부산대학교 도서관

Currently， it is challenging to rapidly and precisely locate faults in the secondary systems of intelligent substations through manual analysis of extensive operational data， making it inadequate for meeting the high reliability requirements of smart grids. To address this issue， a fault parameter mapping model for the secondary systems of intelligent substations is proposed. Firstly， a knowledge base for fault localization reasoning is established based on characteristic information to encode fault types. Subsequently， utilizing historical operational data from intelligent substations， a training set for the model is constructed. The support vector machine （SVM） is enhanced by introducing a multi-classifier approach， and its parameters are optimized using the whale optimization algorithm （WOA）. By taking equipment status as input and secondary system fault types as output， a mapping relationship is established between the parameters of secondary system equipment of intelligent substations and the operational state. Finally， the proposed model is validated using actual data as a test set， demonstrating the effectiveness of the proposed model.