A modified Bayesian network to handle cyclic loops in root cause diagnosis of process faults in the chemical process industry

Kumari, Pallavi; Bhadriraju, Bhavana; Wang, Qingsheng; Kwon, Joseph

doi:10.1016/j.jprocont.2021.12.011

Cited by 36 publications

(14 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The diagnosis performance of the developed method and TE-based multiblock BN is compared for the diagnosis of a root cause, deviation in the valve opening of the separator pot (listed in Table ), in the TEP. To perform diagnosis using the final BNs obtained using DTE-based and TE-based methods, the modified Bayesian network (mBN) technique is utilized . Specifically, the mBN first identifies the weakest causal relation of each cyclic loop and then converts it into a temporal relation.…”

Section: Resultsmentioning

confidence: 99%

“…These root causes are selected for this case study due to their ability to violate safety constraints in TEP, thus leading to rare events. Among these root causes, the faults due to root causes 6−8 are known to propagate via significant cyclic loops of the TEP, 20 and hence, they are also used to generate data for cyclic loop discovery while learning the causal network. Here, the selected root causes (Table 2) are inserted into the TEP after 8 h of normal operation, resulting in deviations in various process variables.…”

Section: Multiblock Bnmentioning

confidence: 99%

“…Here, the accuracy of root cause diagnosis depends on the causal network. For an accurate causal network, it is important to consider cyclic loops into the causal network that are prevalent in the CPI due to intensive material and heat integration, feedback control, and coupling among process variables. , To this end, recently, several works have proposed methods to meticulously incorporate cyclic loops in the causal network. , It is to be noted that cyclic loops incorporated into the causal network through existing methods are known beforehand based on process knowledge. However, it is possible, especially in a complex chemical process, that cyclic loops are not known due to inadequate process knowledge.…”

Section: Introductionmentioning

confidence: 99%

“…17,18 To this end, recently, several works have proposed methods to meticulously incorporate cyclic loops in the causal network. 19,20 It is to be noted that cyclic loops incorporated into the causal network through existing methods are known beforehand based on process knowledge. However, it is possible, especially in a complex chemical process, that cyclic loops are not known due to inadequate process knowledge.…”

Section: ■ Introductionmentioning

confidence: 99%

See 3 more Smart Citations

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

Kumari

Wang

Khan

et al. 2022

Ind. Eng. Chem. Res.

Self Cite

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

Section: Resultsmentioning

confidence: 99%

Section: Multiblock Bnmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: ■ Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

Kumari

Wang

Khan

et al. 2022

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In Ref. [4], a modified Bayesian network was constructed to diagnose the root cause of faulty cases for chemical processes. More related results can be found in Refs.…”

Section: Introductionmentioning

confidence: 99%

Community detection based process decomposition and distributed monitoring for large‐scale processes

et al. 2022

View full text Add to dashboard Cite

Distributed architectures wherein multiple decision-making units are employed to coordinate their decision-making/actions based on real-time communication have become increasingly important for monitoring processes that have large scales and complex structures. Typically, the development of a distributed monitoring scheme involves two key steps, that is, the decomposition of the process into subsystems, and the design of local monitors based on the configured subsystem models. In this article, we propose a distributed process monitoring approach that tackles both steps for large-scale processes. A data-driven process decomposition approach is proposed by leveraging community structure detection to divide variables into subsystems optimally via finding a maximal value of the metric of modularity. A two-layer distributed monitoring scheme is developed where local monitors are designed based on the configured subsystems of variables using canonical correlation analysis. Innersubsystem interactions and inter-subsystem interactions are tackled by the two layers separately, such that the sensitivity of this monitoring scheme to certain types of faults is improved. We utilize a numerical example to illustrate the effectiveness and superiority of the proposed method. It is then applied to a simulated wastewater treatment process.

show abstract

An improved Harris Hawks optimization for Bayesian network structure learning via genetic operators

Liu,

Cai,

Shi

et al. 2023

Soft Comput

View full text Add to dashboard Cite

A modified Bayesian network to handle cyclic loops in root cause diagnosis of process faults in the chemical process industry

Cited by 36 publications

References 51 publications

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

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

Community detection based process decomposition and distributed monitoring for large‐scale processes

An improved Harris Hawks optimization for Bayesian network structure learning via genetic operators

Contact Info

Product

Resources

About