Reliability Engineering & System Safety最新文献_第2页

A novel multi-task causal representation learning approach for interpretable maritime collision severity prediction 一种新的多任务因果表示学习方法用于可解释的海上碰撞严重程度预测

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112236

Miaomiao Wang , Yanfu Wang , Zhicheng Ma , Jin Wang

Accurate prediction and robust interpretation of maritime collision severity are crucial. Prevailing correlation-based methods are non-robust and lack interpretability, struggling with confounding factors, data heterogeneity, and class imbalance. A novel multi-task causal representation learning framework (MCLF) is proposed to address these limitations. Its core is a structured disentanglement mechanism that decomposes effects into direct effects and indirect effects mediated by unsafe factors, reinforced by adversarial training and orthogonality constraints to reduce representation-level confounding induced by observed covariates, thereby improving robustness and interpretability. To address class imbalance, an interactive data synthesis module using the tabular denoising diffusion probabilistic model (TabDDPM) is used, which generates high-quality samples for hard-to-classify cases to enhance model robustness. A dynamic multi-task fusion strategy then adaptively integrates the primary severity prediction with auxiliary tasks (pollution, property loss, and death). This holistic approach achieves superior predictive accuracy and enhances interpretability by providing a structurally-grounded decomposition of effects, advancing towards more transparent decision-support in maritime safety.

海上碰撞严重程度的准确预测和可靠解释至关重要。目前流行的基于相关性的方法不稳健，缺乏可解释性，与混杂因素、数据异质性和类不平衡作斗争。提出了一种新的多任务因果表示学习框架（MCLF）来解决这些限制。其核心是结构化解纠缠机制，该机制将效应分解为不安全因素介导的直接效应和间接效应，并通过对抗性训练和正交性约束进行强化，以减少观测协变量引起的表征水平混淆，从而提高鲁棒性和可解释性。为了解决分类不平衡问题，采用了基于表格去噪扩散概率模型（TabDDPM）的交互式数据综合模块，该模块为难以分类的案例生成高质量的样本，增强了模型的鲁棒性。动态多任务融合策略自适应地将主要严重程度预测与辅助任务（污染、财产损失和死亡）集成在一起。这种整体方法实现了卓越的预测准确性，并通过提供基于结构的影响分解来增强可解释性，从而向更透明的海上安全决策支持迈进。

{"title":"A novel multi-task causal representation learning approach for interpretable maritime collision severity prediction","authors":"Miaomiao Wang , Yanfu Wang , Zhicheng Ma , Jin Wang","doi":"10.1016/j.ress.2026.112236","DOIUrl":"10.1016/j.ress.2026.112236","url":null,"abstract":"<div><div>Accurate prediction and robust interpretation of maritime collision severity are crucial. Prevailing correlation-based methods are non-robust and lack interpretability, struggling with confounding factors, data heterogeneity, and class imbalance. A novel multi-task causal representation learning framework (MCLF) is proposed to address these limitations. Its core is a structured disentanglement mechanism that decomposes effects into direct effects and indirect effects mediated by unsafe factors, reinforced by adversarial training and orthogonality constraints to reduce representation-level confounding induced by observed covariates, thereby improving robustness and interpretability. To address class imbalance, an interactive data synthesis module using the tabular denoising diffusion probabilistic model (TabDDPM) is used, which generates high-quality samples for hard-to-classify cases to enhance model robustness. A dynamic multi-task fusion strategy then adaptively integrates the primary severity prediction with auxiliary tasks (pollution, property loss, and death). This holistic approach achieves superior predictive accuracy and enhances interpretability by providing a structurally-grounded decomposition of effects, advancing towards more transparent decision-support in maritime safety.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112236"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039123","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

Probabilistic assessment of dynamic urban evacuation-sheltering functionality under typhoons based on interdependent road-shelter network 基于相互依存道路-遮蔽网络的台风下城市动态疏散-遮蔽功能概率评估

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112240

Lu Zhang , Shuang Tan , Bo Chen , Qingshan Yang

The functionality of the urban evacuation-sheltering system (UESS) during typhoons depends on the interdependent operational states of both the road network and shelters. Temporal variations in physical damage and traffic demand cause dynamic degradation in UESS functionality. Existing accessibility metrics are limited in capturing this dynamic behavior due to incomplete functional quantification, limited generalizability, lack of standardized outputs, and numerical instability under time-varying traffic demand. To address these issues, this study proposes a holistic functionality metric that captures temporal variations in evacuation timeliness and shelter availability, accompanied by a probabilistic assessment framework for UESS. A case study in Futian District, Shenzhen, reveals a distinct three-stage evolution of UESS functionality during typhoons: an initial sharp decline, a phase of slight recovery and stabilization, and a secondary decline followed by stabilization. The duration and magnitude of UESS functionality changes at each stage vary markedly across typhoon scenarios, with multisource uncertainties further exacerbate spatial disparities in UESS functionality among urban zones. The proposed metric and framework enhance the characterization and uncertainty quantification of dynamic UESS functionality, providing valuable insights for identifying urban areas with weak UESS performance, optimizing evacuation strategies, and strengthening urban resilience to typhoons.

