A Coupled Mechanical–Electrochemical Study of Li-Ion Battery Based on Genetic Programming and Experimental Validation

Abstract

Lithium-ion batteries (LIBs) are the heart of electric vehicle because they are the main source of its power transmission. The current scientific challenges include the accurate and robust evaluation of battery state such as the discharging capacity so that the occurrence of unforeseen dire events can be reduced. State-of-the-art technologies focused extensively on evaluating the battery states based on the models, whose measurements rely on determination of parameters such as the voltage, current, and temperature. Experts have well argued that these models have poor accuracy, computationally expensive, and best suited for laboratory conditions. This forms the strong basis of conducting research on identifying and investigating the parameters that can quantify the battery state accurately. The unwanted, irreversible chemical and physical changes in the battery result in loss of active metals (lithium ions). This shall consequently result in decrease of capacity of the battery. Therefore, measuring the stack stress along with temperature of the battery can be related to its discharging capacity. This study proposes the evaluation of battery state of health (SOH) based on the mechanical parameter such as stack stress. The objective of this study will be to establish the fundamentals and the relationship between the battery state, the stack stress, and the temperature. The experiments were designed to validate the fundamentals, and the robust models are formulated using an evolutionary approach of genetic programming (GP). The findings from this study can pave the way for the design of new battery that incorporates the sensors to estimate its state accurately.