Power output from the PV module changes continuously with time depending upon the climatic condition. This changes are most important in tropical area like Senegal due to the variation of the seasons (dry and rainy). Furthermore, different types of maximum power point tracking (MPPT) algorithm are presented in literature in order to get maximum output from the PV system. They can be summarized in two categories: classical and intelligent methods. The classical methods in no uniform weather condition are not efficient and an important loss of energy is showed. However, faced to this problematics like energy loss and no uniform weather conditions an Adaptative methods is used to optimize the PVs energy. In this study, two intelligent controllers based on artificial neural networks (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) are proposed to optimize the PVs production in non-uniform weather conditions and compared in order to show the most powerful model. For the ANN, the main challenge is to find the optimal neural in the hidden layer and in the paper, it is obtained using evaluator factor like mean squared error (MSE). These techniques using artificial intelligence (AI) algorithms are used for power optimization of a photovoltaic system are trained and validated with real data from a photovoltaic micro power plant in dry and rainy season installed at polytechnic high school of Dakar. The performances of the controllers to optimize the PVs power are evaluated during the dry and rainy seasons. Simulation results show that the ANFIS MPPT controller is more efficient and robust than ANN in non-uniform weather conditions. They have the ability of generalization and adaption to each meteorological conditions. These bullet summarize the applied methodology•A real electrical characteristics of photovoltaic panel are used for learning and validation of the controllers.•A comparative study of the methods in two different season is done.•ANFIS gives best performance in weather conditions compared to the ANN.