Reliability Engineering & System Safety最新文献_第7页

Nested adaptive Kriging-based reliability analysis and computational resource allocation for complex systems 基于嵌套自适应kriging的复杂系统可靠性分析与计算资源分配

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-02-05 DOI: 10.1016/j.ress.2026.112367

Kewei Ye , Xiaobing Ma , Han Wang

Efficient reliability analysis of complex engineering systems faces significant challenges due to the integration of multiple subproblems, multidisciplinary coupling, high-dimensional characteristics, and resource incompatibility. These systems are often decomposed into multiple cascading subsystems, which enables concurrent analysis to manage this complexity. Surrogate-based techniques are widely utilized to alleviate the computational burden associated with time-consuming simulations. This study proposes a nested adaptive Kriging-based method for the reliability analysis of complex systems by integrating system decomposition with adaptive surrogate-based methods. The proposed method operates within a multilayer framework and proceeds in two stages, namely, a sequential updating stage and a resource allocation stage. In the first stage, an efficient nested reliability-oriented acquisition function is developed to guide model updating, and its closed-form expression is derived. In the second stage, a cost-effectiveness strategy that accounts for both simulation costs and modeling costs is introduced to determine which model combinations should be updated at each iteration. Finally, the proposed method is validated to be superior to the benchmark method and strategy through two mathematical examples and two practical applications.

复杂工程系统的多子问题集成、多学科耦合、高维特性和资源不兼容等特点，使高效的可靠性分析面临重大挑战。这些系统通常被分解为多个级联子系统，这使得并发分析能够管理这种复杂性。基于代理的技术被广泛用于减轻与耗时模拟相关的计算负担。将基于自适应代理的方法与系统分解相结合，提出了一种基于嵌套自适应kriging的复杂系统可靠性分析方法。该方法在多层框架内运行，分顺序更新阶段和资源分配阶段两个阶段进行。首先，建立了一个高效的面向可靠性的嵌套获取函数来指导模型更新，并推导了该函数的封闭表达式；在第二阶段，引入考虑模拟成本和建模成本的成本效益策略，以确定在每次迭代中应该更新哪些模型组合。最后，通过两个数学算例和两个实际应用，验证了所提方法优于基准方法和策略。

{"title":"Nested adaptive Kriging-based reliability analysis and computational resource allocation for complex systems","authors":"Kewei Ye , Xiaobing Ma , Han Wang","doi":"10.1016/j.ress.2026.112367","DOIUrl":"10.1016/j.ress.2026.112367","url":null,"abstract":"<div><div>Efficient reliability analysis of complex engineering systems faces significant challenges due to the integration of multiple subproblems, multidisciplinary coupling, high-dimensional characteristics, and resource incompatibility. These systems are often decomposed into multiple cascading subsystems, which enables concurrent analysis to manage this complexity. Surrogate-based techniques are widely utilized to alleviate the computational burden associated with time-consuming simulations. This study proposes a nested adaptive Kriging-based method for the reliability analysis of complex systems by integrating system decomposition with adaptive surrogate-based methods. The proposed method operates within a multilayer framework and proceeds in two stages, namely, a sequential updating stage and a resource allocation stage. In the first stage, an efficient nested reliability-oriented acquisition function is developed to guide model updating, and its closed-form expression is derived. In the second stage, a cost-effectiveness strategy that accounts for both simulation costs and modeling costs is introduced to determine which model combinations should be updated at each iteration. Finally, the proposed method is validated to be superior to the benchmark method and strategy through two mathematical examples and two practical applications.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112367"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal Bayesian maintenance policy for gear shafts under variable operating conditions with partially observable information 具有部分可观测信息的变工况下齿轮轴的最优贝叶斯维修策略

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-02-09 DOI: 10.1016/j.ress.2026.112388

Xin Li , Jing Cai , Weixi Shi , Zhenzhen Liu , Zhendong Zhao , Yan Liu , Hang Fei , Hongfu Zuo

In mechanical transmission systems, gear shafts serve as essential conduits for torque transfer and alignment processes, and their failure can lead to substantial increases in maintenance and supportability costs. Sensor-based condition monitoring yields only partially observable information about the actual health of gear shafts, which complicates maintenance decisions. To overcome this challenge, a novel optimal Bayesian maintenance policy under partially observable information is presented. A hidden semi-Markov model (HSMM) consisting of three states—unobservable healthy and unhealthy states, as well as an observable failure state—is employed to model the performance degradation process of the target system. Considering the nondecreasing characteristics of the system hazard rate in normal operations and wear-out stages in practical scenarios, the Erlang and hyper-Erlang distributions are employed to depict the sojourn times in the healthy and unhealthy states, respectively. An explicit conditional reliability function is updated in real time on the basis of Bayes’ theorem and integrated into a cost-minimizing semi-Markov decision process (SMDP). A control limit algorithm identifies the reliability threshold for optimal downtime scheduling. A validation of the gear shaft life test conducted under variable operating conditions reveals the earlier detection of incipient failures and lower expected average costs than those of other fault detection models. The proposed approach offers both theoretical insights and practical value for enhancing the safety and reliability of high-end mechanical equipment.

