Ai-enhanced reliability centered maintenance for high voltage gas circuit breakers

• Proposes an AI-enhanced reliability-centered maintenance (RCM) framework for high-voltage circuit breakers (HVCBs). • Introduces a three-stage condition assessment method that incrementally improves accuracy using progressively enriched data. • Validates the framework through real-world network data, benchmarking it against conventional CBM strategies. • Demonstrates that AI-integrated RCM improves maintenance planning, reduces operational costs, and enhances reliability. Circuit breakers (CBs) are critical for protection and switching in power systems, requiring effective maintenance strategies to ensure reliability and longevity. Traditional approaches, including time-based, preventive, and condition-based maintenance (CBM), can be costly and inefficient. Reliability-centered maintenance (RCM) offers a more strategic alternative by prioritizing maintenance based on equipment condition and system importance. This paper presents a novel AI-driven method for assessing the condition of high-voltage circuit breakers (HVCBs) and optimizing maintenance schedules, with a particular focus on contact erosion. The proposed framework comprises three stages, each incorporating progressively more data to enhance assessment accuracy. The methodology is validated in a real-world network, comparing CBM and RCM strategies. Results indicate that integrating AI with RCM improves maintenance planning, reduces costs, and enhances system reliability. This approach provides a cost-effective and data-driven solution for CB maintenance in power systems.