A Hybrid Knowledge-Based and Data-Driven Method for Aging-Dependent Reliability Evaluation of High-Voltage Circuit Breaker

High-voltage circuit breakers (HVCBs) are important components for reliable operation of power systems. Due to irreversible degradation, effective estimation of the aging-dependent failure rates of HVCBs is crucial for fit-for-purpose maintenance strategies. Many data-intensive models previously proposed for the aging-dependent failure rate evaluation cannot be extended to CBs without sufficient aging-related condition monitoring data. This article proposes a hybrid knowledge-based and data-driven approach for estimating the heterogeneous aging-dependent failure rate of CBs using a well-established Evaluation-Forecasting-Individualization framework. First, the proposed approach employs a hybrid knowledge-based and data-driven method to evaluate the degradation index (DI) that quantifies the aging degrees of individual CB. Second, a Block Hankel Tensor Autoregressive Integrated Moving Average (BHT-ARIMA) based multiple sequence prediction model is developed to improve the DI forecasting accuracy by combining hidden structure information among correlated CBs. Finally, a DI-based relative degradation function is formulated and integrated into a 2-parameter Weibull distribution to evaluate the relative deterioration level of in-group individual CBs towards the entire group. Evaluation results for a 550 kV CB dataset from an actual power company (Chongqing grid) are used to demonstrate the effectiveness and efficiency of the proposed approach.