Design and implementation of fuzzy neural controlled grid interactive inverter

In this study, a single phase inverter which can operate as grid interactive mode to transfer energy generated by renewable energy sources such as wind, solar, fuel cell was designed, and simulation and experimental studies were carried out. Simulation studies were implemented via MATLAB/Simulink and DS1104 dSPACE controller board was used in experimental studies. Inverter was voltage sourced and line frequency transformer and LCL filter were used at the output of the inverter. Inverter was designed as current controlled to minimize the sensivity to phase errors and fuzzy-neural controller was used to control and shape the output current. Membership functions and the rule base of the fuzzy logic controller that controls the output current were determined by using artificial neural networks. System that simulated and implemented was tested with fuzzy-PI and PI controllers along with fuzzy-neural controller to provide the comparison opportunity between the results. There is a parallelism between the results obtained from simulation and experimental studies. Both simulation and experimental results showed that, fuzzy-neural controlled grid interactive inverter operates synchronously with the utility line phase and frequency; the output current that enjected to the grid is in sinusoidal waveform and harmonic level of this current is compatible with the international standarts. The results also showed that inverter has a fast response to the refence variations.