Understanding Causes of Low Voltage (LV) Faults in Electricity Distribution Network Using Association Rule Mining and Text Clustering

In-depth understanding of a fault cause in electricity distribution network has always been a paramount importance to Distributed Network Operators for reliable power supply. Faults in the network have direct effect on its stability, availability, and maintenance; so, the quick elimination, prevention, and avoidance of fault causes that generated them, are of special interest. Possible opportunity to understanding of the causes and correlation of the factors where future faults may arise can significantly help electricity distribution operators who happen to be accountable to detect and repair such faults. Every distribution network asset has a different level of performance and reliability, and which may vary depend on environmental conditions. Faults identifying in distribution network has rich literature but a very few studies had been done on understanding the factors that contribute to LV Faults using data mining and machine learning techniques. Since there are lack of studies on faults identifying in distribution network with data mining, this study will formulate a starting point. The aim of this paper is to use the association rule mining and clustering techniques to understand the various hidden patterns from the faults database. The uncovered relationship can be represented in the form of Association rules and clusters.