Isolating Multiple Sources of Plant-Wide Oscillations via Independent Component Analysis

Xia, Chunming; Howell, John

doi:10.1016/s1474-6670(17)36590-4

Cited by 9 publications

(16 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. Spectral decomposition methods can also be used to detect and classify spectral features in multivariate datasets. A spectral envelope method was used to detect and categorize process measurements with similar spectral characteristics …”

Section: Introductionmentioning

confidence: 99%

Methods for root cause diagnosis of plant‐wide oscillations

Duan¹,

Chen²,

Shah³

et al. 2014

AIChE Journal

View full text Add to dashboard Cite

Plant-wide oscillations are common in many industrial processes. They may impact the overall process performance and reduce profitability. It is important to detect and diagnose such oscillations. This paper reviews advances in diagnosis of plant-wide oscillations. The main focus of this study is on identifying possible root causes of oscillations using two techniques, one based on data analysis in the temporal and spectral domains and the other based on process connectivity analysis. The process data-based analysis provides an effective way to capture the difference between the root cause variable and the secondary propagated oscillating variables. It is shown that process topology-based methods are capable of finding oscillation propagation pathways and, thus, help in determining the root cause. This paper discusses and compares five such methods-spectral envelope, adjacency matrix, Granger causality, transfer entropy, and Bayesian network inference methods-by application to an industrial benchmark dataset.

show abstract

Section: Introductionmentioning

confidence: 99%

Methods for root cause diagnosis of plant‐wide oscillations

Duan¹,

Chen²,

Shah³

et al. 2014

AIChE Journal

View full text Add to dashboard Cite

show abstract

“…Independent Component Analysis (ICA) is a decomposition of a data matrix that minimises statistical dependence between the basis vectors. It gives basis functions with a good one-to-one relationship with the physical sources of signals, as shown by Xia and Howell [12] who gave the first application of ICA to process spectra. Non-negative matrix factorization (NMF) was introduced in the area of image recognition [17].…”

Section: Detection Of Non-oscillating Disturbancesmentioning

confidence: 96%

Advances and New Directions in Plant-Wide Controller Performance Assessment

Thornhill

Horch²

2006

IFAC Proceedings Volumes

View full text Add to dashboard Cite

“…Spectral independent component analysis (ICA) and non‐negative matrix factorization (NMF) are two decomposition methods used for plantwide oscillation diagnosis. For details, refer to Refs 13–15.…”

Section: Tools For Plantwide Oscillations Detectionmentioning

confidence: 99%

Plantwide oscillations diagnosis—current state and future directions

Choudhury

2011

Asia-Pacific J Chem Eng

View full text Add to dashboard Cite

This paper describes the currently available techniques for root‐cause diagnosis of plantwide oscillation and future directions for research in this area. Plantwide oscillation arises mainly for three reasons: (1) tight controller tuning, (2) loop interactions and (3) presence of nonlinear faults such as valve stiction and sensor faults. Plantwide oscillation is detrimental for plant operation because it causes large variability in the control loops resulting in reduced profitability due to inferior quality products, larger rejection rates, reduction in average throughput and increase in energy requirements. Copyright © 2011 Curtin University of Technology and John Wiley & Sons, Ltd.

show abstract

Isolating Multiple Sources of Plant-Wide Oscillations via Independent Component Analysis

Cited by 9 publications

References 9 publications

Methods for root cause diagnosis of plant‐wide oscillations

Methods for root cause diagnosis of plant‐wide oscillations

Advances and New Directions in Plant-Wide Controller Performance Assessment

Plantwide oscillations diagnosis—current state and future directions

Contact Info

Product

Resources

About