台风期间城市疏散-庇护系统（UESS）的功能取决于道路网络和庇护所的相互依赖的运行状态。物理损坏和交通需求的时间变化导致UESS功能的动态退化。由于功能量化不完整、泛化能力有限、缺乏标准化输出以及时变交通需求下的数值不稳定，现有的可达性指标在捕捉这种动态行为方面受到限制。为了解决这些问题，本研究提出了一个整体的功能指标，该指标可以捕捉疏散及时性和庇护所可用性的时间变化，并伴随着UESS的概率评估框架。以深圳福田区为例，揭示了台风期间UESS功能的明显三个阶段的演变：最初的急剧下降，一个轻微的恢复和稳定阶段，以及第二个下降之后稳定的阶段。不同台风情景下各阶段UESS功能变化的持续时间和幅度存在显著差异，多源不确定性进一步加剧了城市区域间UESS功能的空间差异。所提出的度量和框架增强了动态UESS功能的表征和不确定性量化，为识别UESS性能较弱的城市地区、优化疏散策略和增强城市对台风的抵御能力提供了有价值的见解。

{"title":"Probabilistic assessment of dynamic urban evacuation-sheltering functionality under typhoons based on interdependent road-shelter network","authors":"Lu Zhang , Shuang Tan , Bo Chen , Qingshan Yang","doi":"10.1016/j.ress.2026.112240","DOIUrl":"10.1016/j.ress.2026.112240","url":null,"abstract":"<div><div>The functionality of the urban evacuation-sheltering system (UESS) during typhoons depends on the interdependent operational states of both the road network and shelters. Temporal variations in physical damage and traffic demand cause dynamic degradation in UESS functionality. Existing accessibility metrics are limited in capturing this dynamic behavior due to incomplete functional quantification, limited generalizability, lack of standardized outputs, and numerical instability under time-varying traffic demand. To address these issues, this study proposes a holistic functionality metric that captures temporal variations in evacuation timeliness and shelter availability, accompanied by a probabilistic assessment framework for UESS. A case study in Futian District, Shenzhen, reveals a distinct three-stage evolution of UESS functionality during typhoons: an initial sharp decline, a phase of slight recovery and stabilization, and a secondary decline followed by stabilization. The duration and magnitude of UESS functionality changes at each stage vary markedly across typhoon scenarios, with multisource uncertainties further exacerbate spatial disparities in UESS functionality among urban zones. The proposed metric and framework enhance the characterization and uncertainty quantification of dynamic UESS functionality, providing valuable insights for identifying urban areas with weak UESS performance, optimizing evacuation strategies, and strengthening urban resilience to typhoons.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112240"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039092","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

Bound-constrained nonstationary Gaussian process regression for ventilated cavitation prediction 通风空化预测的约束非平稳高斯过程回归

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112222

Tian Bai , Kuangqi Chen , Dianpeng Wang

This research is motivated by the need to predict cavity length in ventilated cavitation experiments. Given the freestream velocity and ventilation rate, the underlying physical mechanism gives rise to informative bounds and distinct shedding regimes, which in turn induce nonstationarity in the system. In this paper, we propose a novel nonstationary bounded Gaussian process regression model that simultaneously incorporates nonstationarity and bound constraints. We adopt the Gaussian process projection framework to enforce the bound constraints and propose a mixture of bounded Gaussian processes to capture nonstationarity, where each component models a locally stationary behavior consistent with the constraints. The model parameters and mixture components are estimated through a two-stage procedure. Importantly, the proposed model reveals latent physical mechanisms by identifying distinct components, thereby offering deeper scientific insights into the input-output relationship. Numerical results across test functions and the ventilated cavitation experiments validate the superiority of the proposed method. Specifically, the proposed method effectively captures the evolution of ventilated cavity structures, thereby significantly enhancing the adaptability and operational reliability of high-speed underwater vehicles under complex environmental conditions. Codes are available on https://github.com/tbai114/Nonstationary-bounded-Gaussian-process.

本研究的动机是需要预测通风空化实验中的空腔长度。考虑到自由流速度和通风量，潜在的物理机制产生了信息边界和不同的脱落机制，这反过来又导致了系统的非平稳性。本文提出了一种同时包含非平稳约束和有界约束的非平稳有界高斯过程回归模型。我们采用高斯过程投影框架来加强有界约束，并提出了一个有界高斯过程的混合来捕获非平稳性，其中每个组件建模与约束一致的局部平稳行为。模型参数和混合成分通过两阶段估计。重要的是，提出的模型通过识别不同的组成部分揭示了潜在的物理机制，从而为投入产出关系提供了更深入的科学见解。跨测试函数的数值结果和通风空化实验验证了该方法的优越性。具体而言，该方法有效地捕捉了通气腔体结构的演化过程，从而显著提高了高速水下航行器在复杂环境条件下的适应性和运行可靠性。代码可在https://github.com/tbai114/Nonstationary-bounded-Gaussian-process上获得。

