Nonlinear neuro-adaptive MPPT controller and voltage stabilization of PV Systems under real environmental conditions

Most PV systems are equipped with classical algorithms such as Perturb and Observe, Hill climbing and Incremental Conductance for Maximum Power Point Tracking Control (MPPT). The simplicity and ease of implementation of these conventional techniques are seen as the main reason of their utilization in PV systems. However, researchers’ attention has, in recent years, been attracted by artificial intelligence-based techniques which can better perform within the bounds of the nonlinearity of PV system characteristics. In this paper, an adaptive nonlinear technique is developed for both MPPT control and voltage stabilization of a Single-Ended Primary Inductance Converter. This control scheme based on Radial Basis function (RBF) neural network is equally used for approximation of unmeasurable or unmeasured variables of the PV system. The main objective of this nonlinear controller is to tract the maximum power and to stabilize the DC output voltage under real environmental conditions. The proposed technique has been numerically tested in a Matlab/Simulink environment under real climatic conditions and load variations. The close-loop stability of the controller is verified by Lyapunov’s theory and the proposed algorithm gives satisfactory results compared to Extremum Seeking Control-based MPPT used in the same conditions.