Health State Assessment Method for Complex Systems Based on Correlation Decoupled Belief Rule Base
ABSTRACT Belief rule base (BRB) has been widely used in the health state assessment of complex systems because of its excellent performance in complex system modeling. However, most existing health assessment models assume attribute independence, whereas in real‐world engineering applications, corre...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-06-01
|
| Series: | Energy Science & Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/ese3.70076 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | ABSTRACT Belief rule base (BRB) has been widely used in the health state assessment of complex systems because of its excellent performance in complex system modeling. However, most existing health assessment models assume attribute independence, whereas in real‐world engineering applications, correlations between attributes can degrade modeling performance and compromise robustness. To quickly and effectively address the impact of attribute correlation on the model, this paper proposes the BRB model, which is based on correlation decoupling (BRB‐CD). The method introduces a decoupling matrix calculation strategy based on correlation analysis. First, the decoupling matrix is constructed based on the attribute features and its parameters are optimized by minimizing the correlation coefficient between attributes. Second, the input data are transformed using the decoupling matrix, and the data after decoupling is used as the new input to reduce the effects of redundant information and perturbations. Finally, through the experimental validation of the health state assessment of the WD615 diesel engine, it is found that the BRB‐CD model can realize more accurate health state assessment, and the model has better robustness. |
|---|---|
| ISSN: | 2050-0505 |