An Online Data Driven Fault Detection Method in Dynamic Process based on Sparse Representation

作者：Yang Rui; Huang Mengjie^*

来源：IEEE International Conference on Mechatronics and Automation (ICMA), 2017-08-06 To 2017-08-09.

DOI：10.1109/ICMA.2017.8016093

• intermittent fault
• fault detection
• sparse representation

With the development of science and industrial technologies, the intermittent fault has become the main fault of actual system, and the fault diagnosis on intermittent fault has progressed. However, with the increase in the complexity and uncertainty of modern engineering system, it is not feasible to establish accurate mathematical models. Thus, data-driven method is required for fault detection. Based on the sparsity of intermittent faults in some domains, an intermittent fault detection method based on sparse representation is proposed, with the online update of over-complete dictionary and fault detection threshold. With the simulation verification, the proposed method is suitable for intermittent fault detection in dynamic system.

• 出版日期2017
• 单位山东科技大学; 中国石油大学（北京）