A Novel SVP Neural Fusion to Optimize Switching Losses and to Improve Steady State Performance in AC/DC Hybrid Microgrid

Hybrid AC/DC microgrids are gaining growing significance in the integration of renewable energy sources and improving the stability and reliability of power distribution systems. The efficient operation of bidirectional converters within these Microgrids is critical for facilitating power transfer between the AC and DC sections but is often challenged by switching losses. This research paper introduces a novel approach to optimize switching losses in an interlinking bidirectional converter (BDC) within a hybrid AC/DC microgrid by leveraging Space Vector Pulse Width Modulation (SVPWM) and Recurrent Neural Network (RNN). The BDC facilitates power transfer in the AC and DC sections of the microgrid. The proposed method harnesses the predictive control capabilities of RNN to optimize the SVPWM strategy, thereby minimizing switching losses and enhancing overall system efficiency. The novelty of this approach rests in the amalgamation of RNN with SVPWM to dynamically adapt the modulation strategy based on real-time system conditions, resulting in improved performance, reduced switching losses, and enhanced steady-state operation. The efficacy of the proposed approach is substantiated through MATLAB Simulink, underscoring its potential to elevate the performance of hybrid AC/DC Microgrids.