Research on Integrated SVDD Rotating Machinery Vibration Fault Detection Method Based on Deep Autoencoder

There are some problems in vibration equipment of rotating machinery, such as complicated mechanism, confused field data and fewer fault samples. In order to improve the accuracy of fault detection, this paper proposes a rotating mechanical vibration fault detection method based on deep autoencoder and integrated SVDD(DAE-ISVDD). First, use the original normal sample data to train multiple different deep autoencoders, after that the trained autoencoders are used to extract features from the original data, then use the extracted features to train multiple basic SVDD weak classifiers, and finally, a strong classifier is obtained by means of weighted integration. Through the example verification, the resulting strong classifier can effectively improve the accuracy of the vibration fault detection of rotating machinery. Meanwhile, compared with other traditional methods, this method has good anti-noise performance and generalization ability, and can be well applied in practical engineering.