Abstract

Aiming at the problems of data redundancy and data abnormality of multi-source unstructured data such as video, picture, and text in the process of processing quality inspection and equipment status monitoring of discrete intelligent production line, a multi-source unstructured data cleaning method based on dynamic cloud Bayesian network is proposed. We analyze the characteristics of multi-source unstructured data in the processing operation of the discrete intelligent production line and construct a multi-source unstructured data description model. combine dynamic Bayesian network and cloud model theory to design a multi-source unstructured data cleaning framework and processing flow based on dynamic cloud Bayesian network. finally, the feasibility of the proposed method is demonstrated by simulation analysis of arithmetic cases.

This content is only available via PDF.
You do not currently have access to this content.