Connection type intermittent fault diagnosis method based on self-organizing feature mapping neural network

Abstract

The invention discloses a connection type intermittent fault diagnosis method based on a self-organizing feature mapping neural network. The method comprises the following steps: step 1, extracting total element feature parameters; wherein the total element characteristic parameters comprise test signal characteristic parameters and external environment characteristic parameters; step 2, model learning training; wherein the learning training process comprises the steps of initializing, normalizing, calculating an Euclidean distance, determining a minimum distance, adjusting a connection weightand updating a learning rate and a neighborhood; step 3, fault diagnosis; learning and training samples in a normal state, an intermittent state and a fault state to obtain neural network weight vectors of various classification results; calculating the severity degree of the state of the standard sample, and marking the state of the classified result; and inputting the test sample into the model, and obtaining the state of the test sample through the category diagnosed by the SOFM neural network, thereby obtaining an intermittent fault diagnosis result of the test sample.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis, in particular to a connection-type intermittent fault diagnosis method based on a self-organizing feature mapping neural network. Background technique [0002] Intermittent faults are generally subdivided into three types, engineering, test void and connection. In the late stage of equipment life (the late stage of delivery), connection type intermittent faults begin to emerge in large numbers. The main reason is that during use, the external temperature and vibration stress of the device become more and more affected over time, which makes the connection part of the device more prone to poor contact or cracked solder joints. After the intermittent fault occurs in the use of the equipment, it may not be reproduced during ground inspection. There are two main reasons. The first is that such intermittent faults usually require a certain stress condition to be detected; the second is that ev...

AI Technical Summary

Problems solved by technology

[0003] 1. The Hidden Markov (HMM) model is mainly based on statistics and probability transfer matrix. It is a statistical description of the occurrence of random processes, which cannot reflect the characteristics of intermittent faults and has no adaptability;

[0004] 2. The occurrence of intermittent faults has an obvious relationship with the external environment. The Hidden Markov (HMM) model cannot reflect the external environment, such as stress conditions, temperature, etc., so the model cannot be widely used

Images