Methodology to Identify Stressor Impacts on Energy Storage System Degradation

The increasing integration of renewable energy and electrification has introduced new challenges for grid reliability, with energy storage systems (ESS) playing a key role in maintaining stability. Lithium-ion batteries are widely used but their long-term effectiveness is limited by degradation. Conventional battery management systems (BMS) estimate state of health (SOH) but cannot directly link operational stressors such as temperature, ramp rate, and state of charge (SOC) to degradation. Here, we introduce an EKF-based method for realtime SOH estimation that simultaneously adapts a stressorinformed degradation model, directly connecting operational conditions to observed capacity loss. The approach is trained using cycling test results from a lithium iron phosphate (LFP) cell over four years. Results show that the EKF provides nearly stable SOH estimates under intermittent measurements, with a degradation model correlated to experimental stressors. These results lay the groundwork for future algorithms that can extract degradation models from dynamic and uncertain data, leading to BMS designs that can leverage this information over the life of the system.