Research of condition based maintenance evaluation model for the on-post vacuum circuit breaker on least squares support vector machine based on particle swarm optimization algorithm

-With the connections between the safe operation and national economy development getting closer, the original scheduled maintenance has been unable to adapt to the current demand for electricity. The condition-based maintenance, which is an immediate and effective maintenance mode for supply equipment, achieves supply companies general concern. This paper aims at the existing problems in on-post vacuum circuit breaker, builds the device status and risk assessment index system based on actual situation, and proposes on-post vacuum circuit breaker condition-based maintenance overhaul which based on particle swarm optimization and least squares support vector machine. This paper collects 100 box transformer substation data from distribution network in a power company to do empirical analysis, the mean absolute percentage error and mean square error is 0.1296% and 0.0716, respectively. Thus, this method has high accuracy and good generalization ability, which can be applied to the evaluation of condition-based maintenance. Key-Words: -Condition-based maintenance; On-post vacuum circuit breaker; Particle swarm optimization; Least squares support vector machine; Comprehensive evaluation