Inaccurate battery state of health measurements impact revenue, safety, operations

A report from Brussels-based 3E, a renewable energy technology and software-as-a-service (SaaS) company, unpacks why SoH is a critical measure of battery performance, reliability, and lifespan; breaks down SoH estimation techniques; and explores how digital twin models can enhance SoH monitoring and analysis. 

SoH refers to the ratio between a battery’s actual storage capacity and its original, nameplate capacity. Though the metric cannot be directly measured, it plays a large role in how companies maintain BESS and perform financial modeling and valuation.  

The 3E report explains SoH is directly tied to battery degradation, which is influenced by a variety of thermal, chemical, and electrical processes that evolve as a battery changes overtime and experiences calendar and cyclic aging. As batteries degrade for many reasons, including high operating temperatures and high resting states of charge (SOC), finding a way to accurately measure SoH is critical.  

Estimating battery SoH is difficult, however, due to variability in battery chemistries and among manufacturers and operating conditions. 

Experimental, lab-based techniques such as charge and discharge tests can be valuable ways to produce accurate results without high amounts of computation required but their results are generally difficult to replicate in the field. By contrast, modeling is flexible, scalable, and ideal for real-world applications but is data-intensive.

Linear estimates

Some battery management systems (BMS) use linear, cycle-based estimation as a proxy measure. That method provides limited insight, however, as it doesn’t measure depth of discharge or thermal conditions, among other factors. It also masks degradation mechanisms, making it difficult to find problems until it can be too late.

The 3E report noted the risk of inaccurate SoH measurement. Thermal runaway becomes a larger risk as batteries approach their end of life, and SOC calculations are less accurate without firm SoH data. A mismatch between a battery’s estimated and actual SoH can have significant financial consequences, as evidenced by one example cited in the report, where there was an 8% difference. 

“Considering an installation of 50 MWh, day-ahead prices of €50 ($54)/MWh and approximately one cycle per day,” the report stated, “This 8% mismatch can result in around €73,000 of lost revenue because of an inaccurate estimation.” 

The report also suggested several corrective, preventive, and predictive strategies to decrease BESS degradation and proactively monitor SoH.  

Firstly, regularly updating BMS and repairing auxiliary systems such as temperature controls ensures that improvements can be implemented properly. Though it is possible to instantly increase a system’s overall SoH by replacing individual modules with low SoH, the report noted that this can substantially increase operating expenses and should be a last resort.  

Additionally, regular maintenance and frequent capacity testing can optimize resources, improve information sharing among stakeholders, and provide qualitative insights. That does require taking BESS offline, however, so the potential for negative financial implications remains.  

The 3E report also suggested modeling degradation with physics-based digital twins. That cost-effective approach can measure simultaneous thermal, electrical, and chemical changes to provide continuous, measurable monitoring of a battery’s degradation and can map how different battery components interact to improve or worsen degradation rates.

From pv magazine USA.