A Novel Rice Density-Neural Network Controller Implementation in the Electric Vehicle Wireless Charging System for the Performance Enhancement

Abstract

Despite significant advancements in infrastructure, the process of charging electric vehicle batteries (EVBs) continues to face critical challenges that hinder widespread adoption and operational efficiency. Charging stations are often unevenly distributed, leading to accessibility issues, especially in densely populated or remote areas. Moreover, the charging process itself is constrained by slow speeds, energy losses, and limited compatibility across vehicle models and charger types. Batteries with low state of charge (SOC) require prolonged charging durations, further exacerbating inefficiencies. Additionally, the increased load on local power grids during peak charging hours raises concerns about stability and safety of the grid. In order to address these challenges, this article introduces an inductor–capacitor–capacitor (LCC) wireless charging approach for EVBs. The proposed LCC compensation topology overcomes the aforementioned technical limitations by introducing a novel hybrid rice density-neural network (RD-NN) controller to enhance charging efficiency, reduce dependency on wired connectors, and improve overall performance. Also, the switching of converters within the system is expressed by a multi-objective function that studies SOC, voltage, and efficiency under defined constraints. The optimization problem is addressed through metaheuristic algorithms like particle swarm optimization (PSO), genetic algorithm (GA), and the proposed novel RD-NN controller. Compared to conventional optimization techniques such as PSO and GA, the proposed RD-NN controller exhibited superior performance, achieving an energy efficiency of 96.2% and an optimal SOC of 95.2%. Integration of the rice density technique with NNs better optimizes the non-linear behaviors of EVB charging and enhances control responsiveness. The stability of the RD-NN controller is further validated using the Routh-Hurwitz (R-H) criterion, confirming its robustness under dynamic operating conditions. This novel approach of the RD-NN controller, coupled with LCC wireless architecture, presents an optimum solution in EVB charging by realizing superior energy transfer, adaptive control, and robust system stability, positioning it ahead of conventional methods.

Cite this article as: S. Rathee and G. L. Pahuja, "A novel rice density-neural network controller implementation in the electric vehicle wireless charging system for the performance enhancement," Electrica, 26, 0055, doi: 10.5152/electrica.2026.25055.