Small Magnitude PMU Bad Data Detection Based on Data Mining Technology

Abstract

This paper presents a method to detect bad data from small amplitude Phasor Measurement Units (PMUs). First of all, the characteristics of poor PMU data with small amplitude are analyzed. Then, the advantages of long short-term memory (LSTM) network in processing time series data are analyzed, and the characteristics of time series data selection memory are used to extract the characteristic quantities of PMU amplitude data. Next, a method of using the phase angle slope to amplify the characteristics of the bad phase angle data is proposed. Finally, using the characteristics of the amplitude and phase angle of the PMU data, and using the density-based spatial clustering (DBSCAN) algorithm of the noise-based application for clustering analysis, the effective detection of small-amplitude PMU bad data is realized. Simulation results show that this method can improve the quality of PMU data.