在机械传动系统中，齿轮轴是扭矩传递和对准过程的重要管道，它们的故障会导致维护和支持成本的大幅增加。基于传感器的状态监测只能获得有关齿轮轴实际健康状况的部分可观察信息，这使得维护决策变得复杂。为了克服这一挑战，提出了一种新的部分可观察信息下的最优贝叶斯维护策略。采用隐半马尔可夫模型（HSMM）对目标系统的性能退化过程进行建模，该模型由不可观察的健康状态和不健康状态以及可观察的失效状态三种状态组成。考虑到实际场景中系统在正常运行和磨损阶段的危害率不降低的特点，分别采用Erlang和超Erlang分布来描述健康和非健康状态下的停留时间。根据贝叶斯定理实时更新显式条件可靠性函数，并将其集成到成本最小化的半马尔可夫决策过程中。控制极限算法确定了最优停机调度的可靠性阈值。在不同运行条件下进行的齿轮轴寿命测试验证表明，与其他故障检测模型相比，该模型可以更早地检测到早期故障，并且预期平均成本更低。该方法对提高高端机械设备的安全性和可靠性具有重要的理论意义和实用价值。

{"title":"Optimal Bayesian maintenance policy for gear shafts under variable operating conditions with partially observable information","authors":"Xin Li , Jing Cai , Weixi Shi , Zhenzhen Liu , Zhendong Zhao , Yan Liu , Hang Fei , Hongfu Zuo","doi":"10.1016/j.ress.2026.112388","DOIUrl":"10.1016/j.ress.2026.112388","url":null,"abstract":"<div><div>In mechanical transmission systems, gear shafts serve as essential conduits for torque transfer and alignment processes, and their failure can lead to substantial increases in maintenance and supportability costs. Sensor-based condition monitoring yields only partially observable information about the actual health of gear shafts, which complicates maintenance decisions. To overcome this challenge, a novel optimal Bayesian maintenance policy under partially observable information is presented. A hidden semi-Markov model (HSMM) consisting of three states—unobservable healthy and unhealthy states, as well as an observable failure state—is employed to model the performance degradation process of the target system. Considering the nondecreasing characteristics of the system hazard rate in normal operations and wear-out stages in practical scenarios, the Erlang and hyper-Erlang distributions are employed to depict the sojourn times in the healthy and unhealthy states, respectively. An explicit conditional reliability function is updated in real time on the basis of Bayes’ theorem and integrated into a cost-minimizing semi-Markov decision process (SMDP). A control limit algorithm identifies the reliability threshold for optimal downtime scheduling. A validation of the gear shaft life test conducted under variable operating conditions reveals the earlier detection of incipient failures and lower expected average costs than those of other fault detection models. The proposed approach offers both theoretical insights and practical value for enhancing the safety and reliability of high-end mechanical equipment.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112388"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A safety protection method based on trajectory prediction for the operation of virtual coupling trains 基于轨迹预测的虚拟联轴车运行安全保护方法

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-01-30 DOI: 10.1016/j.ress.2026.112328

Ying Zhao , Haijun Li , Xiaobing Liu , Yan Huang

This study proposes a safety protection method based on trajectory prediction (SPTP) for the operation of virtual coupling trains. Specifically, a hybrid TLMA model that integrates Temporal Convolutional Networks (TCN), Long Short-Term Memory (LSTM), and Multi-head Self-attention (MATT) was developed to predict the trajectory of the leading train. Based on the prediction results, the SPTP method was introduced, grounded in principles such as space requirements for the following train’s operation, safety requirements when trains are stationary in the station platform, and operation safety requirements under different adverse conditions. Furthermore, a microscopic multi-state train-following model was constructed to validate the effectiveness of the SPTP method. The comparative results of the prediction model demonstrate that the TLMA model outperforms baseline models, achieving high accuracy and demonstrating excellent applicability for train trajectory prediction. Then, the SPTP method was compared with existing safety protection methods. Numerical simulation results showed that the SPTP method effectively reduced the distance interval between trains by 34.6 %, the speed difference between trains by 7.0 %, and the arrival time deviation by 65.0 %. These findings suggest that the SPTP method could effectively improve operation efficiency for urban rail trains and enhance passenger service quality.

提出了一种基于轨迹预测的虚拟联轴车运行安全保护方法。具体而言，建立了一种结合时间卷积网络（TCN）、长短期记忆（LSTM）和多头自注意（MATT）的混合TLMA模型来预测前导列车的运行轨迹。根据预测结果，从后续列车运行的空间要求、列车在站台静止时的安全要求、不同不利条件下的运行安全要求等原则出发，引入了SPTP方法。在此基础上，构建了微观多状态列车跟踪模型，验证了SPTP方法的有效性。预测模型的对比结果表明，TLMA模型优于基线模型，具有较高的预测精度，对列车轨迹预测具有良好的适用性。然后，将SPTP方法与现有的安全保护方法进行了比较。数值模拟结果表明，SPTP方法可有效地将列车间距缩短34.6%，列车间速度差缩短7.0%，到达时间偏差缩短65.0%。研究结果表明，SPTP方法可以有效提高城市轨道交通运行效率，提高客运服务质量。

