Rotating machinery fault qualitative diagnosis method based on convolutional neural network

Abstract

The invention discloses a rotating machinery fault qualitative diagnosis method based on a convolutional neural network, belonging to the technical field of machinery fault diagnosis. The diagnosis method comprises the steps of extracting enough sample points from rotating machinery vibration data, building a convolutional neural network model, training the convolutional neural network model, randomly intercepting multiple sample points in a test, classifying the test sample points by using the trained convolutional neural network, and completing the qualitative diagnosis of a rotating machinery fault. The artificial extraction of features is needed in a traditional fault qualitative diagnosis method, the accuracy is low, the generalization performance is poor, the method is complex and difficult to understand, and the engineering popularization of the method is difficult. According to the rotating machinery fault qualitative diagnosis method based on the convolutional neural network, the features can be automatically extracted, the accuracy is high, the generalization performance is high, the method is simple and is easy to understand, and the engineering popularization of the method is easy.

Description

Technical field: [0001] The invention belongs to the technical field of mechanical fault diagnosis, and in particular relates to a qualitative diagnosis method for rotating mechanical faults based on a convolutional neural network. technical background: [0002] Rotating machinery is one of the most widely used mechanical parts in the industry, and it is also a vulnerable part. Its operating status directly affects the performance of the entire equipment. Rotating machinery mainly includes bearings and gears. According to incomplete statistics, 30% of mechanical failures are caused by rotating machinery. When a rotating machine fails, it is of great practical significance to detect the type of failure. [0003] The traditional qualitative diagnosis method for rotating machinery faults relies on manual feature extraction, the algorithm is complex, and the effect is not good. Therefore, the present invention proposes a qualitative diagnosis method for rotating machinery faul...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a qualitative diagnosis method for rotating machinery faults based on convolutional neural network, so as to solve the technical problems that the traditional rotating machinery fault detection method has poor diagnostic effect and is difficult to be practically applied.

Images