2017
DOI: 10.1109/tie.2017.2698422
|View full text |Cite
|
Sign up to set email alerts
|

Data-Driven Distributed Local Fault Detection for Large-Scale Processes Based on the GA-Regularized Canonical Correlation Analysis

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
56
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
6
1

Relationship

1
6

Authors

Journals

citations
Cited by 104 publications
(56 citation statements)
references
References 37 publications
0
56
0
Order By: Relevance
“…From (10), it is noticed that the reformulated optimization problem without penalties will reduce to that of KPCA. As can be seen in (8) and 10, there is no approximation and the excessive computation of K 1 2 is avoided. The features χ cannot be expressed with an explicit function.…”
Section: A Optimization Problem Based On Elastic Net Regularizationmentioning
confidence: 99%
See 1 more Smart Citation
“…From (10), it is noticed that the reformulated optimization problem without penalties will reduce to that of KPCA. As can be seen in (8) and 10, there is no approximation and the excessive computation of K 1 2 is avoided. The features χ cannot be expressed with an explicit function.…”
Section: A Optimization Problem Based On Elastic Net Regularizationmentioning
confidence: 99%
“…Due to the rapid development in the fields of storage and sensor technologies, data-driven PM-FD methods have gained a substantial amount of research attention in recent years [3]- [5]. As the representative data-driven PM-FD approaches, principal component analysis (PCA), canonical correlative analysis (CCA), partial least squares (PLS), have been widely applied in various industrial processes [6]- [8]. For example, Zhou et al [9] developed a The associate editor coordinating the review of this manuscript and approving it for publication was Francesco Tedesco.…”
Section: Introductionmentioning
confidence: 99%
“…In modern industrial processes, it is important to guarantee process safety, achieve energy conservation, and improve product quality, all of which often largely depend on the timely monitoring and control of important quality variables . However, due to reasons like harsh measurement environment, cost of expensive instruments, and significant measurement delays, most of these quality variables are often difficult to measure online by hard sensors in actual production processes.…”
Section: Introductionmentioning
confidence: 99%
“…23 In recent years, many monitoring algorithms based on data-driven subspace segmentation have been published. [24][25][26][27] These methods can aid in decreasing process complexity and focus on local data characteristics to reduce the risk of variability being overwhelmed under abnormal conditions and improve monitoring performance. On the basis of the PCA, an approach for monitoring nonlinear processes with multiple linear subspaces is established.…”
Section: Introductionmentioning
confidence: 99%