{"title":"A safety protection method based on trajectory prediction for the operation of virtual coupling trains","authors":"Ying Zhao , Haijun Li , Xiaobing Liu , Yan Huang","doi":"10.1016/j.ress.2026.112328","DOIUrl":"10.1016/j.ress.2026.112328","url":null,"abstract":"<div><div>This study proposes a safety protection method based on trajectory prediction (SPTP) for the operation of virtual coupling trains. Specifically, a hybrid TLMA model that integrates Temporal Convolutional Networks (TCN), Long Short-Term Memory (LSTM), and Multi-head Self-attention (MATT) was developed to predict the trajectory of the leading train. Based on the prediction results, the SPTP method was introduced, grounded in principles such as space requirements for the following train’s operation, safety requirements when trains are stationary in the station platform, and operation safety requirements under different adverse conditions. Furthermore, a microscopic multi-state train-following model was constructed to validate the effectiveness of the SPTP method. The comparative results of the prediction model demonstrate that the TLMA model outperforms baseline models, achieving high accuracy and demonstrating excellent applicability for train trajectory prediction. Then, the SPTP method was compared with existing safety protection methods. Numerical simulation results showed that the SPTP method effectively reduced the distance interval between trains by 34.6 %, the speed difference between trains by 7.0 %, and the arrival time deviation by 65.0 %. These findings suggest that the SPTP method could effectively improve operation efficiency for urban rail trains and enhance passenger service quality.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112328"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Zone-collaborative integrated framework for probabilistic flaw tolerance assessment of aeroengine structure 航空发动机结构概率缺陷容限评估的区域协同集成框架

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-01-31 DOI: 10.1016/j.ress.2026.112338

Jiong-ran Wen , Bai-yang Zheng , Jian Li , Cheng-wei Fei

Traditional flaw tolerance assessment methods for aeroengine turbine blisks suffer from incomplete spatial coverage and inadequate uncertainty quantification, as hotspot-based approaches focus solely on predetermined high-stress regions while neglecting stochastic flaw distributions across structural surfaces. This study develops a Zone-Collaborative Integrated (ZCI) framework that systematically addresses these limitations through three integrated components: zone-probabilistic decomposition using improved Gaussian Mixture Models with K-means initialization for surface partitioning and Combined Sampling Method for flaw coordinate uncertainty quantification; Genetic Algorithm-enhanced Kriging (GA-Kriging) surrogate modeling integrated with series system theory for multi-zone reliability assessment; and systematic implementation algorithm enabling comprehensive spatial coverage with computational efficiency. Validation through notched plate and turbine blisk case studies demonstrate that GA-Kriging achieves 63.3% improvement in computational efficiency and 31.8%/26.7% enhancement in training/testing precision compared to conventional methods, with normalized RMSE below 0.02. The ZCI framework exhibits 94.95-98.70% accuracy relative to direct simulation while predicting 12-76% higher fatigue life than hotspot method at equivalent reliability levels (720 cycles for hotspot vs. 1518 cycles for two-zone ZCI at R = 0.99 in Case 2). Sensitivity analysis reveals flaw geometry parameters dominate reliability outcomes (flaw radius: -1.75, flaw depth: -1.25), providing quantitative guidance for structural design optimization. The proposed framework transforms computationally prohibitive full-scale reliability problems into manageable zone-based assessments, offering a systematic approach for probabilistic flaw tolerance design of critical aerospace components.

传统的航空发动机涡轮叶片缺陷容差评估方法只关注预定的高应力区域，而忽略了结构表面上缺陷的随机分布，存在空间覆盖不全和不确定性量化不足的问题。本研究开发了一个区域协同集成（ZCI）框架，该框架通过三个集成组件系统地解决了这些限制：区域概率分解使用改进的高斯混合模型与K-means初始化进行表面划分和组合采样方法进行缺陷坐标不确定性量化；结合序列系统理论的遗传算法改进Kriging （GA-Kriging）代理模型多区可靠性评估系统的实现算法，以计算效率实现全面的空间覆盖。通过缺口板和涡轮叶片的案例验证，GA-Kriging方法的计算效率比传统方法提高了63.3%，训练/测试精度提高了31.8%/26.7%，归一化RMSE低于0.02。相对于直接模拟，ZCI框架显示出94.95-98.70%的准确性，而在相同的可靠性水平下，预测疲劳寿命比热点方法高12-76%（在案例2中，热点为720次循环，两区ZCI为1518次循环，R = 0.99）。灵敏度分析表明，缺陷几何参数主导了可靠性结果（缺陷半径为-1.75，缺陷深度为-1.25），为结构设计优化提供了定量指导。提出的框架将计算上禁止的全尺寸可靠性问题转化为可管理的基于区域的评估，为关键航空部件的概率缺陷容限设计提供了一种系统的方法。

