Research On Power Quality Pollution Early Warning Model Of Enterpirse Based On Data Mining

In this paper,we propose to set up a exploratory analysis of L TechnologyCompany about power quality pollution early warning model, based on thedatabase of Taigu power company monitoring system and the reference ofpower quality study. Firstly, the paper introduces the current status of researchand basic concept of power quality pollution; In addition, uses the descriptivestatistics to analyze each monitoring point of L Technology Company powersystem from the macroscopic, and takes Apriori algorithm to construct thenetwork of power quality pollution from the microcosmic, in order to search thepotential contact between each monitoring point; Moreover, uses the CHAIDalgorithm of decision tree to perform a detailed analysis of the variousmonitoring points power quality pollution early warning model based on powerquality indicator, power indicator, daily indicator and the created network ofpower quality pollution, in order to explore the important variables affecting themonitoring points and summarize the logical expression of power qualitypollution’s occurrence,and effectively grasp the characteristics of eachmonitoring point and the problem where lies.The paper takes an empirical analysis on each monitoring point of the LTechnology Company through the statistical knowledge and data mining, itconcludes: the network of power quality pollution is a premise to power qualitypollution early warning model; master the important impact variable of themonitoring points is the key to the early warning model; sum up the goallogical expression is an important measure to the model. We hope that theresults of this study can improve the development of the power system, andprovide a reference to a more rational use of power quality market, betterpower service and power quality use.