Power transformer fault diagnosis method and diagnosis device

Abstract

The invention discloses a power transformer fault diagnosis method and diagnosis device. The method comprises the following steps that the fault type is acquired and the dissolved gas sample in oil is clear so as to establish a fault standard sequence; a sample matrix including the sample to be tested is established by integrating the fault standard sequence and an improved three-ratio method; the kernel function type is determined and the corresponding parameters of a kernel fuzzy clustering algorithm are initialized; a membership matrix is solved through the kernel fuzzy clustering algorithm by iteration; and the membership value of the sample to be tested is obtained according to the membership matrix, and the final fault type of the sample to be tested is determined through comparison of the membership value. According to the diagnosis method, the accuracy and the reliability of sample division clusters can be enhanced so that the range of application is wider, and the accuracy of internal potential fault diagnosis of the power transformer can be effectively enhanced.

Description

technical field [0001] The invention relates to the technical field of power transformers, in particular to a fault diagnosis method and a diagnosis device for power transformers. Background technique [0002] As the key equipment of the power grid, the power transformer connects transmission lines of different voltage levels and undertakes the functions of voltage conversion and power distribution. Due to the complex internal structure of the power transformer, the uneven distribution of the electric field, and the stress of electricity, heat, machinery, chemistry, and the environment during operation, various defects and failures inevitably occur. When a power transformer fails and causes an unplanned power outage, it will not only bring huge economic losses to the power production unit, but also seriously disrupt people's daily life and normal social order. [0003] DGA (Dissolved Gas Analysis, dissolved gas analysis in oil) method is a simple and effective transformer f...

AI Technical Summary

Problems solved by technology

However, because the fuzzy clustering method makes hyperspherical assumptions on the clusters, it is only suitable for convex clustering analysis, resulting in further improvement in fault diagnosis accuracy; in addition, the IEC three-ratio method cannot judge multiple fault types, and cannot be used in actual use. Identify all failure types

Images