{"title":"Zone-collaborative integrated framework for probabilistic flaw tolerance assessment of aeroengine structure","authors":"Jiong-ran Wen , Bai-yang Zheng , Jian Li , Cheng-wei Fei","doi":"10.1016/j.ress.2026.112338","DOIUrl":"10.1016/j.ress.2026.112338","url":null,"abstract":"<div><div>Traditional flaw tolerance assessment methods for aeroengine turbine blisks suffer from incomplete spatial coverage and inadequate uncertainty quantification, as hotspot-based approaches focus solely on predetermined high-stress regions while neglecting stochastic flaw distributions across structural surfaces. This study develops a Zone-Collaborative Integrated (ZCI) framework that systematically addresses these limitations through three integrated components: zone-probabilistic decomposition using improved Gaussian Mixture Models with K-means initialization for surface partitioning and Combined Sampling Method for flaw coordinate uncertainty quantification; Genetic Algorithm-enhanced Kriging (GA-Kriging) surrogate modeling integrated with series system theory for multi-zone reliability assessment; and systematic implementation algorithm enabling comprehensive spatial coverage with computational efficiency. Validation through notched plate and turbine blisk case studies demonstrate that GA-Kriging achieves 63.3% improvement in computational efficiency and 31.8%/26.7% enhancement in training/testing precision compared to conventional methods, with normalized RMSE below 0.02. The ZCI framework exhibits 94.95-98.70% accuracy relative to direct simulation while predicting 12-76% higher fatigue life than hotspot method at equivalent reliability levels (720 cycles for hotspot vs. 1518 cycles for two-zone ZCI at R = 0.99 in Case 2). Sensitivity analysis reveals flaw geometry parameters dominate reliability outcomes (flaw radius: -1.75, flaw depth: -1.25), providing quantitative guidance for structural design optimization. The proposed framework transforms computationally prohibitive full-scale reliability problems into manageable zone-based assessments, offering a systematic approach for probabilistic flaw tolerance design of critical aerospace components.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112338"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analysis of urban hydrogen-blended natural gas pipeline leak failure and accident evolution based on the combination of causal inference and probabilistic machine learning 基于因果推理和概率机器学习相结合的城市混氢天然气管道泄漏故障及事故演变分析

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-01-29 DOI: 10.1016/j.ress.2026.112323

Wuyin Lin , Songming Yu , Xinran Yu , Yuxing Li , Cuiwei Liu

Integrating hydrogen into urban gas pipeline networks is a pivotal technology for energy transition yet poses critical safety threats, thus necessitating comprehensive risk assessment of hydrogen-blended natural gas pipelines. This study performs full quantitative risk assessment of leakage failure and accident evolution by proposing a novel framework that integrates causal inference (Bow-Tie analysis) with probabilistic machine learning (Bayesian networks), enabling systematic failure factor identification and dynamic accident progression simulation. Key findings indicate human factors and pipeline material degradation as primary triggers. The studied pipeline exhibits a low baseline failure probability, with dispersion emerging as the most likely consequence of leakage. Higher hydrogen blending ratios significantly elevate jet fire risk due to hydrogen’s low ignition energy, while hydrogen’s inherent buoyancy and high diffusivity notably mitigate the likelihood of flash fire and vapor cloud explosion. The case study verifies the model’s practicability, and macro-micro analyses provide holistic insights, offering a reliable method to guide pipeline safety and reliability improvement amid energy transition.

氢气融入城市燃气管网是能源转型的关键技术，但也存在严重的安全威胁，因此有必要对氢气混合天然气管道进行综合风险评估。本研究通过提出一种将因果推理（Bow-Tie分析）与概率机器学习（贝叶斯网络）相结合的新框架，对泄漏故障和事故演变进行了全面的定量风险评估，从而实现了系统的故障因素识别和动态事故进展模拟。主要研究结果表明，人为因素和管道材料降解是主要诱因。所研究的管道显示出较低的基线失效概率，泄漏最可能的结果是分散。由于氢的点火能较低，较高的氢混合比例显著提高了喷射火灾的风险，而氢固有的浮力和高扩散系数显著降低了闪火和蒸汽云爆炸的可能性。通过实例分析，验证了模型的实用性，宏观微观分析提供了整体洞见，为指导能源转型背景下管道安全可靠性提升提供了可靠的方法。

{"title":"Analysis of urban hydrogen-blended natural gas pipeline leak failure and accident evolution based on the combination of causal inference and probabilistic machine learning","authors":"Wuyin Lin , Songming Yu , Xinran Yu , Yuxing Li , Cuiwei Liu","doi":"10.1016/j.ress.2026.112323","DOIUrl":"10.1016/j.ress.2026.112323","url":null,"abstract":"<div><div>Integrating hydrogen into urban gas pipeline networks is a pivotal technology for energy transition yet poses critical safety threats, thus necessitating comprehensive risk assessment of hydrogen-blended natural gas pipelines. This study performs full quantitative risk assessment of leakage failure and accident evolution by proposing a novel framework that integrates causal inference (Bow-Tie analysis) with probabilistic machine learning (Bayesian networks), enabling systematic failure factor identification and dynamic accident progression simulation. Key findings indicate human factors and pipeline material degradation as primary triggers. The studied pipeline exhibits a low baseline failure probability, with dispersion emerging as the most likely consequence of leakage. Higher hydrogen blending ratios significantly elevate jet fire risk due to hydrogen’s low ignition energy, while hydrogen’s inherent buoyancy and high diffusivity notably mitigate the likelihood of flash fire and vapor cloud explosion. The case study verifies the model’s practicability, and macro-micro analyses provide holistic insights, offering a reliable method to guide pipeline safety and reliability improvement amid energy transition.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112323"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146098522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-dimensional sequence embedding and improved Informer for prediction of industrial alarm events 面向工业报警事件预测的多维序列嵌入和改进的Informer

