Transformer state fuzzy set pair assessment method based on matter-element augmentation extensive correlation

Abstract

The invention discloses a transformer state fuzzy set pair evaluation method based on matter-element augmentation extensive correlation, comprising steps of performing classification on a transformer fault sample set according to a set fault type, performing equivalent augmentation on the classified fault sample sets through a random weighting method, constructing a fault type-fault symptom correlation identification matrix based on augmented fault sample sets to obtain a reduced correlation frequency fault set matrix of each fault type and an optimized fault symptom constant weight coefficient, constructing a transformer fault symptom set data dictionary classical domain and a joint domain through combination with a matter-element extensive theory so as to obtain a matter-element correlation function of each fault symptom and solve a variable weight coefficient corresponding to each fault type, calculating an identical discrepancy contrary evaluation matrix needed by the transformer state evaluation through the ladder-ridge type fuzzy membership function, and obtaining a set pair analysis multi-element relation degree of each fault type and the transformer integral operation state by combining with the optimized variable weight coefficient to realize the analysis and evaluation on the transformer operation state.

Description

technical field [0001] The invention relates to the technical field of transformer state evaluation, in particular to a transformer state fuzzy set pair evaluation method based on matter-element augmented extension association. Background technique [0002] The working status of power transformers, especially oil-immersed power transformers, is an important reference for evaluating the safety and reliability of their operation. Power transformer failures can cause severe power system outages. Therefore, it is particularly important to detect potential initial faults and evaluate the working status of power transformers, which is an important technical means to prevent and avoid serious faults of power transformers. [0003] Among the existing transformer fault detection and condition assessment methods, the dissolved gas analysis (DGA) in oil is commonly used. The working mechanism of the DGA method is that the manifestations of power transformer faults mainly include over...

AI Technical Summary

Problems solved by technology

[0006] However, when the set pair analysis state evaluation method is used to calculate the weight, the expert comprehensive scoring method and the analytic hierarchy process with strong subjective dependence are usually used, so it needs the support of subjective experience and has poor robustness; while the existing fusion association rules The intelligent algorithm only uses simplified support and confidence criteria in the process of association rule mining and weight calculation, especially in the calculation of weight coefficients based on confidence, although it improves the subjectivity of expert comprehensive scoring and AHP, it does not It does not effectively take into account the internal correlation between fault symptoms and the complexity of the rules, nor does it consider the influence of sample balance on the correctness of transformer state feature correlation extraction, and its similarity, difference and reverse evaluation degradation analysis blurs the negative Influenced by boundaries, its triangular membership function is relatively rough when simulating the ambiguity between state levels; while the extension analysis based on matter-element correlation introduces the idea of ​​matter-element correlation in the degradation analysis, effectively considering the state parameters of the transformer in the normal state. However, the evaluation process still needs the accumulation and support of expert systems and empirical knowledge, and the robustness of its state evaluation needs to be improved

Images