Computation-efficient Approach to EIS Feature Extraction for Battery Informatics and Big Data

Electrochemical Impedance Spectroscopy (EIS) has the potential for improved prediction of battery performance and lifespan, but often has costly computation requirements. Current SOC/SOH prediction methods rely on data-driven or model-based matrix approaches. In advancing towards EIS's big data applications, we propose an efficient and unambiguous curve feature extraction method, surpassing traditional ECM fitting.