83 / 2024-04-11 11:13:02

Machine Learning-Based Predictive Model for Steady-State Temperature at Critical Points of a 126 kV Vacuum Circuit Breaker

Vacuum circuit breaker, machine learning, temperature estimation

The structural complexities of vacuum circuit breakers (VCBs) impede the temperature measurement at critical points. This study develops a novel predictive model for a 126 kV VCB, leveraging load current, ambient temperature, and contact resistance at key positions to estimate temperatures at these critical points. The research has two primary objectives: first, to estimate temperatures at measurement points during temperature rise tests to assess compliance with standard limits; and second, to calculate the hot spot temperature within the vacuum interrupter. This paper presents an electromagnetic-thermal-fluid coupled simulation model of a single-phase 126 kV VCB to investigate the temperature distribution under steady-state load conditions. Utilizing the data collected from the simulation, a predictive model was developed using machine learning to estimate temperatures at critical points within the VCB. This research uses external parameters to calculate the internal temperatures of the VCB, thus evaluating whether the temperature rise meets design and operational standards.