Farahnaz Ahmadi - Department of Electrical Engineering, University of Kurdistan, Kurdistan, Sanandaj, Iran
Shahram Karimi - Department of Electrical Engineering, University of Razi, Kermanshah, Iran
Mehdi Kolahkaj - Department of Electrical Engineering, Islamic azad university central Tehran branch

The output power of solar modules depends on many factors such as solar irradiation, shading and ambient temperature. Due to non-linear nature of the photovoltaic (PV) system and variable environmental conditions, it is possible that solar modules of the PV system are affected by partial shading conditions, which results in incomplete operation of the control algorithm. This paper presents an artificial intelligence-based technique for delivering the maximum power of the PV power generation system through an interconnect circuit and in an independent operation. The interface circuit between the PV dc source and the load is accomplished by a DC-DC boost converter. The proposed control technique is the maximum power point tracking (MPPT) using an adaptive neuro-fuzzy inference system (ANFIS). In order to demonstrate the accepted and desirable performance of the ANFIS technique, a comparison between this control technique and the two conventional techniques of perturbation & observation (P&O) and incremental conductance (INC) is carried out. In order to validate the NFIS control scheme ANFIS, a simulation has been carried out by three scenarios and for different weather conditions. The simulation results confirm the rapid and accurate dynamic response of the ANFIS technique.

Photovoltaic module, Adaptive neuro-fuzzy inference system, Maximum power point tracking, Conventional technique, Simulated real time.