Analysis and Investigation on Causes of Voltage Sag Based on A Novel Apriori Algorithm

Hao Liu,Rongrong Ma,Lei Li,Shaohui Liu,Qinghui Zeng,Zhuojun Wu,R Weerasinghe,C.-H Weng,J Wu

doi:10.1051/e3sconf/202019403015

Abstract

As more and more power users are increasingly demanding the quality of electricity, the losses caused by voltage sags are becoming more and more serious. Therefore, it is very important to analyze the cause of the voltage sag to prevent the voltage sag in time. This paper proposes a new algorithm that combines Apriori correlation analysis algorithm and cluster analysis algorithm to analyze the causes of voltage sags. Because some typical climatic conditions also have an important influence on the cause of voltage dips, The data is initially processed using climate factors as clustering indicators, and then the correlation analysis between typical electrical characteristics and voltage sags is performed, and strong association rules are finally obtained. According to the calculation and analysis of examples, some factors with high correlation with the causes of voltage sag are found, which will provide theoretical support for the prevention of voltage sag and provide ideas for further research on the causes of voltage sag.

Highlights

With the high-speed development of smart electric power and energy systems [1]-[3], smart grids become the development trend of future power systems [4], and more and more artificial algorithms are gradually applied to power grids [5]
The hidden features in the data that are strongly related to the occurrence of voltage sag can be discovered, which can provide theoretical support for studying the occurrence of voltage sag and contribute to the improvement of power quality
This article first divides a large number of historical voltage sag data into three categories based on climate factors through cluster analysis

Summary

Introduction

With the high-speed development of smart electric power and energy systems [1]-[3], smart grids become the development trend of future power systems [4], and more and more artificial algorithms are gradually applied to power grids [5] Under this background, high-end manufacturing industry becoming more and more sensitive to the unavoidable voltage sag events in power supply system, voltage sag has caused huge losses to users, so it has become the focus of industry and academia[6]. The results show that climatic factors, the nature of the switching load, voltage level, area, time and other factors have a certain influence on the type of voltage sag. This analysis result provides a more specific theoretical basis for the prevention of voltage sag

Algorithm Principle

Apriori Correlation Analysis Algorithm

Clustering Analysis Based on K-means Algorithm

Case analysis

Findings

Conclusions