Oil-Paper Insulation State Evaluation by Integrating Neighborhood Rough Sets and Evidential Reasoning

In order to accurately diagnose the oil-paper insulation state, this paper proposes a comprehensive assessment model that fuses neighborhood rough set and evidence-based reasoning. The model is constructed based on the reply voltage method multi-feature quantity assessment database, and the evidence fusion theory is used to calculate the confidence level of the insulation state proposition, which overcomes the one-sidedness of the single feature quantity. Meanwhile, in order to reduce the influence of subjective factors, the basic confidence assignment method based on the database is proposed according to the definition of the upper and lower approximations of the neighborhood rough set, and the rationality of the weight assignment is enhanced. The results show that the proposed strategy can reflect the deterioration trend and provide a reference basis for the formulation of overhaul strategy.