{"title":"Bound-constrained nonstationary Gaussian process regression for ventilated cavitation prediction","authors":"Tian Bai , Kuangqi Chen , Dianpeng Wang","doi":"10.1016/j.ress.2026.112222","DOIUrl":"10.1016/j.ress.2026.112222","url":null,"abstract":"<div><div>This research is motivated by the need to predict cavity length in ventilated cavitation experiments. Given the freestream velocity and ventilation rate, the underlying physical mechanism gives rise to informative bounds and distinct shedding regimes, which in turn induce nonstationarity in the system. In this paper, we propose a novel nonstationary bounded Gaussian process regression model that simultaneously incorporates nonstationarity and bound constraints. We adopt the Gaussian process projection framework to enforce the bound constraints and propose a mixture of bounded Gaussian processes to capture nonstationarity, where each component models a locally stationary behavior consistent with the constraints. The model parameters and mixture components are estimated through a two-stage procedure. Importantly, the proposed model reveals latent physical mechanisms by identifying distinct components, thereby offering deeper scientific insights into the input-output relationship. Numerical results across test functions and the ventilated cavitation experiments validate the superiority of the proposed method. Specifically, the proposed method effectively captures the evolution of ventilated cavity structures, thereby significantly enhancing the adaptability and operational reliability of high-speed underwater vehicles under complex environmental conditions. Codes are available on <span><span>https://github.com/tbai114/Nonstationary-bounded-Gaussian-process</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112222"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039664","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 cost-reliability optimization framework for aging water distribution networks with partial demand surges considering time-dependent deterioration effect 考虑时变劣化效应的部分需求激增老化配水网络成本可靠性优化框架

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112239

Liangcheng Yu , Mingyuan Zhang , Haixing Liu , Yunhu Liu

The synergistic impacts of structural deterioration in aging urban water distribution networks (WDNs) and nonlinear demand surges in partial regions pose dual threats to water supply security. Current research inadequately addresses the dynamic time-dependent effects of aging pipelines on network operational reliability. This study proposes a cost-reliability optimization framework for aging WDNs under partial demand surge scenarios, integrating time-dependent reliability theory and evolutionary algorithms. The framework employs a Gaussian stochastic process-based corrosion power-law model to quantify time-dependent pipe failure probabilities, subsequently transformed into deterioration coefficients. These coefficients are systematically integrated with topological connectivity, junction hydraulic importance, and pipeline uniformity coefficients to formulate a novel Time-dependent Reliability Index (TRI). Comprehensively evaluate pipeline stress strength to establish deterioration pressure boundaries, ensuring safe operating conditions for aged WDNs. Within a Non-dominated Sorting Genetic Algorithm II (NSGA-II) architecture, Pareto frontiers are established for operational decisions on variable-speed pump groups, simultaneously minimizing energy cost and maximizing reliability. Implementing Anytown and a Chinese city's large-scale WDN reveals critical insights: TRI provides a more robust assessment than traditional reliability metrics in quantifying the long-term time-dependent system reliability of WDN, and preventive maintenance based on pressure thresholds effectively mitigates the progressive reliability degradation associated with pipeline deterioration. A more critical finding is that differences in WDN topology and scale cause partial demand surges to slightly enhance the reliability of large-scale aged WDNs, while small-scale WDNs exhibit weakened reliability. This framework offers a scientific WDN reliability enhancement pattern that delivers cost-effective and highly reliable operation solutions in aging WDNs.

老化的城市配水网络结构退化和部分地区的非线性需求激增的协同影响对供水安全构成双重威胁。目前的研究还没有充分解决老化管道对网络运行可靠性的动态时效影响。结合时变可靠性理论和进化算法，提出了部分需求激增情景下老化wdn的成本-可靠性优化框架。该框架采用基于高斯随机过程的腐蚀幂律模型来量化随时间变化的管道失效概率，然后将其转换为劣化系数。将这些系数与拓扑连通性、连接处水力重要性和管道均匀性系数进行系统集成，形成一种新的时间相关可靠性指数（TRI）。综合评价管道应力强度，建立劣化压力边界，确保老化水轮机的安全运行条件。在非支配排序遗传算法II （NSGA-II）架构中，为变速泵组的操作决策建立了帕累托边界，同时最小化能源成本和最大化可靠性。实施Anytown和中国城市的大规模WDN揭示了关键的见解：TRI在量化WDN长期依赖时间的系统可靠性方面提供了比传统可靠性指标更稳健的评估，基于压力阈值的预防性维护有效地减轻了与管道退化相关的渐进式可靠性退化。一个更重要的发现是，WDN拓扑和规模的差异会导致部分需求激增，从而略微提高大规模老化WDN的可靠性，而小规模WDN的可靠性则会减弱。该框架提供了一种科学的WDN可靠性增强模式，为老化的WDN提供了经济高效、高可靠性的运行解决方案。

