Intelligent control of soft switching solid-state transformer in renewable microgrids using quokka swarm ANN

With the growing adoption of Renewable Energy Sources (RESs), the demand for efficient power management and conversion in microgrids has become crucial. This study presents a Hybrid Renewable Energy System (HRES)-based microgrid integrating Photovoltaic (PV), Doubly Fed Induction Generator (DFIG)-based Wind Energy Conversion System (WECS) and Battery Energy Storage System (BESS). A Soft Switching Solid-State Transformer (SS-SST)-based power conversion architecture is employed to enhance power stability, improve efficiency, and ensures seamless grid integration by contributing voltage regulation and power flow optimization. The Quokka Swarm Optimized Artificial Neural Network (ANN) Controller is implemented for dynamic duty cycle and voltage regulation, ensuring optimal performance under varying load conditions. The Pulse Width Modulation (PWM) generator is tuned using the Quokka Swarm Optimization (QSO) algorithm, significantly improving the control accuracy and reducing power losses. The proposed system is validated through MATLAB simulations and laboratory implementation, demonstrating enhanced voltage regulation and improved power efficiency compared to conventional controllers. The proposed QSO-ANN achieves a settling time of 55 ms, overshoot of <3%, steady-state error below 1% and efficiency up to 97%. The grid current Total Harmonic Distortion (THD) is reduced to below 0.1% in simulation and maintained within 1.5–1.8% in hardware experiments, complying with IEEE-519 standards.