AUTHORS:
Abhay Chhetri Electrical Cluster, University of Petroleum & Energy Studies, Dehradun, India
Mayank Saklani Electrical Cluster, University of Petroleum & Energy Studies, Dehradun, India
Devender Kumar Saini Electrical Cluster, University of Petroleum & Energy Studies, Dehradun, India
Monika Yadav Electrical Cluster, University of Petroleum & Energy Studies, Dehradun, India
Yogesh Chandra Gupta Electrical Cluster, University of Petroleum & Energy Studies, Dehradun, India

ABSTRACT:
It is known that the State of Charge (SOC) of an electric vehicle’s (EV) battery is the most essential parameter. It provides information about the battery's performance, allowing us to estimate the battery's charging and discharging capacities. Therefore, with an understanding of SOC characteristics, battery life, and EVs range can be enhanced. A variety of approaches to estimate the battery SOC % are presented in the literature, however, the filter-based method provides the most accurate results. This study provides an Extended Kalman Filter (EKF)-based SOC estimation coupled with other errorreduction strategies, including Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Battle Royale Optimization (BRO). The data used for this estimation is derived from the existing data model utilizing a Turnigy battery's LA92 drive cycle. It is observed that based on the given data set, the EKF provides an error percentage of 1.944% whereas, the proposed methodology of EKF with optimization technique especially PSO outperforms the EKF by attaining 1.8% error reducing the EKF SOC estimation error by 0.14%.

Keywords: Proportional Integral Derivative (PID), Swarm Intelligence (SI), Renewable Energy (RE), Statistical Analysis

Conference Venue: Mövenpick Hotel & Apartments Bur Dubai, Dubai-UAE
Conference Date: 29-30 April 2023

ISBN Number: 978-625-00-1526-1
DOI Number: https://doi.org/10.53375/icmame.2023.407


PDF Download