Fault Classification and Warning of Substation Area Based on Fusion IGWO Clustering and Prediction

The ongoing evolution of the conventional energy sector is inextricably linked to the advent of intelligent grid and energy internet technologies. The effective utilisation of generated power data represents a pivotal aspect of this process. The objective of this study is to collect target-related power characteristics through the mining of raw data, with the aim of constructing a substations’ fault classification and warning system through the application of clustering prediction methods. The K-mean clustering algorithm, which incorporates the improved grey wolf algorithm, is employed to cluster and analyse the feature labels in order to obtain the warning boundary value. Concurrently, the IGWO-long and short-term memory network algorithm is utilised to predict the labelled data, and a comparison is made in order to obtain the warning results. To demonstrate the efficacy of the proposed methodology, we present a case study examining the operational data of 256 distribution substations in Shanghai.