Three-stage feature fusion rotary machinery fault diagnosis method based on multi-modal data

Abstract

The invention provides a three-stage feature fusion rotary machine fault diagnosis method based on multi-modal data. The method comprises the steps: obtaining the operation state parameter data of a machine, and obtaining the data of at least two modals; inputting the obtained modal data into a preset neural network model to obtain a final fault classification result; wherein the preset neural network model at least sequentially comprises three stages of first self-feature fusion, mutual feature fusion and second self-feature fusion; according to the method, a three-stage feature fusion method including a one-dimensional convolutional neural network and a two-dimensional convolutional neural network is adopted, multi-scale features are fused and fault diagnosis is carried out, the two-dimensional convolutional neural network extracts correlation between feature mapping maps, an attention mechanism can carry out different weight distribution on the feature maps, and the fault diagnosis accuracy is improved. Important information is highlighted, redundant information is reduced, and the fault diagnosis performance is greatly improved.

Description

technical field [0001] The present disclosure relates to the technical field of mechanical fault diagnosis, in particular to a three-stage feature fusion method for fault diagnosis of rotating machinery based on multimodal data. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] With the continuous development of intelligent manufacturing, industrial systems are becoming more and more complex and non-linear, and the losses caused by equipment damage are also increasing. Early fault detection can not only eliminate faults before they cause huge economic losses, but also avoid major safety accidents. However, due to the complexity and nonlinearity of industrial systems, it is difficult to build an accurate model. Due to the continuous development of information science and technology, industrial systems have produced a large amount of opera...

AI Technical Summary

Problems solved by technology

However, despite the success of these intelligent methods, there are still two disadvantages: (1) These intelligent fault diagnosis methods need to be combined with feature extraction methods, resulting in a strong dependence on feature selection

However, data-level fusion requires a highly consistent data structure, and in decision-level fusion, when each classifier has a large decision-making conflict, the final classification result will be very unstable

Images