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-01-29 DOI: 10.1016/j.ress.2026.112317

Wenbin Jiang , Wenkai Hu , Yupeng Li , Weihua Cao

As an effective alarm monitoring strategy, alarm event prediction helps mitigate the impact of alarm floods and the risk of industrial accidents by providing early warnings of potential future alarms, thereby allowing operators more time to take corrective action. However, in continuous industrial processes, varying operating conditions and abnormal states cause real-time fluctuations in alarm rates, posing challenges for existing methods to achieve satisfactory prediction performance. In view of such issues, this paper proposes a new alarm event prediction method adapting to variable alarm rates over long-term consecutive alarm monitoring periods using multi-dimensional sequence embedding and improved Informer. The contributions are threefold: 1) An adaptive alarm sequence segmentation strategy is designed to generate input alarm sequences adapting to alarm rates; 2) a multi-dimensional sequence embedding method based on both the alarm tags and time intervals is proposed to convert the textual alarm messages into numerical vectors; and 3) an Informer based alarm event prediction model is developed for precise and early alarm event prediction under alarm flood and non-flood periods. A case study based on the Vinyl Acetate Monomer public model is given to prove the effectiveness of the proposed method.

作为一种有效的报警监测策略，报警事件预测通过提供潜在未来报警的早期预警，有助于减轻报警洪水的影响和工业事故的风险，从而使运营商有更多的时间采取纠正措施。然而，在连续的工业过程中，不同的运行条件和异常状态会导致报警率的实时波动，这对现有方法实现令人满意的预测性能提出了挑战。针对这些问题，本文提出了一种基于多维序列嵌入和改进的Informer的适应长期连续报警监测周期内变报警率的报警事件预测新方法。主要贡献有三：1)设计了一种自适应报警序列分割策略，生成适应报警率的输入报警序列；2)提出了一种基于报警标签和时间间隔的多维序列嵌入方法，将文本报警信息转化为数值向量；3)建立了基于Informer的预警事件预测模型，实现了预警洪涝期和非洪涝期预警事件的准确预警。以醋酸乙烯单体公共模型为例，验证了该方法的有效性。

{"title":"Multi-dimensional sequence embedding and improved Informer for prediction of industrial alarm events","authors":"Wenbin Jiang , Wenkai Hu , Yupeng Li , Weihua Cao","doi":"10.1016/j.ress.2026.112317","DOIUrl":"10.1016/j.ress.2026.112317","url":null,"abstract":"<div><div>As an effective alarm monitoring strategy, alarm event prediction helps mitigate the impact of alarm floods and the risk of industrial accidents by providing early warnings of potential future alarms, thereby allowing operators more time to take corrective action. However, in continuous industrial processes, varying operating conditions and abnormal states cause real-time fluctuations in alarm rates, posing challenges for existing methods to achieve satisfactory prediction performance. In view of such issues, this paper proposes a new alarm event prediction method adapting to variable alarm rates over long-term consecutive alarm monitoring periods using multi-dimensional sequence embedding and improved Informer. The contributions are threefold: 1) An adaptive alarm sequence segmentation strategy is designed to generate input alarm sequences adapting to alarm rates; 2) a multi-dimensional sequence embedding method based on both the alarm tags and time intervals is proposed to convert the textual alarm messages into numerical vectors; and 3) an Informer based alarm event prediction model is developed for precise and early alarm event prediction under alarm flood and non-flood periods. A case study based on the Vinyl Acetate Monomer public model is given to prove the effectiveness of the proposed method.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112317"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146098524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automated identification of pilot failures in aviation accidents using a BERT-based classifier and topic modeling 基于bert分类器和主题建模的航空事故飞行员故障自动识别

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-01-24 DOI: 10.1016/j.ress.2026.112284

July B. Macedo , Plínio M.S. Ramos , Caio B.S. Maior , Márcio J.C. Moura , Isis D. Lins

