State Evaluation and Tripping Probability Prediction of Large Power Transformer Based on Health Index

Abstract

Abstract Aiming at the problem of overmaintenance and undermaintenance caused by lack of equipment running state, a method of state evaluation and tripping probability prediction of large power transformer based on health index is proposed. The paper constructs a transformer health evaluation system with a four layer deep architecture, uses the extension cloud theory to evaluate the deterioration of state indicators, combines the Analytic Hierarchy Process and Entropy Weight Method to weight the indicators in indicator level, introduces the improved Dezert-Smarandache (DSmT) theory to effectively integrate the evaluation results of each layer, and reconciles the contradictions and conflicts between conclusions. Using the health index to represent the health state of transformers, constructing a tripping probability prediction model with the health index as input, and obtaining the tripping probability of transformers. The research results indicate that the method proposed in this paper can accurately and effectively evaluate the health state of transformers and their functional components, providing information for the management and maintenance strategy of equipment.