Electromechanical product system or equipment real-time fault diagnosis method based on width learning

Abstract

The invention provides an electromechanical product system or equipment real-time fault diagnosis method based on width learning, which provides a plurality of methods including loss function optimization, namely cost-sensitive learning, random inactivation Dropout and integrated learning on the basis of a traditional width learning system to obtain an improved width learning system. Aiming at the problems of different data types, serious imbalance of the types and the like of actual monitoring data, on the premise of guaranteeing the training and optimization efficiency, the cost weight, the inactivation probability and the like are set as adjustable parameters, integrated learning voting is carried out by imitating a bagging algorithm, and a final result is predicted. The ubiquitous problems of uncertain influence and class imbalance in fault diagnosis are solved, based on the improved width learning system, training is fast, prediction is accurate, stability and robustness are high, the method is applied to real-time monitoring of the health state of a complex system or equipment, faults can be prevented, and maintenance suggestions can be provided.

Description

technical field [0001] This application relates to the field of fault diagnosis, and specifically relates to a fault diagnosis method for fault diagnosis of electromechanical product systems or equipment using a width learning system to realize real-time and accurate monitoring of fault states. Background technique [0002] With the rapid development of information technology and artificial intelligence level, while the degree of integration and intelligence of mechanical and electrical products continues to increase, its complexity and risk factor have also greatly increased. During the operation of electromechanical products, the components are closely connected and easily cross-affected. A small fault is likely to cause a chain reaction, resulting in damage to the entire system. Not only will it bring huge economic losses, but it may also endanger the lives of relevant personnel. Therefore, condition monitoring and fault diagnosis of electromechanical product systems or ...

AI Technical Summary

Problems solved by technology

In order to solve the problem that the model structure and parameters are difficult to determine, and reduce the influence of uncertain factors, the present invention adds a Dropout layer to the BLS and adopts integrated learning to realize the health of the electromechanical product system or equipment on the basis of meeting the diagnostic accuracy requirements. Real-time monitoring of status

Images