Aviation accidents are a significant concern due to the potential loss of life, and human factors have emerged as the primary underlying cause. The aviation industry maintains extensive records, including accident investigation reports, which offer valuable insights for decision-making. This research explores the application of Natural Language Processing (NLP) as a solution for analyzing these documents and conducting risk assessments, empowering experts to manage accidents better and develop effective preventive measures. We propose a novel methodology that leverages a Bidirectional Encoder Representations from Transformers (BERT)-based classifier, combined with topic modeling techniques, to automate the labeling of accident datasets and identify key pilot failures contributing to aviation accidents. This automated labeling process is a critical step in efficiently creating a high-quality dataset essential for training a classifier capable of accurately detecting specific failure types. By applying the methodology to accident reports from the National Transportation Safety Board (NTSB), we successfully trained a classifier that identifies pilot failures, such as skill-based errors, routine violations, and perceptual errors. This study contributes to the field by introducing an innovative integration of contextual embeddings and topic modeling, significantly reducing manual efforts in data preparation while enhancing the precision and efficiency of analyzing aviation accident data. The findings demonstrate the potential of NLP to streamline the analysis of accident reports, assisting experts in developing targeted training programs, procedural improvements, and risk mitigation strategies to address pilot-related errors effectively.

由于可能造成生命损失，航空事故是一个重大问题，而人为因素已成为主要的潜在原因。航空业保存着大量的记录，包括事故调查报告，这些记录为决策提供了有价值的见解。本研究探讨了自然语言处理（NLP）作为分析这些文件和进行风险评估的解决方案的应用，使专家能够更好地管理事故并制定有效的预防措施。我们提出了一种新的方法，利用基于变形金刚（BERT）的双向编码器表示分类器，结合主题建模技术，自动标记事故数据集，并识别导致航空事故的关键飞行员故障。这种自动标记过程是有效创建高质量数据集的关键步骤，这对于训练能够准确检测特定故障类型的分类器至关重要。通过将该方法应用于美国国家运输安全委员会（NTSB）的事故报告，我们成功地训练了一个分类器，该分类器可以识别飞行员的故障，例如基于技能的错误、常规违规和感知错误。本研究通过引入上下文嵌入和主题建模的创新集成，大大减少了数据准备的人工工作，同时提高了分析航空事故数据的精度和效率，从而为该领域做出了贡献。研究结果表明，NLP在简化事故报告分析、协助专家制定有针对性的培训计划、程序改进和风险缓解策略方面具有潜力，可以有效地解决与飞行员相关的错误。

{"title":"Automated identification of pilot failures in aviation accidents using a BERT-based classifier and topic modeling","authors":"July B. Macedo , Plínio M.S. Ramos , Caio B.S. Maior , Márcio J.C. Moura , Isis D. Lins","doi":"10.1016/j.ress.2026.112284","DOIUrl":"10.1016/j.ress.2026.112284","url":null,"abstract":"<div><div>Aviation accidents are a significant concern due to the potential loss of life, and human factors have emerged as the primary underlying cause. The aviation industry maintains extensive records, including accident investigation reports, which offer valuable insights for decision-making. This research explores the application of Natural Language Processing (NLP) as a solution for analyzing these documents and conducting risk assessments, empowering experts to manage accidents better and develop effective preventive measures. We propose a novel methodology that leverages a Bidirectional Encoder Representations from Transformers (BERT)-based classifier, combined with topic modeling techniques, to automate the labeling of accident datasets and identify key pilot failures contributing to aviation accidents. This automated labeling process is a critical step in efficiently creating a high-quality dataset essential for training a classifier capable of accurately detecting specific failure types. By applying the methodology to accident reports from the National Transportation Safety Board (NTSB), we successfully trained a classifier that identifies pilot failures, such as skill-based errors, routine violations, and perceptual errors. This study contributes to the field by introducing an innovative integration of contextual embeddings and topic modeling, significantly reducing manual efforts in data preparation while enhancing the precision and efficiency of analyzing aviation accident data. The findings demonstrate the potential of NLP to streamline the analysis of accident reports, assisting experts in developing targeted training programs, procedural improvements, and risk mitigation strategies to address pilot-related errors effectively.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112284"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-constrained digital twin framework for deformation analysis and safety assessment of high earth-rock dams 高土石坝变形分析与安全评价的物理约束数字孪生框架

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-02-01 DOI: 10.1016/j.ress.2026.112348

Zhihong Huang , Gang Ma , Zhitao Ai , Jiawei Wang , Xiaolin Chang , Wei Zhou

Accurate deformation analysis is crucial for the safety assessment and risk management of high earth-rock dams. While conventional surrogate-assisted optimization improves prediction accuracy, the neglect of intrinsic physical parameter correlations often leads to non-unique solutions, limited accuracy gains, and numerical divergence. This study proposes a physics-constrained digital twin (DT) framework that enables high-fidelity virtual-physical synchronization. The key innovation is a physical constraint mechanism utilizing a β-variational autoencoder (β-VAE) to extract parameter correlations from global experimental datasets as prior knowledge. By integrating this mechanism with multi-objective optimization and an elite archiving strategy, the framework ensures stable and physically consistent model evolution. Validated on the 303 m high LHK dam, the results demonstrate a 28 % improvement in prediction accuracy and a transition to near real-time computational performance. This framework provides a more reliable and physically consistent modeling approach for intelligent dam operation and lifecycle risk assessment.

