Mechanical fault migration diagnosis method based on adaption sharing deep residual network (ASResNet)

Abstract

The invention discloses a mechanical fault migration diagnosis method based on an adaption sharing deep residual network (ASResNet). A labeled monitoring data set of a laboratory device and a monitoring data set of an engineering device are firstly acquired, and stacked residual units are used to extract migration fault characteristics of the monitoring data of the laboratory device and the engineering device; through a fully connected network, the mapping relationship between the migration fault characteristics and sample health labels is built, distribution differences between the migrationfault characteristics are calculated, the probability distribution of the sample labels is predicted, and pseudo labels of the monitoring data samples of the engineering device are generated; the monitoring data sets of the laboratory device and the engineering device are then used, a to-be-trained parameter set of the ASResNet is trained through object functions constructed by maximizing and minimizing, and a migration diagnosis model is acquired; and the monitoring data of the engineering device are inputted to realize mechanical fault migration diagnosis. The differences between the monitoring data of the laboratory device and the monitoring data of the engineering device are narrowed, and ideal effects are achieved for fault diagnosis for the engineering device.

Description

technical field [0001] The invention belongs to the technical field of mechanical fault diagnosis, and in particular relates to a mechanical fault migration diagnosis method based on an adaptive shared deep residual network. Background technique [0002] Large-scale mechanical equipment is gradually developing in the direction of precision, efficiency, and automation. Since large-scale mechanical equipment is often composed of many key parts, and the coordination between the parts, once a part fails, it will cause The failure of the entire mechanical system will affect the manufacturing process in the slightest, and endanger the lives of relevant personnel in severe cases. Therefore, it is necessary to study the fault diagnosis technology of mechanical equipment to detect and deal with faults early to ensure the safe operation of the equipment. The traditional fault diagnosis method relies too much on the experience and professional knowledge of diagnostic experts, and it is...

AI Technical Summary

Problems solved by technology

[0002] Large-scale mechanical equipment is gradually developing in the direction of precision, efficiency, and automation. Since large-scale mechanical equipment is often composed of many key parts, and the coordination between the parts, once a part fails, it will cause The failure of the entire mechanical system will affect the manufacturing process in the slightest, and endanger the lives of relevant personnel in severe cases. Therefore, it is necessary to study the fault diagnosis technology of mechanical equipment to detect and deal with faults early to ensure the safe operation of the equipment.

The traditional fault diagnosis method relies too much on the experience and professional knowledge of diagnostic experts, and it is difficult to meet the analysis requirements of massive monitoring data. However, intelligent fault diagnosis extracts fault feature information from monitoring data and uses intelligent algorithms to identify and classify faults. Automatic judgment and intelligent decision-making, get rid of the situation that traditional fault diagnosis methods rely too much on diagnostic experts, and become an important means to ensure the safe operation of mechanical equipment under the background of mechanical big data

[0003] At present, the relevant research on fault intelligent diagnosis generally requires that the available samples for training the intelligent diagnosis model are sufficient, but this is difficult to meet in engineering practice

On the one hand, mechanical equipment works under normal conditions for a long time, resulting in much more monitoring data obtained under normal conditions than under fault conditions. Therefore, the typical fault samples obtained are insufficient; The monitoring data obtained in the dataset does not contain labeling information, and manual labeling of monitoring data is expensive, resulting in insufficient samples of labeled monitoring data

In summary, it is difficult to train and obtain an intelligent fault diagnosis model with high accuracy in identifying the health status of equipment by only using the monitoring data of mechanical equipment obtained in engineering practice

Images