Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model
IntroductionRisk coupling (RC) analysis of underground gas storage (UGS) leakage accident risks is critical to overall natural gas storage safety. Consequently, the interactions among diverse risk factors need attention.MethodsThis study proposes a novel methodology combining Dynamic Bayesian Networ...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-07-01
|
| Series: | Frontiers in Earth Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/feart.2025.1639790/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839622408883929088 |
|---|---|
| author | Yao Hu Zhilong Ding Liguang Qiao Feng Gu Mengqi Yang |
| author_facet | Yao Hu Zhilong Ding Liguang Qiao Feng Gu Mengqi Yang |
| author_sort | Yao Hu |
| collection | DOAJ |
| description | IntroductionRisk coupling (RC) analysis of underground gas storage (UGS) leakage accident risks is critical to overall natural gas storage safety. Consequently, the interactions among diverse risk factors need attention.MethodsThis study proposes a novel methodology combining Dynamic Bayesian Networks (DBNs) and the N-K model to analyze RC in UGS leakage accidents. First, the causes of leakage accidents are systematically investigated, and risk categories are identified. Second, the categories of coupled risk arising from equipment, human, geological, and management factors are identified. Third, a DBN model is constructed based on leakage risk analysis and the N-K model. Fourth, the setting variables for RC nodes in the proposed DBN are identified through computational results using the N-K. Additionally, the validation of the proposed model is proven utilizing a three-axiom-based technique.ResultsBy integrating the N-K model’s mutual information metric with DBN’s temporal modeling, the approach achieves a mean absolute error (MAE) of 0.032 in predicting coupling probabilities and enables risk reduction of up to 17.4% through targeted interventions, enhancing the accuracy and actionable insights for UGS safety management. In the short term, the coupling of human factors and management factors is the main factor leading to the leakage accidents occurrence, and with the development of time, the coupling of equipment factors, human factors, geological factors, and management factors coupling is the main factor leading to the leakage accidents occurrence.DiscussionThe developed DBN effectively characterizes the dynamic evolution of leakage risks and RC mechanisms in UGS facilities. Furthermore, sensitivity analysis is implemented using the proposed model to investigate the impact of failure probabilities of risk factors on predominant RC types in the short term, we can reduce the human factors and management factors coupling by strengthening personnel training and optimizing the management process and other measures; in the long term, we can reduce the risk of coupling by constructing the whole life cycle management of the equipment, geological dynamic monitoring, and other measures. |
| format | Article |
| id | doaj-art-c47eb546b42f4e909a7bbce6d32a99f1 |
| institution | Matheson Library |
| issn | 2296-6463 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Earth Science |
| spelling | doaj-art-c47eb546b42f4e909a7bbce6d32a99f12025-07-22T05:29:55ZengFrontiers Media S.A.Frontiers in Earth Science2296-64632025-07-011310.3389/feart.2025.16397901639790Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K modelYao Hu0Zhilong Ding1Liguang Qiao2Feng Gu3Mengqi Yang4College of Civil Aviation Safety Engineering, Civil Aviation Flight University of China, Guanghan, Sichuan, ChinaCollege of Civil Aviation Safety Engineering, Civil Aviation Flight University of China, Guanghan, Sichuan, ChinaCollege of Economics and Management, Civil Aviation Flight University of China, Guanghan, Sichuan, ChinaLogistics service company, Civil Aviation Flight University of China, Guanghan, Sichuan, ChinaBusiness School, China University of Political Science and Law, Beijing, ChinaIntroductionRisk coupling (RC) analysis of underground gas storage (UGS) leakage accident risks is critical to overall natural gas storage safety. Consequently, the interactions among diverse risk factors need attention.MethodsThis study proposes a novel methodology combining Dynamic Bayesian Networks (DBNs) and the N-K model to analyze RC in UGS leakage accidents. First, the causes of leakage accidents are systematically investigated, and risk categories are identified. Second, the categories of coupled risk arising from equipment, human, geological, and management factors are identified. Third, a DBN model is constructed based on leakage risk analysis and the N-K model. Fourth, the setting variables for RC nodes in the proposed DBN are identified through computational results using the N-K. Additionally, the validation of the proposed model is proven utilizing a three-axiom-based technique.ResultsBy integrating the N-K model’s mutual information metric with DBN’s temporal modeling, the approach achieves a mean absolute error (MAE) of 0.032 in predicting coupling probabilities and enables risk reduction of up to 17.4% through targeted interventions, enhancing the accuracy and actionable insights for UGS safety management. In the short term, the coupling of human factors and management factors is the main factor leading to the leakage accidents occurrence, and with the development of time, the coupling of equipment factors, human factors, geological factors, and management factors coupling is the main factor leading to the leakage accidents occurrence.DiscussionThe developed DBN effectively characterizes the dynamic evolution of leakage risks and RC mechanisms in UGS facilities. Furthermore, sensitivity analysis is implemented using the proposed model to investigate the impact of failure probabilities of risk factors on predominant RC types in the short term, we can reduce the human factors and management factors coupling by strengthening personnel training and optimizing the management process and other measures; in the long term, we can reduce the risk of coupling by constructing the whole life cycle management of the equipment, geological dynamic monitoring, and other measures.https://www.frontiersin.org/articles/10.3389/feart.2025.1639790/fullunderground gas storageleakage accidentN-K modelrisk couplingdynamic Bayesian network |
| spellingShingle | Yao Hu Zhilong Ding Liguang Qiao Feng Gu Mengqi Yang Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model Frontiers in Earth Science underground gas storage leakage accident N-K model risk coupling dynamic Bayesian network |
| title | Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model |
| title_full | Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model |
| title_fullStr | Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model |
| title_full_unstemmed | Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model |
| title_short | Risk coupling analysis of underground gas storage leakage accidents based on dynamic Bayesian network and N-K model |
| title_sort | risk coupling analysis of underground gas storage leakage accidents based on dynamic bayesian network and n k model |
| topic | underground gas storage leakage accident N-K model risk coupling dynamic Bayesian network |
| url | https://www.frontiersin.org/articles/10.3389/feart.2025.1639790/full |
| work_keys_str_mv | AT yaohu riskcouplinganalysisofundergroundgasstorageleakageaccidentsbasedondynamicbayesiannetworkandnkmodel AT zhilongding riskcouplinganalysisofundergroundgasstorageleakageaccidentsbasedondynamicbayesiannetworkandnkmodel AT liguangqiao riskcouplinganalysisofundergroundgasstorageleakageaccidentsbasedondynamicbayesiannetworkandnkmodel AT fenggu riskcouplinganalysisofundergroundgasstorageleakageaccidentsbasedondynamicbayesiannetworkandnkmodel AT mengqiyang riskcouplinganalysisofundergroundgasstorageleakageaccidentsbasedondynamicbayesiannetworkandnkmodel |