Reliability Engineering & System Safety最新文献_第5页

Time-dependent fragility analysis and risk assessment of nuclear containment structure subjected to internal pressure considering chloride-induced corrosion of reinforcing steel 考虑钢筋氯化物腐蚀的核安全壳结构内压脆弱性时变分析与风险评估

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-18 DOI: 10.1016/j.ress.2025.111736

Song Jin , Shanjie Xu , Tong Shen , Hao Wang , Xuyang Cao , Hongbin Jia

{"title":"Time-dependent fragility analysis and risk assessment of nuclear containment structure subjected to internal pressure considering chloride-induced corrosion of reinforcing steel","authors":"Song Jin , Shanjie Xu , Tong Shen , Hao Wang , Xuyang Cao , Hongbin Jia","doi":"10.1016/j.ress.2025.111736","DOIUrl":"10.1016/j.ress.2025.111736","url":null,"abstract":"<div><div>This study carried out time-dependent fragility analysis and risk assessment of nuclear containment structure subjected to internal pressure based on Kernel Density Estimation (KDE). The accuracy and efficiency of the proposed methodology were demonstrated by fragility analysis of ring-stiffened cylinder structure subjected to hydrostatic pressure. Then, the proposed methodology is applied to time-dependent fragility analysis and risk evaluation of the nuclear containment structure during the service life. Detailed time-dependent chloride-induced corrosion model considering uncertainties involved in environment and material properties is formulated. Latin Hypercube Sampling (LHS) is employed to generate stochastic samples of nuclear containment structure, and nonlinear finite element simulation of nuclear containment structure for different service time is performed. The proposed KDE methodology is adopted to analyze time-dependent fragility, with the Cumulative Conditional Failure Probability (CCFP) serving as the metric to quantify probabilistic safety performance of nuclear containment structure for different service time. Results indicated that the KDE approach provides higher accuracy in fitting fragility curves compared to conventional parametric method. Nuclear containment structure in this study meets probabilistic safety margins during long-term service conditions.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111736"},"PeriodicalIF":11.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120907","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

ESFDA: Energy-based Source-Free domain adaptation method for rotating machinery fault diagnosis 基于能量的旋转机械故障诊断无源域自适应方法

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-18 DOI: 10.1016/j.ress.2025.111708

Ke Yan , Yuzhen Ma , Changfu He , Mei Hua , Xin Li

{"title":"ESFDA: Energy-based Source-Free domain adaptation method for rotating machinery fault diagnosis","authors":"Ke Yan , Yuzhen Ma , Changfu He , Mei Hua , Xin Li","doi":"10.1016/j.ress.2025.111708","DOIUrl":"10.1016/j.ress.2025.111708","url":null,"abstract":"<div><div>The source-free unsupervised domain adaptation diagnosis (SF-UDAD) of rotating machinery aims at transferring diagnostic knowledge from a well-labeled source domain to an unlabeled target domain without accessing the source domain data. This approach excels over conventional domain adaptation techniques in safeguarding data privacy and reducing data transmission and storage costs. Nevertheless, existing methods cannot address this distribution shift between domains due to their lack of connection with marginal distributions, resulting in unstable adaptation. This paper explores SF-UDAD from an energy perspective and introduces energy-based source-free domain adaptation (ESFDA). Specifically, during the pretraining phase, an attention mechanism is integrated to enhance the model’s capability in capturing domain-invariant features. In the following adaptation phase, the deep learning model is converted to an energy-based model (EBM), and the Langevin Dynamics is employed to construct contrastive divergence to optimize the EBM. To enhance the robustness of the EBM, high-confidence predictions of the target domain are incorporated as additional information to facilitate the model’s adaptation. The stability and effectiveness of the proposed method are demonstrated through experiments on three opensource datasets.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111708"},"PeriodicalIF":11.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145096668","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

Post-risk assessment model for gas explosion accidents based on the coupling effect of disaster-causing factors 基于致灾因素耦合效应的瓦斯爆炸事故后风险评价模型

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-18 DOI: 10.1016/j.ress.2025.111733

