Model-based state of health estimation of a lead-acid battery using step-response and emulated in-situ vehicle data

Abstract

Abstract Lead-acid (PbA) batteries are one the most prevalent battery chemistries in low voltage automotive applications. In this work, we have developed an equivalent circuit model (ECM) of a 12V PbA battery while preserving the major dynamics of a semi-empirical model we have developed previously. Thereafter, two batteries are aged according to a modified IEC 60095-1 standard and the evolution of ECM parameters investigated. The trend derived from these aging campaigns are used to determine the state of health (SOH) of three either new or aged on the vehicle batteries. The contribution from this approach stems from the fact that the ECM model while being simple is capable of approximating the sum of charge transfer resistances of both electrodes from a short step response. In addition, another SOH estimation technique is developed based on frequency response analysis of a metric dubbed instantaneous dynamic resistance derived from voltage response of the battery during engine cranking. Real world voltage response of the battery during cranking is imitated in the lab and superposed on IEC 60095-1 standard profile, according to which another battery is aged. This data is used to assess the performance of SOH estimation technique.