Modeling of Grid-Connected SPV System and Performance Analysis Using Advanced Controller Under Varying Environmental Conditions

This study investigates the dynamic performance of a 255kW grid-connected solar photovoltaic (SPV) system through mathematical modeling and simulation. A novel hybrid algorithm, the Adaptive Grasshopper Optimization Algorithm combined with a Recurrent Neural Network (AGO-RNN), is employed to evaluate the system's performance under three distinct scenarios: (i) standard temperature and irradiance (G), (ii) varying temperature and irradiance (G), and (iii) simultaneous changes in temperature and irradiance (G). The AGO-RNN approach is utilized to calculate the Total Harmonic Distortion (THD) in each scenario, with the results compared against those achieved using the Particle Swarm Optimization (PSO) method. The proposed system incorporates a 255 kW solar PV model, Maximum Power Point Tracking (MPPT), and a three-level Neutral Point Clamped (NPC) inverter. Simulation outcomes demonstrate that the AGO-RNN controller reduces THD by 1.6% compared to PSO, meeting the requirements of IEC 61727 by minimizing current magnitude and ensuring improved grid performance.