Feng Li , Baoyan Duan , Yue Zhang , Dongdong Liang

{"title":"Post-risk assessment model for gas explosion accidents based on the coupling effect of disaster-causing factors","authors":"Feng Li , Baoyan Duan , Yue Zhang , Dongdong Liang","doi":"10.1016/j.ress.2025.111733","DOIUrl":"10.1016/j.ress.2025.111733","url":null,"abstract":"<div><div>The coal mining industry—a typical high-risk industry—involves multiple positions, processes, and disasters. In particular, gas explosion accidents are susceptible to the domino effect due to the strong explosion shock wave generated, which can severely threaten the safety of personnel. Therefore, the objective of this study is to propose an comprehensive and effective method to assess the post-risk and provide help for the risk management and control. First, based on the cognitive reliability and error analysis method<span><span><sup>1</sup></span></span> and Offshore Reliability Data Handbook,<span><span><sup>2</sup></span></span> Bayesian network<span><span><sup>3</sup></span></span> technology is employed to determine the risk value of each disaster-causing factor (DCF) .<span><span><sup>4</sup></span></span> Subsequently, the DCFs involved in each position are integrated, and a superimposed risk model is established. Finally, based on ArcGIS software, the spatial distribution of risk is simulated, and the risk level is determined. The results show that: (1) The posterior probability of most DCFs has an order of magnitude increase, and the nodes with higher sensitivity are primarily related to ventilation being blocked, methods of stopping and starting local fans, and blasting techniques. (2) The risk hierarchy changes after considering the risk superimposed effect. The first-level risk posts are mainly coal miners and mining electrical workers. (3) The risk distribution map shows that the key control areas of coal mine gas explosion accidents are the mining face and tunneling face, power distribution room, and pyrotechnic storehouse.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111733"},"PeriodicalIF":11.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109303","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

Multidimensional dynamic disturbance factor identification framework for maritime networks of strategic materials 战略物资海上网络多维动态干扰因子识别框架

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-17 DOI: 10.1016/j.ress.2025.111725

Jianing Yu , Zhixiang Fang , Hengzu Liu

{"title":"Multidimensional dynamic disturbance factor identification framework for maritime networks of strategic materials","authors":"Jianing Yu , Zhixiang Fang , Hengzu Liu","doi":"10.1016/j.ress.2025.111725","DOIUrl":"10.1016/j.ress.2025.111725","url":null,"abstract":"<div><div>Maritime networks of strategic materials (i.e., iron ore, crude oil, coal, liquefied natural gas, liquefied petroleum gas, and soybeans) face multiple external disturbances in fluctuating international economic and geopolitical environments. This study is set against the backdrop of major events (the international crude oil price plunge, the initiation of Brexit, the COVID-19 pandemic, and the Russia-Ukraine conflict), and presents a multidimensional, dynamic disturbance factor combination identification framework for maritime networks of strategic materials under several major events, which extends the Hidden Markov Model and stepwise optimization strategies to identify the disturbance factor combinations faced by different countries and various strategic materials. The results show that under different major event scenarios, disturbance factors exhibit distinct combinations and spatiotemporal distributions on strategic materials. This study reveals maritime systems' complex challenges and response opportunities under multidimensional external disturbances. It provides a reference for further risk management, capacity scheduling, and policy formulation.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111725"},"PeriodicalIF":11.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109302","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

Fragility assessment of FRP-strengthened RC beams with chloride-induced corrosion damage under impact loads 冲击载荷下氯化物腐蚀损伤frp加固RC梁易损性评价

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-17 DOI: 10.1016/j.ress.2025.111728

Yu Liu , Yifei Hao , Hong Hao , Yun Zhou

{"title":"Fragility assessment of FRP-strengthened RC beams with chloride-induced corrosion damage under impact loads","authors":"Yu Liu , Yifei Hao , Hong Hao , Yun Zhou","doi":"10.1016/j.ress.2025.111728","DOIUrl":"10.1016/j.ress.2025.111728","url":null,"abstract":"<div><div>Chloride-induced corrosion threatens the durability of reinforced concrete (RC) structures, posing substantial challenges from both engineering and maintenance perspectives throughout the lifecycle. This study aims to assess the fragility of RC beams with chloride-induced corrosion damage with or without CFRP strengthening under varying impact intensities. The impact mass-velocity diagram was developed using a validated high-fidelity physics-based finite element model. The effectiveness of fibre-reinforced polymer (FRP) strengthening of the corrosion-damaged beams on the impact resistance was examined. The probabilistic fragility assessment of corroded beams under impact load was carried out by using the Monte Carlo simulation. A suitable strengthening decision on the impact resistance of corroded beams was then proposed. The numerical results indicate that the probability of impact damage in RC beams increases prominently with the advancement of corrosion deterioration. The application of FRP strengthening significantly reduces the probability of impact damage in corroded beams. Among the methods studied, the monolithic FRP strengthening approach proves to be more effective than the segmental method in mitigating damage and enhancing impact resistance. For instance, the impact damage probabilities of corroded beams are reduced from 0.87 to 0.38 and 0.17, respectively, by segmental and monolithic FRP strengthening when the corrosion degree is 29%.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111728"},"PeriodicalIF":11.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145096771","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 dynamic reliability assessment method for multi-state manufacturing system by merging imprecise observational information 一种融合不精确观测信息的多状态制造系统动态可靠性评估方法

