trading revenues in ERCOT.
The improvement comes from more precise dispatch, better spread capture, and more accurate accounting for tradable energy that EMS estimations would otherwise leave as buffer. At the same time, deviation risk declines, reducing the risk of non-compliance penalties and limiting revenue volatility. The result is higher top-line performance with stronger risk mitigation.
Cloud-based predictive battery analytics
The key to improving SOC accuracy beyond the capability of the BMS is to adopt cloud-based
State-of-charge (%)
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
