S-Transform and Scale Zooming Based Classification for Short Duration Power Quality Disturbance According to Partial Similarity

Classification for short duration power quality disturbance (SDPQD) is the premise for electric power quality improvement. A new classification method for SDPQDs is proposed in this paper. It is based on partial similarity and scale zooming of the module time-frequency matrix (MTFM) by Stransform. All types of disturbances with different durations are standardized by time-frequency scale zooming. For templates, one certain part of the MTFM is extracted as the standard template which can represent the disturbance characters to the greatest extent. By comparing the corresponding part of the zoomed MTFM of the test signal with the standard templates, the disturbance type can be well classified according to the principle of maximum similarity. In this paper, six types of single power disturbances and two types of complex power disturbances are discussed and well classified, there is no need to use other complicated classifiers in the whole classification process. The simulation and field test results show that the proposed method is effective to classify disturbance signals in power system.