基于加权朴素贝叶斯的GIS 隔离开关接触状态评估方法

For improving the reliability of contact condition assessment of GIS disconnector，in this paper a contact condition assessment method of disconnector based on weighted naive Bayes algorithm with consideration of its operation condition and environmental information is proposed. The ambient temperature，wind speed，sunlight intensity，current of conductive rod and temperature of enclosure are taken as the input characteristics of the state assessment model，the electromagnetic-thermal-flow multi physical field coupling model of GIS disconnector is setup，and the relationship between the characteristics is simulated and analyzed. In order to reduce the influence of feature correlation on the assessment results，a contact condition assessment model of GIS disconnector based on weighted naive Bayes is set up，and the particle swarm optimization algorithm is used to solve the optimal weight of features in the evaluation model. The assessment model is tested，and the effectiveness and correctness of the method propoed in this paper are proved by the result.