Industrial equipment fault detection and classification method based on tensor data dimension reduction

Abstract

The invention provides an industrial equipment fault detection and classification method based on tensor data dimension reduction in the field of Internet-of-things. The method comprises the followingsteps: 1) data acquisition: various types of sensors are adopted for data acquisition, a data source is provided for prediction, and the data comprises structured data such as operation parameters inthe production process of industrial equipment and unstructured data such as videos or images during operation; 2) data preprocessing: fusing the data of different structures, and performing dimension reduction on the fused data; and 3) data analysis: performing health detection on the production condition of the industrial equipment according to the data sent by the sensors around the productionprocess after a large amount of data is trained by using the stacked denoising auto-encoder of the server, so that the processing efficiency is improved, and the accuracy of the detection result is improved.

Description

technical field [0001] The invention relates to a fault detection method, in particular to a method for detecting and classifying industrial equipment faults based on dimension reduction of tensorized data, which belongs to the technical field of the Internet of Things. Background technique [0002] The Industrial Internet of Things and data-driven technologies are revolutionizing manufacturing by enabling computer networks to collect vast amounts of data from connected machines and transform large-scale mechanical data into actionable information. As a key component of modern manufacturing systems, machine health monitoring has fully embraced the big data revolution. In contrast to the top-down modeling provided by traditional physics-based models, data-driven machine health monitoring systems provide a bottom-up solution paradigm for when certain failures occur (diagnostics) and predict Future operating conditions and remaining service life. To extract useful knowledge f...

AI Technical Summary

Problems solved by technology

[0008] The disadvantage of this solution is that it only collects continuous or discrete structured data as the basis for evaluating the health status of equipment, and the reliability is not high.

Due to the singleness of the data, the prediction model in the above scheme only uses the linear regression algorithm to predict the target parameter data, and the prediction method is too simple

In terms of evaluating the health status, by comparing the predicted value with the actual value, according to whether the residual error of the two exceeds a certain threshold, only the health of the industrial equipment can be obtained, and the health of the industrial equipment cannot be accurately judged, that is, it cannot Detailed judgment of the health status of industrial equipment

Images