Hybrid Approach for the Development of Compensation Topologies in Wireless Power Transfer Systems

Abstract

The benefits of a wireless power transfer (WPT) system are simplicity, safety, and dependability. Energy transfer coils should be adjusted by electrical measures in applications to increase the system's output power and decrease its input capacity. This manuscript presents a hybrid approach for compensation topologies of WPT system. The hybrid technique is the combination of Lyrebird Optimization Algorithm (LOA) with the Dual Attention Graph Convolutional Network (DAGCN), and then it is together named as LOA-DAGCN technique.The objective of the proposed is to increase the power efficiency in WPT system. The proposed LOA method is used to optimize the optimal controller parameters and generates a set of control signals. The DAGCN is used to predict the power delivered to the load. The proposed technique's performance is put to the test using the MATLAB working environment and contrasted with other methods that are currently in use. The current techniques include Particle Swarm Optimization (PSO), Recurrent Neural Networks (RNN), and Artificial Neural Networks (ANN). According to the results, the suggested method's efficiency is 97%, the current Ann's is 87%, the RNN's is 77%, and the PSO's is 68%.