Power equipment defect diagnosis method based on multi-source feature information fusion

Abstract

The invention discloses a power equipment defect diagnosis method based on multi-source feature information fusion. The method comprises the steps of acquiring a first image, a second image and sound data of a to-be-detected area in the same detection range; based on a wavelet transform method, according to the first image and the second image, generating a fused image, and according to the sound data, labeling the fused image to generate a sample data set; generating an initial defect diagnosis model based on a BP neural network, and training the initial defect diagnosis model according to the sample data set to determine model parameters; and generating a power equipment defect diagnosis model according to the model parameters and the initial defect diagnosis model, the power equipment defect diagnosis model being used for performing fault identification according to the first image, the second image and the sound data of the to-be-detected power equipment so as to generate a fault prediction box. According to the technical scheme, fault detection can be effectively carried out on the power equipment, and the operation stability of the power grid is improved.

Description

technical field [0001] This application relates to the technical field of electric equipment detection, in particular, to a method for diagnosing electric equipment defects based on multi-source feature information fusion. Background technique [0002] The domestic demand for power energy is getting higher and higher, and the maintenance of power equipment is a problem that cannot be ignored. In the prior art of electric equipment defect detection, there are generally three detection means, including: infrared thermal image detection, visible light image detection, and sound data detection. [0003] 1) The infrared thermal image can provide complete target temperature information, but the background information is blurred, and the specific part of the equipment overheating cannot be accurately identified; [0004] 2) Visible light images can provide more comprehensive background information, but the target defect information is not obvious; [0005] 3) The sound data can p...

AI Technical Summary

Problems solved by technology

[0003] 1) The infrared thermal image can provide complete target temperature information, but the background information is blurred, and the specific part of the equipment overheating cannot be accurately identified;

[0004] 2) Visible light images can provide more comprehensive background information, but the target defect information is not obvious;

[0005] 3) The sound data can provide sound parameter information such as frequency, and other parameter information such as position information corresponding to the image taken at the same angle of view. Combining defect types, although a suitable solution can be found in the terminal database, the sound data alone cannot accurately obtain information such as the defect type of power equipment, and there is a problem that the obtained defect detection information of power equipment is incomplete

[0006] In the prior art, infrared thermal images and visible light images are usually fused to detect electrical equipment defects in the electrical environment. However, the frequency and other parameter information provided by the sound data is not used in it, that is, the defect type is obtained. At the same time, it is impossible to find a solution in real time according to the degree of defect of the power equipment, which is not conducive to accurate maintenance of defective equipment

Images