Research On Health Assessment Method Of Power Transformer Based On PCA And Analytic Hierarchy Process

With the continuous expansion of the size of the power grid,the capacity and number of power transformers in high-voltage transmission networks and low-voltage distribution networks have grown rapidly.As one of the core equipments of the power system,power transformers play an important role in the power grid.Their health status is very important for the formulation of later maintenance plans and life prediction.In order to ensure the stability and reliability of the power system,reduce unnecessary repairs and unplanned repairs,improve the operating efficiency of the power grid and the life of power transformers,and accurately assess the health status of power transformers has become the focus of current electrical equipment research.Therefore,this paper selects power transformers as the Research object.First,build an evaluation system.A scientific,comprehensive and reasonable evaluation system is the basis for accurate assessment of health status.According to the structure and actual situation of power transformers,the health assessment of power transformers is divided into quantitative index assessment and qualitative index assessment.At the same time,reference to relevant data and literature to determine specific indicators for quantitative assessment and qualitative assessment and quantify.Among them,quantitative index evaluation is the core of power transformer health assessment,so it is the focus of this paper.Secondly,it analyzes the quantitative index evaluation methods for the health status of power transformers.Considering the uncertainty in the quantitative evaluation index data collection and the ambiguity of the index health status classification,the PCA-based power transformer health state cloud model evaluation method and the hierarchical analysis-based power transformer health state fuzzy evaluation method were used for Describe the health status of power transformers and compare the advantages and disadvantages of the two assessment methods.Finally,a fusion of evaluation methods based on evidence theory.Each evaluation method has certain limitations.For the same data,there will be differences in the evaluation results.Therefore,the focus of evidence theory is used to highlight the parts that are supported by different evaluation methods and weaken the differences.At the same time,the evaluation weights are optimized using the least square method,and the optimized evaluation results are merged to further improve the reliability of the evaluation,the rationality of the maintenance strategy and the accuracy of the life prediction.The evaluation model proposed in this paper is verified by examples,and the effectiveness and practicability of the evaluation model are verified by comparing different methods.