AI Driven (VSI) Voltage Source Inverter for Renewable Energy Applications

Abstract

An Artificial Intelligence (AI)--based controller has been developed for enhancing the power quality of a single-phase Voltage Source Inverter (VSI). This controller employs Artificial Neural Network (ANN) technology trained using Sliding Mode Control (SMC), known for its robustness and performance. Data from SMC is utilized to train the ANN model across diverse load scenarios, encompassing both linear and non-linear loads. The ANN model generates switching signals for Pulse Width Modulation (PWM) switches in the VSI, ensuring optimal transient responses in a short timeframe, thereby enhancing performance and robustness without relying on mathematical models. This approach significantly reduces Total Harmonic Distortion (%THD), a crucial indicator of power quality, even under extreme load conditions, showcasing the durability of the ANN-based controller. This method offers flexibility to adapt to varying load conditions and has the potential to substantially enhance power quality in VSIs. Comprehensive comparisons with conventional and AI-based controllers confirm the superiority of the proposed controller in optimizing power quality. Furthermore, the evaluation demonstrates the stability and effectiveness of the proposed controller in maintaining power quality, even with increased load demands.