摘要

A new method combining the advantages of time domain and frequency domain analysis methods is proposed for improving the speed and accuracy of PQ disturbances identification. The method consists of seven function modules. By use of d-q transform, generalized morphological filter, Fourier transform and other simple signal processing methods, the amplitude, disturbance time, disturbance frequency domain singular entropy and other features of disturbances are identified and classified layer-by-layer, then a comprehensive judgment is made by classification results of every layer. Simulation results of seven common kinds of single disturbances and some mixed disturbances indicate that the method is effective. ?2009 State Grid Electric Power Research Institute Press.

全文