Intelligent MPPT for photovoltaic panels using a novel fuzzy logic and artificial neural networks based on evolutionary algorithms

Abstract

Abstract Maximum power point tracking (MPPT) represents one of the significant challenges for designing photovoltaic (PV) systems. Thus, an effective MPPT method of solar panels is required to make them more efficient. Here, four intelligent methods have been applied for MPPT. Fuzzy logic (FL) has been used without the knowledge of an expert to create membership functions and rules. Also, the artificial neural network (ANN) has been employed based on three meta-heuristic algorithms, including genetic algorithm (GA), particle swarm optimization (PSO) algorithm, and imperialist competitive algorithm (ICA). The required data have been received from a solar panel and utilized in the designed systems in MATLAB software. In this case, the ambient temperature and irradiance were considered the systems’ inputs, while the maximum power was regarded as the output. The systems’ accuracy was evaluated using two statistical indices, root mean square error (RMSE) and mean absolute error (MAE). Additionally, they were compared based on stability, speed, and complexity. Eventually, the obtained results specified that the creatively designed fuzzy system provides faster, more accurate, and more stable performance than the other methods. It is also less complicated to implement. Regarding the hybrid methods, the results showed that the ANN-based on ICA is faster however more complicated in implementation compared to the ANN-based on PSO and GA. Yet, in terms of accuracy and stability, the hybrid methods are not significantly different.