{"title":"A cost-reliability optimization framework for aging water distribution networks with partial demand surges considering time-dependent deterioration effect","authors":"Liangcheng Yu , Mingyuan Zhang , Haixing Liu , Yunhu Liu","doi":"10.1016/j.ress.2026.112239","DOIUrl":"10.1016/j.ress.2026.112239","url":null,"abstract":"<div><div>The synergistic impacts of structural deterioration in aging urban water distribution networks (WDNs) and nonlinear demand surges in partial regions pose dual threats to water supply security. Current research inadequately addresses the dynamic time-dependent effects of aging pipelines on network operational reliability. This study proposes a cost-reliability optimization framework for aging WDNs under partial demand surge scenarios, integrating time-dependent reliability theory and evolutionary algorithms. The framework employs a Gaussian stochastic process-based corrosion power-law model to quantify time-dependent pipe failure probabilities, subsequently transformed into deterioration coefficients. These coefficients are systematically integrated with topological connectivity, junction hydraulic importance, and pipeline uniformity coefficients to formulate a novel Time-dependent Reliability Index (TRI). Comprehensively evaluate pipeline stress strength to establish deterioration pressure boundaries, ensuring safe operating conditions for aged WDNs. Within a Non-dominated Sorting Genetic Algorithm II (NSGA-II) architecture, Pareto frontiers are established for operational decisions on variable-speed pump groups, simultaneously minimizing energy cost and maximizing reliability. Implementing Anytown and a Chinese city's large-scale WDN reveals critical insights: TRI provides a more robust assessment than traditional reliability metrics in quantifying the long-term time-dependent system reliability of WDN, and preventive maintenance based on pressure thresholds effectively mitigates the progressive reliability degradation associated with pipeline deterioration. A more critical finding is that differences in WDN topology and scale cause partial demand surges to slightly enhance the reliability of large-scale aged WDNs, while small-scale WDNs exhibit weakened reliability. This framework offers a scientific WDN reliability enhancement pattern that delivers cost-effective and highly reliable operation solutions in aging WDNs.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112239"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039666","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-scale signal transformer with signal processing-based attention interpretation for fault diagnosis of rotating machinery under variable speed conditions 基于信号处理的多尺度信号变压器在旋转机械变速故障诊断中的应用

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112243

Sungjong Kim , Seungyun Lee , Jiwon Lee , Minjae Kim , Heonjun Yoon , Byeng D. Youn

Fault diagnosis of rotating machinery remains a critical challenge due to the need for precise localization of fault-related patterns in time-series signals, which are often influenced by variable speed conditions. Moreover, the limited interpretability of deep learning models restricts their deployment in practical industrial applications. To address these issues, this study proposes the Multi-Scale Signal Transformer (MSSiT), a novel deep learning architecture that adaptively captures multi-resolution features using a multi-scale self-attention mechanism. In addition, a Signal Processing-Based Attention Interpretation (SPAI) method is developed to enhance model interpretability by analyzing attention weight in both time and frequency domains through signal processing techniques.

The proposed framework is validated through two case studies involving various and time-varying speed conditions. Experimental results demonstrate that MSSiT outperforms existing fault diagnosis methods under variable speed conditions. The SPAI results reveal that multi-scale attention mechanism focuses on high-frequency components for fault-related features and low-frequency components for speed-dependent trends at each scale. Furthermore, interpretation results are quantitatively validated, confirming that attention weight reliably highlights physically meaningful features. These findings demonstrate the effectiveness and robustness of the proposed framework and its potential applicability in real-world fault diagnosis under complex operational conditions.

由于需要精确定位时间序列信号中与故障相关的模式，旋转机械的故障诊断仍然是一个关键的挑战，而时间序列信号通常受变速条件的影响。此外，深度学习模型有限的可解释性限制了它们在实际工业应用中的部署。为了解决这些问题，本研究提出了多尺度信号变压器（MSSiT），这是一种新型的深度学习架构，可以使用多尺度自注意机制自适应捕获多分辨率特征。此外，提出了一种基于信号处理的注意解释（SPAI）方法，通过信号处理技术在时域和频域分析注意权重，提高模型的可解释性。通过两个涉及各种时变速度条件的案例研究验证了所提出的框架。实验结果表明，该方法在变速条件下优于现有的故障诊断方法。SPAI结果表明，在每个尺度上，故障相关特征的多尺度注意机制主要集中在高频分量上，速度相关趋势的多尺度注意机制主要集中在低频分量上。此外，对解释结果进行了定量验证，证实了注意权重可靠地突出了物理上有意义的特征。这些结果证明了所提出的框架的有效性和鲁棒性，以及它在复杂操作条件下的实际故障诊断中的潜在适用性。

{"title":"Multi-scale signal transformer with signal processing-based attention interpretation for fault diagnosis of rotating machinery under variable speed conditions","authors":"Sungjong Kim , Seungyun Lee , Jiwon Lee , Minjae Kim , Heonjun Yoon , Byeng D. Youn","doi":"10.1016/j.ress.2026.112243","DOIUrl":"10.1016/j.ress.2026.112243","url":null,"abstract":"<div><div>Fault diagnosis of rotating machinery remains a critical challenge due to the need for precise localization of fault-related patterns in time-series signals, which are often influenced by variable speed conditions. Moreover, the limited interpretability of deep learning models restricts their deployment in practical industrial applications. To address these issues, this study proposes the Multi-Scale Signal Transformer (MSSiT), a novel deep learning architecture that adaptively captures multi-resolution features using a multi-scale self-attention mechanism. In addition, a Signal Processing-Based Attention Interpretation (SPAI) method is developed to enhance model interpretability by analyzing attention weight in both time and frequency domains through signal processing techniques.</div><div>The proposed framework is validated through two case studies involving various and time-varying speed conditions. Experimental results demonstrate that MSSiT outperforms existing fault diagnosis methods under variable speed conditions. The SPAI results reveal that multi-scale attention mechanism focuses on high-frequency components for fault-related features and low-frequency components for speed-dependent trends at each scale. Furthermore, interpretation results are quantitatively validated, confirming that attention weight reliably highlights physically meaningful features. These findings demonstrate the effectiveness and robustness of the proposed framework and its potential applicability in real-world fault diagnosis under complex operational conditions.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112243"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039121","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