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-17 DOI: 10.1016/j.ress.2025.111722

Tudi Huang , Qin Zhang , Michael Beer , Yu Liu , Hong-Zhong Huang

{"title":"A dynamic reliability assessment method for multi-state manufacturing system by merging imprecise observational information","authors":"Tudi Huang , Qin Zhang , Michael Beer , Yu Liu , Hong-Zhong Huang","doi":"10.1016/j.ress.2025.111722","DOIUrl":"10.1016/j.ress.2025.111722","url":null,"abstract":"<div><div>Accurate reliability assessment of advanced manufacturing systems is essential for ensuring production efficiency, reducing downtime, and enabling intelligent maintenance strategies. In practical industrial environments, state observations obtained from sensors or expert evaluations are often imprecise. Effectively utilizing this uncertain information can substantially improve the precision of reliability evaluations. Conventional methodologies often encounter limitations in addressing this challenge, as manufacturing systems are generally characterized by networked production line configurations rather than traditional serial or parallel structures. Moreover, the effective integration of imprecise observational data is essential for the continuous updating of system reliability. This study introduces a novel approach for dynamic reliability evaluation of a multi-state manufacturing system (MSMS), incorporating both rework mechanisms and buffer elements to enhance the accuracy and applicability of system reliability assessments. The MSMS model can effectively depict the gradual degradation processes and diverse performance levels of manufacturing systems, allowing for a more realistic and detailed representation of system behavior over time compared to traditional binary-state models. This study employs the multistate flow network (MFN) model to construct the MSMS reliability assessment framework from a network structure perspective. Dynamic Bayesian networks (DBNs) are developed to update the reliability function of an individual MSMS by incorporating evidential observational data. An illustrative case study on the reliability update of an aluminum alloy wheel production line is presented to demonstrate the proposed methodology. The case study results further confirm the effectiveness of the approach.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111722"},"PeriodicalIF":11.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120909","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

Dual-channel dynamic spline graph convolutional network for bearing remaining useful life prediction 双通道动态样条图卷积网络用于轴承剩余使用寿命预测

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-17 DOI: 10.1016/j.ress.2025.111731

Yubei Jin , Dongdong Liu , Yongchang Xiao , Lingli Cui

{"title":"Dual-channel dynamic spline graph convolutional network for bearing remaining useful life prediction","authors":"Yubei Jin , Dongdong Liu , Yongchang Xiao , Lingli Cui","doi":"10.1016/j.ress.2025.111731","DOIUrl":"10.1016/j.ress.2025.111731","url":null,"abstract":"<div><div>Accurate prediction of the remaining useful life (RUL) of bearings is crucial for predictive maintenance in industrial systems; however, traditional methods frequently fail to account for dynamic topological changes and nonlinear degradation patterns within vibration data. To address these challenges, we propose the Dual-Channel Dynamic Spline Graph Convolutional Network (DDSGNet), different from traditional methods that rely on discrete feature aggregation, this paper models the spatial correlation among vibration features using a global topology aggregation module. The continuous local evolution operator captures stationary degraded dynamic features, whereas the time-dependent learner maintains long-range sequence information. This approach addresses the challenge of modeling continuous local feature changes. Furthermore, a novel activation function, DSAF, is proposed to dynamically adjust to nonlinear signal changes by smoothing gradients, thereby resolving issues of gradient roughness and vanishing. Lastly, a loss function named PhyMAE, which is based on physical constraints, is proposed to align with the physical characteristics of bearing degradation, ensuring accurate and physically consistent RUL prediction. Experiments on two public datasets demonstrate that DDSGNet outperforms state-of-the-art methods in prediction accuracy, offering a robust solution for bearing RUL estimation.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111731"},"PeriodicalIF":11.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109304","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

Revealing higher-order interactions through multimodal irreversibility in flood-affected transportation networks 揭示洪水影响交通网络中多式联运不可逆性的高阶相互作用

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-16 DOI: 10.1016/j.ress.2025.111726

