Name
Title | ||
|---|---|---|
A Novel Distributed Data-Driven Strategy for Fault Detection of Multi-Source Dynamic Systems |
Abstract | ||
|---|---|---|
In this brief, a distributed data-driven fault detection method based on sparse integrated dynamic principal component analysis is designed to enhance the fault detection performance of multi-source dynamic systems. First, by dividing the complex system into multiple sub-sources according to the structural principle, the fusion and complementation of process mechanism knowledge and data-driven are realized. Second, by introducing the Kronecker product covariance structure, the individual modes within each sub-source as well as joint modes shared between sub-sources are captured. Notably, by developing a truncated contribution power method, key variables are selected and the interpretability of principal components is improved. Finally, experimental results demonstrate that the proposed strategy can significantly improve the detection performance compared with the existing PCA-based methods. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/TCSII.2022.3181009 | IEEE Transactions on Circuits and Systems II: Express Briefs |
Keywords | DocType | Volume |
Fault detection,distributed,multi-source systems,sparse integrated dynamic principal component analysis,truncated contribution power method | Journal | 69 |
Issue | ISSN | Citations |
11 | 1549-7747 | 0 |
PageRank | References | Authors |
0.34 | 14 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yi-Fang Zhang | 1 | 0 | 0.34 |
| Bing Han | 2 | 108 | 13.00 |
| Min Han | 3 | 761 | 68.01 |