Predictive risk analysis for leakage accidents with dynamic behaviour 动态泄漏事故的预测风险分析

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-14 DOI: 10.1016/j.ress.2026.112238

Xiangyu Kong , Ruishu Huang , He Li , Jichuan Kang , Yan Dong , C. Guedes Soares , Jin Wang

A predictive risk analysis approach is proposed for modelling leakage accidents with dynamic behaviour based on time-series simulations in order to enhance data foundation of risk analysis tasks. Critical failure items are first identified by the Failure Mode and Effects Analysis model and the occurrence probabilities of which are subsequently computed by the Bayesian and Event Tree Analysis methods. The dynamic behaviour of the accidents is simulated to reveal their time-series failure consequences. With the probabilities and consequences obtained, a predictive risk analysis approach is established as a basis to calculate the risk index of accidents with the consideration of dynamic behaviours. The applicability and superior performance of the proposed approach are illustrated by a leakage risk analysis of offshore hydrogen storage systems. Overall, the proposed approach extends the existing inductive risk analysis concepts to predictive patterns and contributes to leakage accidents analysis and prevention with the situation of data and knowledge scarcities.

为了增强风险分析任务的数据基础，提出了一种基于时间序列模拟的具有动态行为的泄漏事故预测风险分析方法。关键故障项首先由失效模式和影响分析模型确定，然后由贝叶斯和事件树分析方法计算其发生概率。模拟了事故的动态行为，揭示了事故的时间序列破坏后果。在得到概率和后果的基础上，建立了一种考虑动态行为的预测风险分析方法，作为计算事故风险指数的基础。通过海上储氢系统的泄漏风险分析，说明了该方法的适用性和优越性。总体而言，该方法将现有的归纳风险分析概念扩展到预测模式，有助于在数据和知识匮乏的情况下进行泄漏事故分析和预防。

{"title":"Predictive risk analysis for leakage accidents with dynamic behaviour","authors":"Xiangyu Kong , Ruishu Huang , He Li , Jichuan Kang , Yan Dong , C. Guedes Soares , Jin Wang","doi":"10.1016/j.ress.2026.112238","DOIUrl":"10.1016/j.ress.2026.112238","url":null,"abstract":"<div><div>A predictive risk analysis approach is proposed for modelling leakage accidents with dynamic behaviour based on time-series simulations in order to enhance data foundation of risk analysis tasks. Critical failure items are first identified by the Failure Mode and Effects Analysis model and the occurrence probabilities of which are subsequently computed by the Bayesian and Event Tree Analysis methods. The dynamic behaviour of the accidents is simulated to reveal their time-series failure consequences. With the probabilities and consequences obtained, a predictive risk analysis approach is established as a basis to calculate the risk index of accidents with the consideration of dynamic behaviours. The applicability and superior performance of the proposed approach are illustrated by a leakage risk analysis of offshore hydrogen storage systems. Overall, the proposed approach extends the existing inductive risk analysis concepts to predictive patterns and contributes to leakage accidents analysis and prevention with the situation of data and knowledge scarcities.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112238"},"PeriodicalIF":11.0,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039127","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 real-time reliability assessment framework for marine mechanical equipment integrating machine learning and physical knowledge: Toward applications in maritime autonomous surface ships 集成机器学习和物理知识的船舶机械设备实时可靠性评估框架：面向海上自主水面舰艇的应用

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-13 DOI: 10.1016/j.ress.2026.112233

Hongqiang Li , Xiangkun Meng , Wenjun Zhang , Xiang-Yu Zhou , Xue Yang

With the rapid development of maritime autonomous surface ships (MASS), the reliability of onboard mechanical equipment has become increasingly critical for safe and efficient operation. Motivated by these emerging requirements, this study proposes a real-time reliability assessment framework for marine mechanical equipment that integrates data-driven models with physical knowledge. By combining physical knowledge with a Wasserstein generative adversarial network (WGAN) to construct a synthetic dataset, the HI is predicted using principal component analysis and a long short-term memory network (PCA-LSTM) model, and the prediction results are optimized using Savitzky-Golay filtering. Finally, real-time reliability quantification is achieved based on the Weibull distribution and maximum likelihood estimation. The case study of a ship propulsion system demonstrates that this method can identify the accelerating trend of reliability reduction after approximately 400 h of operation, and the reliability remains at 99.36% until 720 h. The capability of ML to predict real-time reliability, combined with physical knowledge, reflects real-world conditions. The results provide real-time predictions of the health state and reliability of mechanical equipment, enabling early fault detection and suggesting the formulation of maintenance planning, thereby supporting the reliable operation of MASS.

