.png){kind=icon}
返回
A Fault Diagnosability Evaluation Method for Dynamic Systems Without Distribution Knowledge
.png){kind=icon}
PRE
AI
DOI：10.1109/TCYB.2020.3027549.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
This article addresses the quantitative fault diagnosability evaluation for dynamic systems without distribution knowledge. The system dynamics in different cases are first characterized by mean vectors and covariance matrices. Then, fault detectability and isolability are defined based on the constructed characteristics. On this basis, the Mahalanobis distance (MD) is employed to propose a fault diagnosability analysis measure. Furthermore, model-based and data-driven algorithms for fault diagnosability evaluation are given according to the MD-based measure. In addition, the reliabilities of the evaluation results are considered by taking advantage of ambiguity sets of mean vectors and covariance matrices of the system dynamics. Finally, two examples are employed to verify the effectiveness of the proposed algorithms.
Keyword:
Covariance matrices
Dynamical systems
System dynamics
Fault detection
Reliability
Uncertainty
Noise measurement
Fault diagnosability
Kullback-Leibler divergence (KLD)
Mahalanobis distance (MD)
quantitative evaluation
期刊
IF:
10.5
论文数:
1.1W
被引数:
5.0W
机构
引用论文
BANDWIDTH SELECTION IN KERNEL DENSITY ESTIMATION: ORACLE INEQUALITIES AND ADAPTIVE MINIMAX OPTIMALITY核密度估计中的带宽选择: ORACLE不等式和自适应极大极小最优性.png)
ANNALS OF STATISTICS
OAAI
ANNALS OF STATISTICSIF3.7
OAAI分享收藏Data-driven realizations of kernel and image representations and their application to fault detection and control system designAUTOMATICA
AUTOMATICAIF5.9
PREAI
分享收藏
查看更多内容