准确的变形分析对高土石坝的安全评价和风险管理至关重要。虽然传统的代理辅助优化提高了预测精度，但忽略了内在的物理参数相关性通常会导致非唯一解、有限的精度增益和数值发散。本研究提出了一个物理约束的数字孪生（DT）框架，实现高保真的虚拟物理同步。关键创新是利用β-变分自编码器（β-VAE）从全局实验数据集中提取参数相关性作为先验知识的物理约束机制。通过将该机制与多目标优化和精英存档策略相结合，该框架确保了模型演化的稳定性和物理一致性。在303 m高的LHK大坝上进行了验证，结果表明预测精度提高了28%，并向接近实时的计算性能过渡。该框架为大坝智能运行和生命周期风险评估提供了一种更加可靠和物理一致的建模方法。

{"title":"Physics-constrained digital twin framework for deformation analysis and safety assessment of high earth-rock dams","authors":"Zhihong Huang , Gang Ma , Zhitao Ai , Jiawei Wang , Xiaolin Chang , Wei Zhou","doi":"10.1016/j.ress.2026.112348","DOIUrl":"10.1016/j.ress.2026.112348","url":null,"abstract":"<div><div>Accurate deformation analysis is crucial for the safety assessment and risk management of high earth-rock dams. While conventional surrogate-assisted optimization improves prediction accuracy, the neglect of intrinsic physical parameter correlations often leads to non-unique solutions, limited accuracy gains, and numerical divergence. This study proposes a physics-constrained digital twin (DT) framework that enables high-fidelity virtual-physical synchronization. The key innovation is a physical constraint mechanism utilizing a <em>β</em>-variational autoencoder (<em>β</em>-VAE) to extract parameter correlations from global experimental datasets as prior knowledge. By integrating this mechanism with multi-objective optimization and an elite archiving strategy, the framework ensures stable and physically consistent model evolution. Validated on the 303 m high LHK dam, the results demonstrate a 28 % improvement in prediction accuracy and a transition to near real-time computational performance. This framework provides a more reliable and physically consistent modeling approach for intelligent dam operation and lifecycle risk assessment.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112348"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Robustness of disaster response systems: Hypergraph modelling, key local structures, and fortification methods 灾难响应系统的稳健性：超图建模、关键的局部结构和加固方法

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-02-07 DOI: 10.1016/j.ress.2026.112372

Chong Gao , Hui Jiang , Mang Li

Disaster response systems (DRS) operate in an uncertain environment and are continuously tested by external perturbations and internal failures. The ability to withstand disruptions and mitigate the risk of structural disintegration is a defining aspect of a resilient DRS. The current research on the resilience of DRS aims to reveal its robustness performance using metrics from social network analysis. However, the correspondence between these metrics and the underlying ability to maintain structural integrity is neither direct nor well-established. In this paper, we first construct faithful representations of DRS using hypergraph modelling. We demonstrate that hypergraph-based representations provide a principled basis for robustness analysis, revealing local structures key to structural integrity. We establish comprehensive robustness metrics and introduce a principled analysis framework to formalise these key local structures. We also perform a stability decomposition and obtain stability indicators that are used to identify the unstable local structures with weak internal coherence and fragile external embedding. Then we formulate the robustness-enhancing problem and develop fortification methods aiming at maximising the total robustness gain. Extensive simulations demonstrate that our approach substantially outperforms alternative strategies, including coreness-based and graph-based methods, across a wide range of fortification budgets. In addition, the proposed method has an efficient numerical implementation, and we validate it in large-scale synthetic hypergraphs.

灾害响应系统（DRS）在不确定的环境中运行，并不断受到外部扰动和内部故障的考验。抵御破坏和减轻结构解体风险的能力是弹性DRS的一个决定性方面。目前对DRS弹性的研究旨在利用社会网络分析的指标来揭示其鲁棒性。然而，这些指标与维护结构完整性的潜在能力之间的对应关系既不直接也不完善。在本文中，我们首先使用超图建模构造了DRS的忠实表示。我们证明了基于超图的表示为鲁棒性分析提供了原则基础，揭示了对结构完整性至关重要的局部结构。我们建立了全面的稳健性指标，并引入了一个原则性的分析框架来形式化这些关键的局部结构。我们还进行了稳定性分解，得到了稳定性指标，用于识别内部相干性弱、外部嵌入脆弱的不稳定局部结构。然后，我们提出了鲁棒性增强问题，并开发了旨在最大化总鲁棒性增益的强化方法。广泛的模拟表明，我们的方法大大优于其他策略，包括基于核心和基于图形的方法，在广泛的防御预算。此外，该方法具有高效的数值实现，并在大规模合成超图中进行了验证。

