Mining Method of Voltage Sag Association Rules Based on Multi-sources Monitoring Data

With the development of digital technology, multi-platform monitoring data have changed from processed objects into basic resources. This paper summarizes the rules of voltage sag events caused by lightning and mines the association rules with multi-platform data. These rules improve the safety and reliability of the distribution network. However, there is a lack of a method to mine the association rules with lightning and voltage sags. By using data of lightning location system and power quality monitoring system, an association rule mining method, which consists of two main steps, is proposed: 1) establish a database based on multi-platform and build a decision table by discrete attributes, 2) propose the improved AprioriTid algorithm considering nonuniform data to mine rules. A comparative study of multi-platform monitoring data is carried out on the use of the traditional and improved AprioriTid algorithms. And the accuracy of the proposed algorithm is validated by measured data.