Optimal Maintenance Scheduling of Power Substation Equipment: A Condition-Informed Risk Assessment Approach

An effective maintenance scheduling of power substation equipment helps to reduce maintenance cost, mitigate outage impacts, and enhance cost efficiency of the power grid. While existing risk-based maintenance strategies primarily rely on statistical failure data collected from similar equipment populations, they often overlook individual health condition data specific to individual equipment. This paper introduces a novel condition-informed risk assessment approach using Quantile Mapping (QM), which effectively integrates trending information from statistical failure data and individual health information from condition monitoring data. The risk assessment results inform optimal distribution level power equipment maintenance scheduling to maximize cost-effectiveness. Initially, an equipment failure model is constructed based on statistical failure data. Subsequently, utilizing the QM method, condition data is incorporated into the failure model, facilitating more accurate condition-informed risk assessment. Finally, risk assessment outcomes and maintenance expenses are integrated into a Mixed Integer Linear Programming (MILP) framework to derive optimal maintenance decisions. Case studies based on distribution power grids show a remarkable 25% reduction in total maintenance cost, and a 24.4% improvement in cost-effectiveness compared to existing time-based maintenance strategies, verifying the effectiveness and significance of the proposed method.