{"title":"Robustness of disaster response systems: Hypergraph modelling, key local structures, and fortification methods","authors":"Chong Gao , Hui Jiang , Mang Li","doi":"10.1016/j.ress.2026.112372","DOIUrl":"10.1016/j.ress.2026.112372","url":null,"abstract":"<div><div>Disaster response systems (DRS) operate in an uncertain environment and are continuously tested by external perturbations and internal failures. The ability to withstand disruptions and mitigate the risk of structural disintegration is a defining aspect of a resilient DRS. The current research on the resilience of DRS aims to reveal its robustness performance using metrics from social network analysis. However, the correspondence between these metrics and the underlying ability to maintain structural integrity is neither direct nor well-established. In this paper, we first construct faithful representations of DRS using hypergraph modelling. We demonstrate that hypergraph-based representations provide a principled basis for robustness analysis, revealing local structures key to structural integrity. We establish comprehensive robustness metrics and introduce a principled analysis framework to formalise these key local structures. We also perform a stability decomposition and obtain stability indicators that are used to identify the unstable local structures with weak internal coherence and fragile external embedding. Then we formulate the robustness-enhancing problem and develop fortification methods aiming at maximising the total robustness gain. Extensive simulations demonstrate that our approach substantially outperforms alternative strategies, including coreness-based and graph-based methods, across a wide range of fortification budgets. In addition, the proposed method has an efficient numerical implementation, and we validate it in large-scale synthetic hypergraphs.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112372"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimization of isolation valve operation and identification of critical components for enhancing the resilience of water distribution systems 提高配水系统弹性的隔离阀操作优化和关键部件识别

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-09-01 Epub Date: 2026-02-09 DOI: 10.1016/j.ress.2026.112387

Xiaoli Liu , Mingyuan Zhang , Haixing Liu

This study proposes a framework to enhance the resilience of water distribution systems (WDSs) through optimized isolation valve operation and identification of critical components. First, the WDS topology is converted into a segment-valve model to efficiently identify the isolation valves for isolating failed pipes. Second, quantitative metrics are established from both hydraulic and water quality perspectives to assess service performance losses caused by isolation valve closures (IVCs). Subsequently, an optimization model is developed to determine the IVC scheme that minimizes cumulative performance loss (CPL) during failures. Finally, a criticality assessment framework is introduced to accurately identify segments and isolation valves that significantly impact system resilience. The proposed framework was validated on two real-world WDSs in China with distinct topological configurations. The results indicate that the optimized IVC scheme reduces average CPL by approximately 15% and 26.63%, respectively, compared to the minimum isolation time scheme and its reverse scheme. The distribution characteristics of critical components vary across WDSs with different topologies. Furthermore, implementing N-valve and N-1 valve configurations reduces the average criticality of segments by 71.02% and 64.69%, respectively, thereby enhancing system resilience. This study provides decision support for developing efficient isolation valve operation schemes and precise component maintenance strategies.

本研究提出了一个框架，通过优化隔离阀操作和识别关键部件来增强配水系统（wds）的弹性。首先，将WDS拓扑转换为分段阀模型，有效识别隔离阀，隔离失效管道；其次，从水力和水质的角度建立定量指标，以评估隔离阀关闭（IVCs）造成的服务性能损失。随后，开发了一个优化模型来确定IVC方案，以最小化故障期间的累积性能损失（CPL）。最后，引入了一个临界评估框架，以准确识别显著影响系统弹性的部分和隔离阀。该框架在中国两个具有不同拓扑结构的现实世界wds上进行了验证。结果表明，优化后的IVC方案与最小隔离时间方案和相反隔离时间方案相比，平均CPL分别降低约15%和26.63%。在具有不同拓扑结构的wds中，关键组件的分布特征各不相同。此外，采用n阀和N-1阀配置，将分段的平均临界度分别降低了71.02%和64.69%，从而增强了系统的弹性。该研究为制定有效的隔离阀操作方案和精确的部件维护策略提供决策支持。

{"title":"Optimization of isolation valve operation and identification of critical components for enhancing the resilience of water distribution systems","authors":"Xiaoli Liu , Mingyuan Zhang , Haixing Liu","doi":"10.1016/j.ress.2026.112387","DOIUrl":"10.1016/j.ress.2026.112387","url":null,"abstract":"<div><div>This study proposes a framework to enhance the resilience of water distribution systems (WDSs) through optimized isolation valve operation and identification of critical components. First, the WDS topology is converted into a segment-valve model to efficiently identify the isolation valves for isolating failed pipes. Second, quantitative metrics are established from both hydraulic and water quality perspectives to assess service performance losses caused by isolation valve closures (IVCs). Subsequently, an optimization model is developed to determine the IVC scheme that minimizes cumulative performance loss (CPL) during failures. Finally, a criticality assessment framework is introduced to accurately identify segments and isolation valves that significantly impact system resilience. The proposed framework was validated on two real-world WDSs in China with distinct topological configurations. The results indicate that the optimized IVC scheme reduces average CPL by approximately 15% and 26.63%, respectively, compared to the minimum isolation time scheme and its reverse scheme. The distribution characteristics of critical components vary across WDSs with different topologies. Furthermore, implementing <em>N</em>-valve and <em>N</em>-1 valve configurations reduces the average criticality of segments by 71.02% and 64.69%, respectively, thereby enhancing system resilience. This study provides decision support for developing efficient isolation valve operation schemes and precise component maintenance strategies.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112387"},"PeriodicalIF":11.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0