Electric energy quality composite disturbance classification method

Abstract

The invention discloses a power quality composite disturbance classification method, and belongs to the field of power system automation. According to the method, firstly, traditional empirical wavelet transform is improved, and an MEWT is adopted to analyze and process a power quality composite disturbance signal; secondly, 12-dimensional feature vectors capable of effectively describing different disturbance characteristics are extracted from an MEWT analysis result of the disturbance signals; then, a multi-label composite disturbance classification model with XGBoost as a sub-classifier is constructed based on a problem conversion strategy; and finally, inputting the feature vector into a trained classification model so as to obtain an recognition result of the power quality disturbance. The method can effectively identify normal signals, voltage sag, voltage interruption, voltage sag, harmonic waves, transient oscillation, transient pulses, voltage flicker and composite disturbance formed by the above disturbances, is excellent in classification precision and noise robustness, and has good application prospects and popularization values.

Description

technical field [0001] The invention relates to the field of electric power system automation, in particular to a method for classifying composite disturbances of power quality. Background technique [0002] In recent years, with the widespread access of a large number of nonlinear loads and distributed power sources, the probability of power quality (Power Quality, PQ) disturbance events in the power grid has risen sharply, and the cross-coupling between different disturbances has also made the disturbance types increasingly complex. It poses a serious threat to the safe and economical operation of the power system. In the face of massive disturbance data recorded by monitoring devices, realizing real-time and accurate classification of complex power quality disturbances is a key prerequisite for targeted analysis and governance of power quality problems, so it has important research significance. [0003] After years of research and accumulation, various PQ perturbation c...

AI Technical Summary

Problems solved by technology

Existing research shows that EWT has become one of the effective ways to analyze PQ signals, but when dealing with complex disturbance signals, the accuracy of its frequency band division still needs to be improved

Images