IOT-BASED ENERGY MANAGEMENT SYSTEM IN ELECTRIC VEHICLES USING OPTIMIZED DEEP LEARNING

Abstract

Electric vehicles are becoming the backbone of smart mobility in smart city applications because of their potential to reduce carbon footprints. In this research, an IoT-based energy management system for EVs by combining the harbor seal whisker optimization (HSWO) and the improved Elman spike neural network (IESNN) has been proposed. The proposed method uses voltage and current sensors on the battery and supercapacitor to transmit real-time energy parameters to the Blynk IoT platform for remote control and monitoring. The proposed IESNN method is used to predict system power demand. Moreover, HSWO is used to tune the network's weight parameters to improve prediction accuracy. IoT integration enables predictive maintenance and real-time data monitoring, enabling users to remotely assess motor performance, energy usage, and battery health. Compared with existing techniques, HSWO-IESNN improves SoC by 11.9%, 9.3%, 7.3%, 5.6%, and 4.4% over IWHO-DL, SCSO-RERNN, EMCABN-ROA, MRA-SDRN, and FBPINN-SAO, respectively.