Xuhui Lin , Long Chen , Qiuchen Lu , Pengjun Zhao , Tao Cheng

{"title":"Revealing higher-order interactions through multimodal irreversibility in flood-affected transportation networks","authors":"Xuhui Lin , Long Chen , Qiuchen Lu , Pengjun Zhao , Tao Cheng","doi":"10.1016/j.ress.2025.111726","DOIUrl":"10.1016/j.ress.2025.111726","url":null,"abstract":"<div><div>Climate change and extreme weather events increasingly threaten urban transportation systems, challenging their ability to maintain essential mobility services. Current analytical approaches primarily focus on individual modes or simplified interactions, failing to capture the complex, non-equilibrium dynamics that emerge when multiple transportation modes interact under stress. This research introduces a novel Multi-modal Visibility Graph Irreversibility (MmVGI) framework for analysing transportation system behaviour during extreme weather events. By integrating concepts from non-equilibrium dynamics with visibility graph analysis, our approach quantifies complex interactions between different transportation modes and reveals the underlying mechanisms driving system non-equilibrium characteristics. Through a case study in the City of London during an extreme rainfall event, we demonstrate that transportation system adaptation exhibits clear hierarchical patterns across different road types. While primary roads maintain stable dynamics dominated by motorised transport, secondary networks show complex patterns of modal interaction, with cycling emerging as a crucial component in system adaptation. The strong correlation between unique and combined irreversibility measurements provides evidence for genuine higher-order interactions that cannot be reduced to simpler modal combinations. These findings advance both theoretical understanding of urban system dynamics and practical approaches to transportation management, offering valuable insights for urban planners and policymakers in developing more resilient, adaptive transportation systems for future climate challenges.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111726"},"PeriodicalIF":11.0,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145096662","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

Fatigue life prediction and uncertainty quantification of aerospace metals: A Bayesian physics-informed neural network model 航空航天金属疲劳寿命预测与不确定性量化：贝叶斯物理信息神经网络模型

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-15 DOI: 10.1016/j.ress.2025.111724

Qianling Wang , Guowen Yao , Mingxun Hou

{"title":"Fatigue life prediction and uncertainty quantification of aerospace metals: A Bayesian physics-informed neural network model","authors":"Qianling Wang , Guowen Yao , Mingxun Hou","doi":"10.1016/j.ress.2025.111724","DOIUrl":"10.1016/j.ress.2025.111724","url":null,"abstract":"<div><div>Aerospace metals exhibit significant variability in fatigue life due to inherent material heterogeneity and manufacturing-induced defects, posing major challenges to the reliability of aircraft structures. Traditional data-driven approaches, such as probabilistic neural networks (PNNs), often rely on predefined prior distributions (e.g., normal distribution) for fatigue life prediction. However, these methods generally fail to capture the true distributional characteristics and lack physical interpretability. To address these limitations, this study proposes a Bayesian Physics-Informed Neural Network (BPINN) framework that integrates physical constraints into a Bayesian neural architecture for simultaneous fatigue life prediction and uncertainty quantification. The model leverages variational inference to transform prior assumptions into data-informed posterior distributions, thereby enhancing both predictive accuracy and physical consistency. Extensive evaluations demonstrate that BPINN not only outperforms conventional neural networks in point-wise prediction accuracy, but also produces more reliable confidence intervals than PNNs. Importantly, the framework exhibits robust generalization and stable performance across multiple external datasets involving different material systems and physical priors. These results underscore the flexibility and reliability of BPINN, and further reveal that fatigue life distributions may deviate significantly from normality—highlighting the critical importance of physically grounded and distribution-aware uncertainty quantification.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111724"},"PeriodicalIF":11.0,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145096709","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

Probably approximately correct-Bayesian Criterion (PBC) in Wiener processes selection 维纳过程选择中的近似正确贝叶斯准则（PBC）

IF 11 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-09-15 DOI: 10.1016/j.ress.2025.111609

Bingxin Yan , Xiaobing Ma , Han Wang

{"title":"Probably approximately correct-Bayesian Criterion (PBC) in Wiener processes selection","authors":"Bingxin Yan , Xiaobing Ma , Han Wang","doi":"10.1016/j.ress.2025.111609","DOIUrl":"10.1016/j.ress.2025.111609","url":null,"abstract":"<div><div>In recent decades, various models have been proposed for specific degradation scenarios, using benchmark models like the Wiener process and general path models. However, model evaluation and selection often rely on available degradation data from tests, limiting the assessment of performance on unseen data. This scenario could be encountered when we have a limited sample size in the degradation test and are concerned about the model’s performance on unseen data. It could also arise when we only have degradation data from a similar product but need to assess the performance of the model on a new product lacking degradation data. To this end, we propose a probably approximately correct-Bayesian criterion (PBC) to select the model with the best generalization performance. The proposed PBC can set a theoretical bound for the generalization error of a model on unseen data and offer a closed-form upper bound when the degradation model is a Wiener process. The proposed criterion shows potential for extending to other degradation models, such as the general path model. Comprehensive experiments and case studies further illustrate the effectiveness of the PBC model selection criterion over the existing Akaike information criterion (AIC) and Bayesian information criterion (BIC).</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"266 ","pages":"Article 111609"},"PeriodicalIF":11.0,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145096711","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