Oil-paper insulation aging diagnosis method based on insulating oil Raman spectrum wavelet packet energy entropy

Abstract

The invention, which belongs to the field of electrical equipment safety assessment, particular relates to an oil-paper insulation aging diagnosis method based on an insulating oil Raman spectrum wavelet packet energy entropy. According to the method, an insulating oil Raman spectrum wavelet packet energy entropy is used as an aging characteristic quantity; an aging diagnosis model that is obtained by combining a Kennard-Stone algorithm, multi-class support vector machine construction, optimization based on a genetic algorithm is used to realize an overall aging state diagnosis of field oil-paper insulating equipment rapidly and effectively. Therefore, a quantitative analysis of dissolve aging characteristic concentration in a certain kind of oil is not needed and the fault-tolerant performance is good. The method has the good growth. An oil-paper insulating sample obtained in the field operating transformer can be used for expanding a sample base of the diagnosis model in work in future, so that the diagnosis model can be improved and corrected continuously and thus the operation and maintenance level of the oil-paper insulation equipment can be improved.

Description

technical field [0001] The invention belongs to the field of electric equipment safety state evaluation, and relates to a transformer aging diagnosis method, in particular to an oil-paper insulation aging diagnosis method based on the energy entropy of insulating oil Raman spectrum wavelet packets. Background technique [0002] On-site rapid diagnosis of the aging state of oil-paper insulation equipment is the key to improving equipment operation safety and preventing major accidents, and is of great significance to ensuring power system operation safety. At present, the existing methods for diagnosing the aging state of oil-paper insulation mainly evaluate the changes in physical and chemical parameters and electrical performance parameters caused by insulation aging. The degree of polymerization and tensile strength among the physical and chemical parameters can best reflect the aging strength of oil-paper insulation. However, when measuring, it is necessary to suspend the...

AI Technical Summary

Problems solved by technology

The degree of polymerization and tensile strength among the physical and chemical parameters can best reflect the aging strength of oil-paper insulation. However, when measuring, it is necessary to suspend the power of the transformer and take samples from several most representative parts of the winding. damage to insulation

The electrical performance parameters mainly refer to the threshold values ​​of single characteristic parameters such as gas analysis in insulating oil and the content of furfural, methanol and acetone. The detection effect of the spectroscopic method is unstable and is easily interfered by organic matter in the oil. The colorimetric method is not only unsatisfactory, but also the toluidine used in the detection is considered to be one of the strongest carcinogens; while CO and CO in the oil 2 The gas chromatography method usually used for content detection, the performance of the chromatographic column will gradually decrease with the increase of the use time, and the gas chromatography detection system requires high-level testing personnel to operate in the laboratory, which limits its application in on-site detection

At present, there is no effective on-site diagnosis method for the overall aging condition of the insulation. Therefore, it is not only of great significance to study a method that can quickly and effectively diagnose the aging degree of oil-paper insulation equipment, but also has a huge application prospect.

Images