Estimation of State of Charge (SOC) for Lithium Ion Batteries using Tunicate Swarm Optimization Technique

Abstract

Lithium ion batteries are immensely used in Electric Vehicles (EVs) because of their special characteristics like high energy density, low self-discharging capability, high voltage, and long life spans. State of Charge (SOC) of a battery is a determining factor of charge left in the battery. A Battery Management System (BMS) is a model that monitors the SOC of the battery from draining out. Keeping track of the SOC requires adaptive estimation of SOC from time to time. An accurate estimation of SOC is needed for prolonging the battery life and for preventing the battery failure. In this research work, a Resistor-Capacitor (RC) circuit based Lithium ion battery model is developed, and a new metaheuristic model based SOC estimation is done by employing Tunicate Swarm Optimization (TSO) technique. The TSO algorithm is found to be superior over several metaheuristic algorithms because of its robustness, innate multi-swarm and self-adaptive nature. The proposed novel TSO based SOC estimation model aims to attain a balance between speed, accuracy, and complexity of SOC estimation. The proposed model is evaluated for its performance and the SOC error of estimation is found within the range [-1, 1]. The proposed SOC estimation framework is compared with existing works to prove its superior performance