A new Approach of Power Quality Disturbance Classification Based on Rough Membership Neural Networks

Building a power quality monitoring and analysis system is important to improve power quality and avoid equipment damage. A new approach for power quality disturbance classification based on linear time-frequency distribution and rough membership neural networks is presented in this paper. Taken the advantages of windowed Fourier transform and S-transform, the approach presented five features to characterize the disturbance signals, than classify them with rough membership neural networks. The simulation results of 7 common kinds of disturbances indicate that the method has good performance of accuracy and efficiency.