Insulator fault diagnosis apparatus and method based on fuzzy cerebellar model neural network

Abstract

The invention provides an insulator fault diagnosis apparatus and method based on a fuzzy cerebellar model neural network. The insulator fault diagnosis apparatus is composed of an MCU, a first pre-processing circuit, a second pre-processing circuit, a third pre-processing circuit, an electric field sensor, a leakage current sensor, and a temperature sensor, a remote transceiver module and a remote computer. The electric field sensor, the leakage current sensor and the temperature sensor collect signals; denoising processing is carried out based on a Kalman filter algorithm to obtain a featuresample with fault information; an insulator fault information training sample library is established; classification training is carried out on samples based on FCMAC; training is carried out by using a BP algorithm to obtain a weight value and a threshold value that enable network optimization to be realized; and when a new information sample is inputted into the network, a fault type of an insulator is determined rapidly and accurately. On the basis of combination of the Kalman algorithm and the CMAC, a few of weight values need to be corrected each time, the learning is accelerated, and the certain generalization ability is enhanced. Besides, the efficiency and accuracy of the insulator fault analysis are improved.

Description

technical field [0001] The invention belongs to the field of insulator detection, and in particular relates to an insulator fault diagnosis device and method based on a fuzzy cerebellum model neural network. Background technique [0002] Traditional insulator detection methods include infrared temperature measurement, ultraviolet imaging, ultrasonic detection, and leakage current method, but these detection methods have disadvantages such as high cost, low safety, or poor applicability. With the continuous development of computer technology and sensor technology, machine learning has been successfully applied to fault detection in many industries. There have been studies on the fault diagnosis of insulators, but many studies only identify a specific fault. Although the identification results are highly accurate, the identification ability is limited and there are still certain defects. Contents of the invention [0003] The purpose of the present invention is to provide a...

AI Technical Summary

Problems solved by technology

There have been studies on the fault diagnosis of insulators, but many studies only identify a specific fault. Although the

Images