Neural Network Controlled Grid Synchronization of Single-Phase Inverter Using SOGI-PLL for Efficient Solar Power Integration

This paper presents a comprehensive Artificial Neural Network (ANN)-based control scheme for single-phase grid-connected inverters, emphasizing efficient and accurate synchronization. Using Echo State Networks (ESN) within the ANN framework, this method generates a precise modulating wave for Pulse Width Modulation (PWM) control. Combined with a Second order generalised algorithms (SOGI) Phase-Locked Loop (PLL) system, the approach effectively aligns the inverter's output with grid phase and frequency, achieving low Total Harmonic Distortion (THD). Simulations in MATLAB/Simulink verify the superior performance of this method over traditional Proportional Integral (PI) and PR control systems in terms of steady-state and transient performance.