ANN-Robust Backstepping MPPT Based on High Gain Observer for Photovoltaic System

A hybrid maximum power point tracking (MPPT) technique has been proposed in this paper for a standalone photovoltaic (PV) system in order to extract the maximum power from PV panel. This technique is composed of two performant controllers, the first one, which is an intelligent method based on the Artificial Neural Network (ANN), is trained to rapidly estimate the optimum voltage under different changes of meteorological conditions, while the second one, that is the robust backstepping controller, is conceived to track the optimum voltage by offering high robustness against disturbances as well as the desired tracking criteria. To minimize the PV system cost, the required sensors' number, their measurement error and the system complexity, a high gain observer (HGO) has been proposed and applied to estimate the state variables of the system by observing the boost inductor current, the PV voltage and the load voltage basing only on data provided by the control law and the PV current and voltage. This approach minimizes the need for additional sensors in practical scenarios, as they come with several drawbacks. By avoiding the use of these sensors, the system can avoid being bulky and costly. The studied PV system was simulated in MATLAB/Simulink to verify its efficiency and robustness even under severe and different weather conditions. Additionally, the proposed MPPT technique was compared with two other techniques: the conventional Perturb and Observe (P&O) technique and the hybrid technique incorporating the Incremental Conductance and Backstepping control (INC-BSC) to prove its efficiency. The results demonstrate how effectively the suggested control system can rapidly detect and precisely track the desired maximum power point, with a response time of approximately 0.05 seconds.