Electrified equipment fault diagnosis method based on neural network model

Abstract

The invention discloses an electrified equipment fault diagnosis method based on a neural network model. The electrified equipment fault diagnosis method comprises the following main steps: S1, acquiring an infrared image of a measured object; S2, performing image processing, and establishing an image model library; S3, associating object names of the detected objects, and extracting detection features; S4, setting a threshold upper limit, and formulating a diagnosis rule; S5, constructing a data set of the defect sample image, constructing a convolutional neural network model, and training the convolutional neural network model; and S6, acquiring an infrared image of the measured object online, and realizing online automatic diagnosis through the convolutional neural network model, thereby identifying and diagnosing faults of the electrified equipment. According to the method, the trained convolutional neural network model is applied to identification of the defect image of the electrified equipment, so that accurate identification of the fault of the electrified equipment is realized, the operation is simple and convenient, the data is standard and unified, the working difficultyof electric power inspection can be reduced, the inspection efficiency is improved, and the fault detection rate is increased.

Description

technical field [0001] The invention relates to the technical field of infrared diagnosis, in particular to a neural network model-based fault diagnosis method for live equipment. Background technique [0002] Infrared detection is a conventional and effective detection method in live detection of power grid power equipment. It has the advantages of long-distance, non-contact and real-time imaging. In the whole detection process, infrared thermal imaging technology is used to detect power equipment. Infrared thermal images can detect possible thermal defects in equipment, predictively detect faults in electrical equipment, and avoid serious damage to electrical equipment. At present, infrared detection is widely used in power transmission and distribution systems of various voltage levels. [0003] Although power workers can judge infrared thermal defects of traditional power equipment based on the "Application Specification for Infrared Diagnosis of Live Equipment", the st...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the complex and cumbersome application specifications existing in the infrared diagnosis method of charged equipment in the prior art, the non-standardized infrared shooting process, the ineffective use of a large amount of infrared thermal image data, and the differences in the load, environment, materials, etc. of the same equipment. , the temperature data obtained by infrared detection has deviations, which will directly affect the results of infrared diagnosis, the fault detection rate is low, and the technical problems cannot be warned in time. A neural network model-based fault diagnosis method for live equipment is provided, which is easy to operate and has standardized data. Unification can reduce the difficulty of power inspection work, improve inspection efficiency, and improve fault detection rate

Images