Power transformer state evaluate method and apparatus based on data fusion

Abstract

The invention discloses a power transformer state evaluate method and apparatus based on data fusion. The method comprises the following steps: selecting a state evaluation variable of an operation state of a power transformer, and determining an initial value and a threshold; performing normalization processing on the state evaluation variable through combination with a relative deterioration degree formula so as to obtain a normalization parameter; based on a fuzzy theory, obtaining a state quantity membership grade of the normalization parameter at each power transformer state grade; through a hierarchy analytical method, obtaining a weight of the state evaluation variable in each sub evidence body; according to the state quantity membership grade and weight of the state evaluation variable, obtaining each power transformer state grade membership grade of each sub evidence body; and through a Dempster-Shafer evidence theory, fusing the different sub evidence bodies, determining that the operation state of the power transformer is a normal state, an attention state, an abnormal state or a severe state. According to the method, the real-time operation state of the power transformer can be more comprehensively analyzed, and the reliability of a determining result of the real-time operation state of the power transformer is effectively improved.

Description

technical field [0001] The invention relates to the technical field of electrical equipment, in particular to a method and device for evaluating the state of a power transformer based on data fusion. Background technique [0002] It is pointed out in related technologies that power transformers are the core equipment of the power transmission and transformation system, and their health status affects the safe and stable operation of the power system; scientific and effective status assessment is helpful for timely discovery of latent defects and accident prevention, and reasonable arrangement of maintenance plans , reduce the probability of equipment failure and maintenance costs, and ensure long-term reliable operation of the power system. [0003] In related technologies, the operating status of power transformers is calculated by many methods such as Bayesian network, fuzzy comprehensive evaluation, gray target theory, gray correlation analysis, extension theory, and evid...

AI Technical Summary

Problems solved by technology

However, various methods have their own deficiencies. The Bayesian network requires knowledge of the prior probability and the independence of related attributes, which may not be true in practical applications. Neural network state evaluation requires samples of sufficient scale and many parameters to be set, and the output The results are difficult to interpret; gray relational analysis is easily affected by human subjective factors, and the objectivity needs to be improved; due to the shortcomings of the method itself, the reliability of the state evaluation needs to be further improved; at the same time, although the online monitoring technology has been vigorously promoted and applied, the related The monitoring data cannot be deeply and effectively involved in the transformer status evaluation, and the real-time evaluation results are difficult to meet the current needs

Images