Steady-state Power Quality Anomaly Recognition Based on Time Series Trend

Abstract

In order to improve the quality of power supply to the park and conduct differentiated service and management, the influence of the power quality (PQ) level of the park needs to be considered. At the same time, the traditional steady-power quality anomaly identification method only compares the value with the limit value and does not consider the change trend of the data, which has some limitations. The characteristics of time-series trend changes of steady-state power quality data is focused on this paper and a steady-state power quality anomaly identification method based on time series trend analysis is proposed. Firstly, data preprocessing is carried out through piecewise linearization to filter out data fluctuations and retain the main trend change characteristics of data. Secondly, the trend change of data is represented by the trend pattern, and the similarity between different trend sequences is calculated by the pattern distance. Finally, combined with the amplitude anomaly index, the comprehensive anomaly index of the data to be identified relative to the normal data segment is calculated to identify whether there are anomalies in the steady-state power quality of the corresponding measurement point. Through simulation examples and case analysis, it is proved that the proposed method is accurate, applicable and easy to implement, and can be easily integrated into the existing power quality monitoring system.