2018
DOI: 10.1016/j.chemolab.2018.07.011
|View full text |Cite
|
Sign up to set email alerts
|

Representation learning based adaptive multimode process monitoring

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
8
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
7

Relationship

0
7

Authors

Journals

citations
Cited by 15 publications
(8 citation statements)
references
References 27 publications
0
8
0
Order By: Relevance
“…Compared with SI [34], MSI provides two distinctions: a) MSI utilizes the total loss (9) of all modes to estimate the importance measure while SI uses the loss (7) of the current mode. Because (9) is the exact optimization objective, it is easier to calculate than (7); b) The initialization of MSI is random while the initial setting of SI is the optimal value of the last mode. Since the objective of SDiPCA-MSI is nonconvex and nonconcvae, random initialization is beneficial for seeking the appropriate model parameters.…”
Section: A Modified Simentioning
confidence: 99%
See 1 more Smart Citation
“…Compared with SI [34], MSI provides two distinctions: a) MSI utilizes the total loss (9) of all modes to estimate the importance measure while SI uses the loss (7) of the current mode. Because (9) is the exact optimization objective, it is easier to calculate than (7); b) The initialization of MSI is random while the initial setting of SI is the optimal value of the last mode. Since the objective of SDiPCA-MSI is nonconvex and nonconcvae, random initialization is beneficial for seeking the appropriate model parameters.…”
Section: A Modified Simentioning
confidence: 99%
“…Industrial systems generally operate under multiple modes due to changing of raw materials, market demands, etc [9]- [11]. Therefore, multimode process monitoring has undergone tremendous development recently [12]- [14], which can be divided into single-model schemes and multiple-model approaches [10], [15].…”
Section: Introductionmentioning
confidence: 99%
“…The authors of many works, for example [16,17], focus on working with high-level features, adapting the processing system to the specific analysis method they use. However, the General list of methods used in industrial CPS information security monitoring is quite large [8,9,[18][19][20][21][22]. Each method has its own advantages and disadvantages.…”
Section: Related Workmentioning
confidence: 99%
“…Deep learning with deep structure is a promising technology in abstract representation learning, which has been applied to industrial image classification and soft‐sensor‐based chemical product quality prediction . Several deep learning methods have also been applied in FDD of chemical processes, such as deep belief network (DBN), stacked auto encoder (SAE), and convolutional neural network (ConvNet) . A DBN‐based fault diagnosis model was constructed to classify 20 faults in Tennessee Eastman (TE) process .…”
Section: Introductionmentioning
confidence: 99%
“…SAE was applied in representation learning to detect faults in TE process . Zhang et al incorporated k ‐nearest neighbor (kNN) rule into sparse denoising auto encoder (SDAE) to perform nonlinear process monitoring. ConvNet has shown great successes in recognition and detection tasks .…”
Section: Introductionmentioning
confidence: 99%