Alarm Event Classification Based on Edit Distance and Hierarchical Clustering in Electricity Substation

Substation alarm events are numerous and not classified effectively, which makes operation and maintenance personnel face great challenges. Aiming at the problems of various categories of substation alarm information, a classification method of substation alarm events based on edit distance and hierarchical clustering is proposed. Firstly, the SCADA system configured in the substation is used to collect the alarm messages and construct the alarm information data set. Secondly, the alarm information is translated and processed. Then, the string distance between every two alarm messages in the alarm message data set is calculated, and a string distance matrix is constructed. On this basis, the hierarchical clustering method is used to cluster the alarm information and obtain the tree diagram. Finally, the classification results of substation alarm events are given according to the number of required categories. The method can be used in the technical field of automatic operation and maintenance of the transformer substation to help maintenance personnel sort out alarm events and improve the operation and maintenance efficiency of the transformer substation.