Rotating machine fault diagnosis method, system and equipment and storage medium

Abstract

The invention provides a rotating machine fault diagnosis method, system and device and a storage medium. The method comprises the following steps: extracting a rotating machine fault signal by adopting a lightweight convolutional neural network to obtain spatial feature information, and performing feature recognition on the spatial feature information; realizing fault diagnosis of the rotating machine according to the feature recognition result; building a novel lightweight convolution block, so that the parameters of the model can be effectively reduced; building a lightweight convolutional neural network fault diagnosis model, so that the problem of data compression is solved to a great extent. And meanwhile, the network structure is effectively improved, and the fault identification precision of the whole rotating machine is improved.

Description

technical field [0001] The present application relates to the field of fault diagnosis, identification and classification, and in particular to a method, system, device and storage medium for fault diagnosis of rotating machinery. Background technique [0002] With the development of industrial Internet of Things technology, the use of deep learning technology to realize mechanical fault diagnosis has been widely used. In addition to realizing the automation and accuracy of fault diagnosis, reducing the amount of parameters and storage space of the model has also been gradually paid attention to. Although traditional convolutional neural networks have good feature extraction effects in data processing, they cannot be applied to devices with limited hardware performance due to high computational complexity. In order to improve the above shortcomings, replacing traditional convolution operations with lightweight convolution operations can effectively improve accuracy and redu...

AI Technical Summary

Problems solved by technology

[0003] At present, convolutional neural network technology has been widely used in the field of mechanical fault diagnosis, but it is not enough to ensure the automation and accuracy of fault diagnosis while reducing the number of parameters and storage space of the model.

Images