Power Quality Improvement using DVR based Adaptive Neuro-Fuzzy Inference System (ANFIS)

PI controller is very common in the control of DVRs. However, one disadvantage of this conventional controller is its inability to still working well under a wider range of operating conditions. So, as a solution fuzzy controller is proposed in literature. But, the main problem with the conventional fuzzy controllers is that the parameters associated with the membership functions and the rules depend broadly on the intuition of the experts. To overcome this problem, Adaptive Neuro-Fuzzy Inference System (ANFIS) based controller design is proposed. The resulted controller is composed of Sugeno fuzzy controller with two inputs and one output. According to the error and error rate of the control system and the output data, ANFIS generates the appropriate fuzzy controller. The simulation results have proved that the proposed design method gives reliable powerful fuzzy controller with a minimum number of membership functions. The Dynamic Voltage Restorer (DVR) is fast, flexible and efficient solution to voltageproblem. The DVR is designed for protecting the whole plant with loads in the range of some MVA. The DVR can restore the load voltage within few milliseconds. The simulation results are obtained through MATLAB/SIMULINK software.