The modern equipment is developing in the direction of complexity and integration, and the traditional diagnosis methods can’t adapt to the changes of equipment. With the development of deep learning, deep learning algorithms are gradually used in the field of fault diagnosis. Most of these algorithms ignore the correlation between data when extracting features from multi-source data. In response to the above problem, this paper proposes a fault diagnosis algorithm based on Bi-directional Long Short-Term Memory, which can extract the correlation features between multi-source data. Through the experiment on the simulated operating data of the heating furnace of steel rolling, the algorithm of this paper has a high fault recognition rate.
Research on Fault Diagnosis Algorithm Based on Bi-directional Long Short-Term Memory
Lect. Notes Electrical Eng.
International Workshop of Advanced Manufacturing and Automation ; 2020 ; Zhanjiang, China October 12, 2020 - October 13, 2020
2021-01-23
10 pages
Article/Chapter (Book)
Electronic Resource
English