Proceedings. 5th IEEE International Symposium on Intelligent Control 1990
DOI: 10.1109/isic.1990.128602
|View full text |Cite
|
Sign up to set email alerts
|

Dynamic fault detection and diagnosis using neural networks

Abstract: A neural network methodology for dynamic fault diagnosis has been proposed. Moving windows cut the dynamic data into overlapping pieces. Then the segmented data are presented to the networks for training and generalization purposes. Some unique features associated nith this methodology, namely the length of the moving window, the sampling rate, and the construction of the training data set, have been studied. The proposed method was successfully applied to a binary distillation process and showed superiority o… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
9
0

Publication Types

Select...
3
3
1

Relationship

0
7

Authors

Journals

citations
Cited by 15 publications
(9 citation statements)
references
References 4 publications
0
9
0
Order By: Relevance
“…In mid 1980's, researchers in the area of process safety proposed a potential approach to detect and diagnose fault using neural network. Venkatasubramanian and Chan [5] were among the earliest researchers to apply ANN in PFD&D. Since then many researchers have studied the implementation of neural network in this area such as Li et al [6], Becraft and Lee [7] and Leung and Ramagnoli [8]. ANN was also used to perform several functions that might also be carried by integrating with other fault detection techniques such as knowledge based expert system [9].…”
Section: Artificial Neural Networkmentioning
confidence: 99%
See 2 more Smart Citations
“…In mid 1980's, researchers in the area of process safety proposed a potential approach to detect and diagnose fault using neural network. Venkatasubramanian and Chan [5] were among the earliest researchers to apply ANN in PFD&D. Since then many researchers have studied the implementation of neural network in this area such as Li et al [6], Becraft and Lee [7] and Leung and Ramagnoli [8]. ANN was also used to perform several functions that might also be carried by integrating with other fault detection techniques such as knowledge based expert system [9].…”
Section: Artificial Neural Networkmentioning
confidence: 99%
“…From heat exchanger (E100) to VLV 102 before pump around back to column. 6 From heat exchanger (E100) to VLV 100 before discharge to storage tank. 7…”
Section: Plant Modelmentioning
confidence: 99%
See 1 more Smart Citation
“…A moving window is an indispensable technique to track dynamic data and widely used for on-line fault diagnosis [5], [6], [10]. Usually it is important to select the proper time span of the moving window for fault diagnosis, which is the product of the sampling 'number (window length) in each moving window and time increment (sampling rate).…”
Section: Structure Of the Variable Moving Window Based Fault Diagnosimentioning
confidence: 99%
“…Large window sizes can consider more information, but may lead to large time delays for the classification of various process operating conditions. In most cases [5], [6],[10], after the time span of the moving is selected, it may not be changed any more. The length of the moving window can be changed with time, if HMM is used for classification.…”
Section: Structure Of the Variable Moving Window Based Fault Diagnosimentioning
confidence: 99%