Adaptive neuro-fuzzy inference systems based estimation of photovoltaic cell parameters using data set obtained outdoor measurements of photovoltaic module

Solar energy, which will play an important role in energy supply in the near future, is the primary energy source among renewable energy sources. Recently, solar energy systems have become one of the most interesting topics due to their many advantages. By using experimental data, to extract the parameters of photovoltaic (PV) panels which is a solar energy system plays an important role in the design, evaluation and efficiency of PV panels. In this context, studies on obtaining different operating conditions and performance characteristics of PV panels have increased remarkably in recent years. In this study, unknown parameters of single diode model equivalent circuit, also known as five parameter models of photovoltaic cell; photo current (I_ph), diode saturation current (I_0), diode ideality factor (n), serial resistance (R_s) and parallel resistance (R_p) were tried to be estimated with ANFIS using experimental data. Manufacturer datasheet information and ANFIS output results are compared in the MATLAB/Simulink environment; I-V and P-V characteristics have been obtained. Simulation results showed that the use of ANFIS in the extraction of unknown parameters of the PV module is a useful tool. Key words: PV Model, Parameters of PV Cell, ANFIS