随着海上自主水面舰艇（MASS）的快速发展，舰载机械设备的可靠性对安全高效运行变得越来越重要。在这些新兴需求的推动下，本研究提出了一种将数据驱动模型与物理知识相结合的船舶机械设备实时可靠性评估框架。将物理知识与Wasserstein生成对抗网络（WGAN）相结合，构建合成数据集，利用主成分分析和长短期记忆网络（PCA-LSTM）模型对HI进行预测，并利用Savitzky-Golay滤波对预测结果进行优化。最后，基于威布尔分布和极大似然估计实现实时可靠性量化。船舶推进系统的实例研究表明，该方法可以识别出运行约400 h后可靠性降低的加速趋势，并且可靠性保持在99.36%，直到720 h。机器学习预测实时可靠性的能力与物理知识相结合，反映了现实情况。研究结果提供了机械设备健康状态和可靠性的实时预测，可以早期发现故障并建议制定维修计划，从而支持MASS的可靠运行。

{"title":"A real-time reliability assessment framework for marine mechanical equipment integrating machine learning and physical knowledge: Toward applications in maritime autonomous surface ships","authors":"Hongqiang Li , Xiangkun Meng , Wenjun Zhang , Xiang-Yu Zhou , Xue Yang","doi":"10.1016/j.ress.2026.112233","DOIUrl":"10.1016/j.ress.2026.112233","url":null,"abstract":"<div><div>With the rapid development of maritime autonomous surface ships (MASS), the reliability of onboard mechanical equipment has become increasingly critical for safe and efficient operation. Motivated by these emerging requirements, this study proposes a real-time reliability assessment framework for marine mechanical equipment that integrates data-driven models with physical knowledge. By combining physical knowledge with a Wasserstein generative adversarial network (WGAN) to construct a synthetic dataset, the HI is predicted using principal component analysis and a long short-term memory network (PCA-LSTM) model, and the prediction results are optimized using Savitzky-Golay filtering. Finally, real-time reliability quantification is achieved based on the Weibull distribution and maximum likelihood estimation. The case study of a ship propulsion system demonstrates that this method can identify the accelerating trend of reliability reduction after approximately 400 h of operation, and the reliability remains at 99.36% until 720 h. The capability of ML to predict real-time reliability, combined with physical knowledge, reflects real-world conditions. The results provide real-time predictions of the health state and reliability of mechanical equipment, enabling early fault detection and suggesting the formulation of maintenance planning, thereby supporting the reliable operation of MASS.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112233"},"PeriodicalIF":11.0,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039668","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

Response and reliability of MDOF Duhem hysteretic system excited by wideband random excitations 宽带随机激励下mof Duhem滞回系统的响应与可靠性

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-13 DOI: 10.1016/j.ress.2026.112226

Qiangfeng Lü , Maolin Deng , Danyu Li

Engineering structures subjected to random external loads often exhibit hysteretic nonlinear behavior. Analyzing the dynamic response of systems with hysteretic restoring forces is crucial for structural fatigue assessment and reliability evaluation. However, theoretical analysis of hysteretic systems remains challenging. Since hysteretic forces cannot be directly treated mathematically, equivalent linearization techniques are typically required. This becomes particularly difficult for multi-degree-of-freedom (MDOF) systems, as most existing studies focus on single-degree-of-freedom (SDOF) systems, and a systematic methodology for simplifying hysteretic forces in MDOF systems has yet to be established. Few studies have derived analytical equivalent expressions for MDOF hysteretic systems. To address this gap, this paper proposes a novel strategy for MDOF Duhem hysteretic systems. By combining the equivalent method with stochastic averaging method, the complex MDOF hysteretic system can be dimensionally reduced, enabling analytical solutions for both stationary response and system reliability. Two numerical examples are presented to demonstrate the proposed methodology. In both cases, analytical equivalent expressions for the Duhem hysteretic restoring force are derived, leading to analytical solutions for the stationary response. Furthermore, the second example additionally provides calculations of the conditional reliability function (CRF) and mean first-passage time (MFPT). Monte Carlo simulations conducted for both examples validate the theoretical predictions, confirming the effectiveness of the proposed method.

