Model-Free Non-Invasive Health Assessment for Battery Energy Storage Assets

With the increasing application of battery energy storage in buildings, networks and transportation, an emerging challenge to overall system resilience is understanding constituent asset health. Current considerations focus on adhering technical specification service short term, rather than long-term consequences However, accurately determining health generally requires invasive measurements or computationally expensive physics-based models which do not scale up a fleet assets cost-effectively. This paper alternatively proposes capturing cumulative maloperation through physics model-free proxy for cell health, articulated via strong influence misuse has internal chemical state. A Hidden Markov Chain approach used automatically recognize violations chemistry specific usage preferences from sequences observed charging actions. The resulting methodology demonstrated distribution network level electrical demand generation data, predicting under number technology scenarios.