Hybrid Neural Network and Adaptive Terminal Sliding Mode MPPT Controller for Partially Shaded Standalone PV Systems

This research presents a new method for controlling the maximum power point tracking (MPPT) of solar photovoltaic (PV) systems that are partially shaded. The proposed approach uses a neural network and an adaptive terminal sliding mode controller (NN-ATSMC) to ensure that the PV system operates at optimal performance under uncertain conditions. The NN-ATSMC controller is applied to a DC/DC boost converter to drive the system to the maximum power point (MPP). This method ensures that the error will converge in finite time and the chattering effect will be minimized without losing robustness under various disturbances and load conditions. Simulation results show that the proposed NN-ATSMC controller performs better than other types of controllers existing in the literature, such as a sliding mode controller (SMC) and a conventional proportional-integral controller (CPI). For the validation of the proposed controller, control hardware-in-the-loop (C-HIL) experimental implementation has been carried out through Texas Instruments digital signal processor C2000. The experimental results show the viability of real-time implementation and verify the effectiveness of the proposed method, which ensures the low cost and stability of the standalone PV systems.