工程结构在随机外荷载作用下往往表现出滞回非线性行为。分析具有滞回恢复力的系统的动态响应是结构疲劳评估和可靠性评估的关键。然而，滞回系统的理论分析仍然具有挑战性。由于迟滞力不能直接用数学方法处理，因此通常需要等效的线性化技术。这对于多自由度（MDOF）系统来说尤其困难，因为大多数现有的研究都集中在单自由度（SDOF）系统上，并且尚未建立一个系统的方法来简化mof系统中的滞回力。很少有研究推导出多自由度滞回系统的解析等效表达式。为了解决这一问题，本文提出了一种新的mof - Duhem滞回系统策略。通过将等效方法与随机平均方法相结合，可以对复杂的多自由度滞回系统进行降维，从而实现平稳响应和系统可靠性的解析解。给出了两个数值例子来证明所提出的方法。在这两种情况下，导出了Duhem滞回恢复力的解析等效表达式，从而得到了平稳响应的解析解。此外，第二个例子还提供了条件可靠度函数（CRF）和平均首次通过时间（MFPT）的计算。对两个例子进行了蒙特卡罗模拟，验证了理论预测，证实了所提出方法的有效性。

{"title":"Response and reliability of MDOF Duhem hysteretic system excited by wideband random excitations","authors":"Qiangfeng Lü , Maolin Deng , Danyu Li","doi":"10.1016/j.ress.2026.112226","DOIUrl":"10.1016/j.ress.2026.112226","url":null,"abstract":"<div><div>Engineering structures subjected to random external loads often exhibit hysteretic nonlinear behavior. Analyzing the dynamic response of systems with hysteretic restoring forces is crucial for structural fatigue assessment and reliability evaluation. However, theoretical analysis of hysteretic systems remains challenging. Since hysteretic forces cannot be directly treated mathematically, equivalent linearization techniques are typically required. This becomes particularly difficult for multi-degree-of-freedom (MDOF) systems, as most existing studies focus on single-degree-of-freedom (SDOF) systems, and a systematic methodology for simplifying hysteretic forces in MDOF systems has yet to be established. Few studies have derived analytical equivalent expressions for MDOF hysteretic systems. To address this gap, this paper proposes a novel strategy for MDOF Duhem hysteretic systems. By combining the equivalent method with stochastic averaging method, the complex MDOF hysteretic system can be dimensionally reduced, enabling analytical solutions for both stationary response and system reliability. Two numerical examples are presented to demonstrate the proposed methodology. In both cases, analytical equivalent expressions for the Duhem hysteretic restoring force are derived, leading to analytical solutions for the stationary response. Furthermore, the second example additionally provides calculations of the conditional reliability function (CRF) and mean first-passage time (MFPT). Monte Carlo simulations conducted for both examples validate the theoretical predictions, confirming the effectiveness of the proposed method.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112226"},"PeriodicalIF":11.0,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981141","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

Health assessment for transmission wire ropes subject to dependent failure modes 受相关失效模式影响的传输钢丝绳的健康评估

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-13 DOI: 10.1016/j.ress.2026.112228

Rui Zheng , Zhenglong Liu , Yuan Xing , Tong Niu , Chengcheng Cai , Ercan Altinzoy , Xiangyun Ren

As a crucial component of a wire-driven structure, the transmission wire rope may experience plastic stretching and sliding friction during operation, resulting in lifetime reduction and potential medical accidents. Therefore, it is essential to assess the health status of transmission wire ropes. This paper investigates the health assessment of transition wires subject to dependent failure modes based on experimental data. A fatigue wear experiment is designed based on actual working conditions to test the tension loss of transmission wire ropes. Experimental results show that the transmission wire ropes are subject to two main failure modes: a soft failure indicating that the tension loss exceeds a predetermined level and a hard failure signifying the random fracture of some wire threads. The tension loss process is described by a Wiener process. The hard failure time, dependent on time and tension loss, is characterized as a proportional hazards model. A recursive approach is used to derive recursive formulas for various health indices such as conditional reliability, soft and hard failure probabilities, and remaining useful time. The results of health assessment can support the health management and maintenance decision-making of transmission wire ropes in surgical instruments. Comparison with existing methods demonstrates that the proposed method can produce accurate assessment results with higher efficiency and less memory.

传输钢丝绳作为钢丝绳驱动结构的关键部件，在运行过程中可能会发生塑性拉伸和滑动摩擦，从而导致使用寿命缩短和潜在的医疗事故。因此，对传输钢丝绳的健康状况进行评估是十分必要的。基于实验数据，研究了不同失效模式下过渡导线的健康评估。根据实际工况设计了疲劳磨损试验，对钢丝绳的张力损失进行了测试。试验结果表明，传输钢丝绳主要有两种失效模式：张力损失超过预定水平的软失效和部分钢丝绳随机断裂的硬失效。张力损失过程用维纳过程来描述。硬失效时间依赖于时间和张力损失，其特征为比例风险模型。采用递归方法推导了条件可靠性、软、硬故障概率、剩余使用时间等健康指标的递归公式。健康评估结果可为手术器械传动钢丝绳的健康管理和维修决策提供依据。与现有方法的比较表明，该方法能够以更高的效率和更小的内存产生准确的评估结果。

