Hierarchical hybrid distributed PCA for plant-wide monitoring of chemical processes

Cao, Yue; Yuan, Xiaofeng; Wang, Yalin; Gui, Weihua

doi:10.1016/j.conengprac.2021.104784

“…Since the system state, x, is unknown, the data-based model (11) is tractable. As an estimation, ŷ can be obtained through a nonlinear estimator o, such as a mapping generated by neural networks.…”

Section: Data-based State Estimationmentioning

confidence: 99%

“…Some common FD techniques utilize statistical information (e.g., correlation) between the input-output data by using linear models [5], such as principal component analysis [2,11], support vector machine (SVM) [12][13][14][15]. Moreover, linear dynamic models have been developed through subspace identification [16][17][18].…”

Section: Introductionmentioning

confidence: 99%

A Segmental Autoencoder-based Fault Detection for Nonlinear Dynamic Systems: An Interpretable Learning Framework

Chen¹,

Pan²,

Dogru³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

This paper presents a segmental autoencoder-based fault detection (FD) framework for nonlinear dynamic systems. The basic idea behind the proposed FD scheme is to identify a generalized kernel representation based on the representation knowledge learned from an autoencoder. By using the system data, several cascades, linking nonlinear operators, are employed to obtain a data-based model which describes the nonlinear dynamic behaviors. With the help of the segmental structure of an autoencoder, a residual generator is then constructed. Rigorous mathematical analysis and an application on a continuous stirred tank reactor demonstrate the effectiveness of the proposed FD method.

show abstract

“…Further, to demonstrate the effectiveness of the proposed methodology, it is compared with a method based on the state-of-the-art TE score, i.e., TE-based multiblock BN methodology. Note that although there exist various approaches to segment processes into blocks in other areas, − they cannot be directly applied to block formation in the proposed work due to not considering shared variables between blocks that are required for the purpose of fusing block-level BNs.…”

Section: Introductionmentioning

confidence: 99%

A Direct Transfer Entropy-Based Multiblock Bayesian Network for Root Cause Diagnosis of Process Faults

Kumari

¹

,

Wang

²

,

Khan

³

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

In chemical processes, Bayesian network (BN)-based approaches have been extensively applied for process fault diagnosis. Generally, BN is learned using score and search algorithms where search algorithms create candidate networks whose fitness to data is measured by scores. However, existing approaches cannot utilize cyclic loop knowledge while learning BN. Since cyclic loops are prevalent in chemical processes, their unaccountability results in inaccurate BN and reduces diagnosis accuracy. Therefore, for accurate diagnosis, we propose direct transfer entropy (DTE)-based multiblock BN to discover cyclic loops while learning BN. First, the process is segmented into multiple blocks. Next, block-level BNs are learned using DTE-based score and Greedy search. By eliminating the common source variable effect, DTE finds correct causality between process variables and obtains accurate block-level BNs, which are fused to discover significant cyclic loops that were otherwise unachievable when finding BN. The performance of the developed methodology is demonstrated through a benchmark process.

show abstract

“…Some common FD techniques utilize statistical information (e.g., correlation) between the input-output data by using linear models [5], such as principal component analysis [2,11], support vector machine (SVM) [12][13][14][15]. Moreover, linear dynamic models have been developed through subspace identification [16][17][18].…”

Section: Introductionmentioning

confidence: 99%

A Segmental Autoencoder-based Fault Detection for Nonlinear Dynamic Systems: An Interpretable Learning Framework

Chen

¹

,

Pan

²

,

Dogru

³

et al. 2022

2022 IEEE 11th Data Driven Control and Learning Systems Conference (DDCLS)

Self Cite

View full text Add to dashboard Cite

This paper presents a segmental autoencoder-based fault detection (FD) framework for nonlinear dynamic systems. The basic idea behind the proposed FD scheme is to identify a generalized kernel representation based on the representation knowledge learned from an autoencoder. By using the system data, several cascades, linking nonlinear operators, are employed to obtain a data-based model which describes the nonlinear dynamic behaviors. With the help of the segmental structure of an autoencoder, a residual generator is then constructed. Rigorous mathematical analysis and an application on a continuous stirred tank reactor demonstrate the effectiveness of the proposed FD method.

show abstract

Hierarchical hybrid distributed PCA for plant-wide monitoring of chemical processes

Cited by 39 publications

References 36 publications

A Segmental Autoencoder-based Fault Detection for Nonlinear Dynamic Systems: An Interpretable Learning Framework

A Segmental Autoencoder-based Fault Detection for Nonlinear Dynamic Systems: An Interpretable Learning Framework

A Direct Transfer Entropy-Based Multiblock Bayesian Network for Root Cause Diagnosis of Process Faults

A Segmental Autoencoder-based Fault Detection for Nonlinear Dynamic Systems: An Interpretable Learning Framework

Contact Info

Product

Resources

About