| Abstract: |
Solar Photovoltaic (PV) power systems are widely integrated with distribution networks providing autonomous and microgrid functionality for improved performance, stability, and energy efficiency under sufficiently capable intelligent control techniques. The study was conducted for a 500 kWp rooftop PV installation (twin beam PV technology) commissioned at an industrial site in central India, with real-time SCADA data acquisition at one-minute intervals over twelve consecutive months. In this contribution, a hybrid control architecture, with the ICA-MPC framework for grid-side power regulation and the PSO–INC hybrid algorithm for MPPT, is proposed. Experimental results show that the proposed intelligent control scheme accomplishes an overall maximum conversion efficiency of 98.2%, a total harmonic distortion (THD) of only 1.43%, a voltage regulation within ±0.8% and a system-wise operating availability of 98.4%, resulting in a 15.9% increase in annual energy yield and a 73.5% reduction in reactive power losses compared to conventional PI-based control. The improved technique is shown to outperform four benchmark controllers through statistical analysis with multiple linear regression (R² = 0.971) and one-way ANOVA (F = 312.47, p < 0.001). A comparative evaluation with the state-of-the-art literature validates that the proposed ICA-MPC hybrid architecture enhances all the relevant performance indices over the existing approaches. This study provides a validated, data-based design framework for large grid-connected renewables systems in developing countries. |
