An Efficient and Accurate Method for Detection and Classification of Power Quality Disturbances

Aiming to the problem of large computation amount and high redundancy of traditional S-transform methods available in the detection and classification of transient power quality disturbances, an efficient and accurate method for the detection and classification is proposed. First, the maximum power spectrum dynamics method is used to extract the main frequency points. Then, a called modified incomplete S-transform proposed in the paper is performed on the main frequency points so as to obtain a modulus time-frequency matrix involving the characteristic information of disturbances. The modified incomplete S-transform is implemented by introducing a bi-Gaussian window function with two parameters to replace the Gaussian window function of the traditional S-transform. And then, the characteristic quantities for detecting power quality disturbance and identifying the classification of the disturbances are extracted based on the matrix. Finally, the disturbance parameters such as the amplitude, start time, and stop time of the disturbances are calculated based on the relevant theories and rules, and also the classification of the disturbances is identified by extracting some important characteristic quantities. A great number of simulation experiments are conducted. The results obtained show that by the method, the start time, stop time, and amplitude changes of power quality disturbance signals can be fast and accurately detected, and the classification of the disturbances can be fast and accurately identified, and that the method has the characteristics of less operation time and strong anti-interference ability. (c) 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.