{"title":"Health assessment for transmission wire ropes subject to dependent failure modes","authors":"Rui Zheng , Zhenglong Liu , Yuan Xing , Tong Niu , Chengcheng Cai , Ercan Altinzoy , Xiangyun Ren","doi":"10.1016/j.ress.2026.112228","DOIUrl":"10.1016/j.ress.2026.112228","url":null,"abstract":"<div><div>As a crucial component of a wire-driven structure, the transmission wire rope may experience plastic stretching and sliding friction during operation, resulting in lifetime reduction and potential medical accidents. Therefore, it is essential to assess the health status of transmission wire ropes. This paper investigates the health assessment of transition wires subject to dependent failure modes based on experimental data. A fatigue wear experiment is designed based on actual working conditions to test the tension loss of transmission wire ropes. Experimental results show that the transmission wire ropes are subject to two main failure modes: a soft failure indicating that the tension loss exceeds a predetermined level and a hard failure signifying the random fracture of some wire threads. The tension loss process is described by a Wiener process. The hard failure time, dependent on time and tension loss, is characterized as a proportional hazards model. A recursive approach is used to derive recursive formulas for various health indices such as conditional reliability, soft and hard failure probabilities, and remaining useful time. The results of health assessment can support the health management and maintenance decision-making of transmission wire ropes in surgical instruments. Comparison with existing methods demonstrates that the proposed method can produce accurate assessment results with higher efficiency and less memory.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112228"},"PeriodicalIF":11.0,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039094","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

NGBoost-Naïve Bayes collaborative deep learning for structural safety evaluation of bridges NGBoost-Naïve基于Bayes协同深度学习的桥梁结构安全评价

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-13 DOI: 10.1016/j.ress.2026.112230

Jin-Ling Zheng , Sheng-En Fang

In structural safety evaluation, machine learning (ML) based methods often exhibit strong data-fitting capabilities but struggle to effectively handle uncertainties in structural response data. Fortunately, Bayesian deep learning (BDL) algorithms can address this drawback by integrating the Bayesian theory with ML algorithms, thereby unifying perception and inference tasks within a single framework. For this purpose, a BDL framework has been proposed combining natural gradient boosting (NGBoost) and the Naïve Bayes theory. The NGBoost serves as the perception component, capturing correlations between deflections at various measurement locations of a healthy structure, while the Shapley Additive Explanation (SHAP) is employed to enhance interpretability. During the training process, the optimal hyperparameters of the NGBoost is objectively determined through Bayesian optimization (BO). The predicted probability distributions of these deflections are treated as hinge variables. By applying the triple standard deviation principle, a structural safety interval is defined to identify scenarios requiring further evaluation. The task-specific component, based on the Naïve Bayes theory, is then utilized to evaluate the structural condition. A bridge benchmark model was used to verify the safety assessment performance under the limited training samples. In addition, a continuous box-girder bridge was employed to further validate the effectiveness of the proposed structural condition indicator. As the structural degradation increased, the condition indicator accurately reflected the degradation variation.

在结构安全评估中，基于机器学习（ML）的方法通常表现出强大的数据拟合能力，但难以有效处理结构响应数据中的不确定性。幸运的是，贝叶斯深度学习（BDL）算法可以通过将贝叶斯理论与ML算法集成来解决这一缺陷，从而在单个框架内统一感知和推理任务。为此，提出了一个结合自然梯度增强（NGBoost）和Naïve贝叶斯理论的BDL框架。NGBoost作为感知组件，捕获健康结构不同测量位置的偏转之间的相关性，而Shapley加性解释（SHAP）用于增强可解释性。在训练过程中，通过贝叶斯优化（Bayesian optimization， BO）客观确定NGBoost的最优超参数。预测这些挠度的概率分布被视为铰链变量。通过应用三标准偏差原则，定义了结构安全区间，以确定需要进一步评估的情况。然后利用基于Naïve贝叶斯理论的任务特定组件来评估结构状况。利用桥梁基准模型验证了在有限训练样本下的安全评价效果。并以某连续箱梁桥为例，进一步验证了所提结构状态指标的有效性。随着结构退化程度的增加，工况指标准确反映了结构退化程度的变化。

{"title":"NGBoost-Naïve Bayes collaborative deep learning for structural safety evaluation of bridges","authors":"Jin-Ling Zheng , Sheng-En Fang","doi":"10.1016/j.ress.2026.112230","DOIUrl":"10.1016/j.ress.2026.112230","url":null,"abstract":"<div><div>In structural safety evaluation, machine learning (ML) based methods often exhibit strong data-fitting capabilities but struggle to effectively handle uncertainties in structural response data. Fortunately, Bayesian deep learning (BDL) algorithms can address this drawback by integrating the Bayesian theory with ML algorithms, thereby unifying perception and inference tasks within a single framework. For this purpose, a BDL framework has been proposed combining natural gradient boosting (NGBoost) and the Naïve Bayes theory. The NGBoost serves as the perception component, capturing correlations between deflections at various measurement locations of a healthy structure, while the Shapley Additive Explanation (SHAP) is employed to enhance interpretability. During the training process, the optimal hyperparameters of the NGBoost is objectively determined through Bayesian optimization (BO). The predicted probability distributions of these deflections are treated as hinge variables. By applying the triple standard deviation principle, a structural safety interval is defined to identify scenarios requiring further evaluation. The task-specific component, based on the Naïve Bayes theory, is then utilized to evaluate the structural condition. A bridge benchmark model was used to verify the safety assessment performance under the limited training samples. In addition, a continuous box-girder bridge was employed to further validate the effectiveness of the proposed structural condition indicator. As the structural degradation increased, the condition indicator accurately reflected the degradation variation.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112230"},"PeriodicalIF":11.0,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981144","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