Improved wild horse optimizer with deep learning enabled battery management system for internet of things based hybrid electric vehicles

Internet of Things (IoT) become an emergent platform in wireless technologies in design of electric vehicles (EVs) and hybrid electric vehicles (HEVs). Dynamic energy storage systems, batteries can be damaged due to overcharging/discharging and their mass penetration deeply affects the grid. For circumventing the likelihood of damage, the EVs and HEVs require an accurate state of charge (SOC) estimation approach for improving the lifetime and safety. An efficient battery management system (BMS) remains a challenging problem in HEVs, commonly utilized to indicate the battery state-of-charge (SOC). As over-charge and over-discharge outcome from predictable damage to the batteries, precise SOC estimation model is needed for HEVs. In this aspect, this study presents an improved wild horse optimizer with deep learning enabled battery management system (IWHODL-BMS) for IoT based HEVs. The presented IWHODL-BMS employs attention based bidirectional long short-term memory (ABiGRU) approach to accurately estimate SOC in HEVs. For enhancing SOC estimation performance of the ABiGRU technique, the IWHO algorithm is utilized as hyperparameter optimizer. The application of the ABiGRU model results in simpler and accurate representation of the input. A comprehensive simulation result portrayed the enhanced outcomes of the IWHODL-BMS model over the other methods under varying measures.