Multi-fault Signal Feature Extraction of Mud Pump Based on Parallel Factorization
- DOI
- 10.2991/mcei-18.2018.15How to use a DOI?
- Keywords
- Parallel factor; Fault diagnosis; Feature extraction; Multiple faults; Nonlinear
- Abstract
The multi-source dynamic feature recognition of mechanical nonlinear multi-failure mode is a technical bottleneck and problem encountered in the application of fault diagnosis in the process industrial production line. It needs not only to extract the time-frequency characteristics of the single-source fault signal but also to ensure the corresponding relationship between the nonlinear variable, multi-fault mode and multi-source fault features in time, frequency and space after feature extraction. Mechanical multi-source signals also have coupling, aliasing, and other phenomena that cause the characteristic signals between the channels to interfere with each other and overlap. In order to meet the requirements of automatic monitoring and fault diagnosis of industrial process production lines, this paper develops multi-source dynamic feature recognition and adaptive diagnosis. This paper studies the multi-scale parallel factorization theory and proposes a three-dimensional time-frequency space model reconstruction algorithm for multi-source feature factors, which improves the accuracy of mechanical fault detection and intelligent levels
- Copyright
- © 2018, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Liu Yang AU - Hanxin Chen AU - Wenjian Huang AU - Jinmin Huang AU - Chenghao Cao PY - 2018/06 DA - 2018/06 TI - Multi-fault Signal Feature Extraction of Mud Pump Based on Parallel Factorization BT - Proceedings of the 2018 8th International Conference on Mechatronics, Computer and Education Informationization (MCEI 2018) PB - Atlantis Press SP - 73 EP - 78 SN - 2352-538X UR - https://doi.org/10.2991/mcei-18.2018.15 DO - 10.2991/mcei-18.2018.15 ID - Yang2018/06 ER -