ANN-based MPPT for photovoltaic systems: performance analysis and comparison with nonlinear and classical control techniques

Abstract

In photovoltaic energy systems, maximum power point tracking (MPPT) techniques are essential for optimizing power output under changing climatic conditions. Several techniques have been proposed in the literature, including classical techniques such as perturb and observe (P&O) and incremental conductance (INC), nonlinear controllers such as backstepping, and artificial intelligence-based techniques like fuzzy logic. This study compares the performance of an artificial neural network (ANN)-based MPPT approach with these nonlinear and classical MPPT techniques. It analyses the advantages and limitations of the various techniques to evaluate their performance in terms of efficiency, accuracy, and output power stability under changing climatic conditions. The study aims to help researchers select the most effective technique to improve the efficiency of photovoltaic systems. The simulation was carried out using MATLAB/Simulink. The simulation results indicated that the artificial neural network achieved better performance than the other techniques in terms of tracking speed, with an efficiency of up to 99.94%, while maintaining stable output power under changing climatic conditions. The backstepping controller also showed stable output power compared to traditional techniques. Fuzzy logic had a lower efficiency than both the artificial neural network and backstepping. Perturbation and observe and incremental conductance are easy to implement, but they showed oscillations around the maximum power point, which reduces the overall efficiency of the system.