Innovative neural network and fuzzy logic control techniques for single-phase grid-connected photovoltaic systems using dual-core DSP microcontroller in smart home applications

The growing global demand for renewable energy has increased the need for efficient and reliable control systems in photovoltaic (PV) applications, ensuring optimal energy extraction and stable grid integration under varying environmental conditions. This paper conducts a detailed analysis of both simulated and practical implementations of a system that integrates a photovoltaic (PV) panel, a DC-to-DC boost converter, and a DC-to-AC inverter. The control of the boost converter is handled by an Intelligent Artificial Neural Network (IANN), and the inverter operation is managed by a Fuzzy Logic Controller (FLC). Notably, the FLC achieves a Total Harmonic Distortion (THD) of just 1.63%, substantially better than the 2.56% THD observed with the MPC algorithm, and it maintains stable output voltage even under variable shading conditions, outperforming both PSO and P&O methods. Extensive simulations carried out in MATLAB-Simulink provide a comprehensive analysis and discussion of both the simulation and experimental results. Furthermore, the development, implementation, and evaluation of electronic circuits (PCB boards) demonstrate their effectiveness in facilitating the seamless integration of PV systems with the electrical grid. The process-in-the-loop (PIL) structure employed in the programming phase significantly aids in debugging, thus enhancing the operational efficiency of the system and simplifying the resolution of software issues. The proposed hybrid technique shows superior results in various performance metrics, achieving a maximum power efficiency of 99.99%, a relative error of 0.000001, and a minimum tracking acceleration of 0.013 s. This study showcases the latest developments in control strategies, enhancing grid compatibility and overall system performance in photovoltaic applications.