Research on Data Association Method Based on Directed Graph

Nowadays, with the development of information technology and continuous improvement of industrial automation, the big data industry continue to penetrate all aspects of manufacturing industry, the data in industry contains many kinds of data, complex structure and wide range of sources. How to develop and make use of it has become a great challenge for enterprises.At the same time, the data pile up makes the processing cycle longer, and the complexity of the structure makes the processing difficulty and calculation increase.However, these data are discrete and can not find the correlated relationship between them. Leads to failure to retrieve their hidden information and to search for potential behind faults or threats. It can not provide effective decision support for fault diagnosis and value discovery.Therefore, this paper aims at the data obtained by sensors and monitoring systems in industry. A data association method based on directed graph is proposed, and the calculation of relational strength and temporal proximity is added. Finally, experiments show the validity and accuracy of the proposed method.