A novel data-driven method for maintenance prioritization of circuit breakers based on the ranking SVM

Abstract As a critical switching component, circuit breakers (CBs) require regular maintenance. The decision-making of prioritization and optimization for CB maintenance is an essential and challenging issue, and limitations exist, including incomplete considerations, subjective rules, and the use of equal weights of the parameters in the integration process. Therefore, this paper proposes a data-driven based method for creating a more reasonable maintenance priority queue for CBs. The control circuit time features are utilized in conjunction with impact indicators to comprehensively evaluate the maintenance requirements of the CBs in the system. Moreover, a hybrid algorithm that combines fuzzy C-means (FCM) clustering with a ranking support vector machine (SVM) is proposed to determine the recommended maintenance priority of CBs. The simulation results demonstrate that the proposed method exhibits better performance than the conventional method and develops rational and objective maintenance prioritization for CBs, especially in the absence of prior knowledge. The method can also deal with missing and invalid data, thereby providing a reliable foundation for CB operation and maintenance management.