Reliability Engineering & System Safety最新文献

{"title":"Critical node identification of dynamic-load wireless sensor networks for cascading failure protection","authors":"Yifan Yuan ,&nbsp;Xiaohong Shen ,&nbsp;Lin Sun ,&nbsp;Yongsheng Yan ,&nbsp;Haiyan Wang","doi":"10.1016/j.ress.2025.111351","DOIUrl":"10.1016/j.ress.2025.111351","url":null,"abstract":"<div><div>Wireless Sensor Networks (WSNs), as complex and dynamic systems, are highly susceptible to cascading failures. To enhance network resilience, this study addresses the identification of critical nodes that drive failure propagation. Unlike prior studies that often ignore the impact of varying network load, we highlight that node importance can change significantly under dynamic load conditions. To tackle this, we introduce a method for identifying critical nodes in dynamic-load WSNs. We first construct a cascading failure model that links network load with link capacity, analyzing how fluctuations in load affect failure propagation. Building on this model, we propose an EW-TOPSIS-based node evaluation method grounded in node deletion, where the influence of each node under different load conditions is considered as distinct evaluation criteria. To verify the proposed method, we conduct simulations of low-rate underwater WSNs in ns-3 under dynamic load conditions. Results show that, as an attack node selection strategy, our method achieves up to 30% and 25% greater degradation in failure severity and PDR, respectively, across varying network topology, densities and traffic conditions, compared to five baseline techniques. This work provides insights for designing effective mitigation strategies against cascading failures in resource-constrained networks.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111351"},"PeriodicalIF":9.4,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144365419","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}

{"title":"Bayesian Optimized Deep Ensemble for Uncertainty Quantification of Deep Neural Networks: a System Safety Case Study on Sodium Fast Reactor Thermal Stratification Modeling","authors":"Zaid Abulawi ,&nbsp;Rui Hu ,&nbsp;Prasanna Balaprakash ,&nbsp;Yang Liu","doi":"10.1016/j.ress.2025.111353","DOIUrl":"10.1016/j.ress.2025.111353","url":null,"abstract":"<div><div>Deep neural networks (DNNs) are increasingly important to scientific computing and engineering system simulations. Accurate uncertainty quantification (UQ) for DNNs is critical in safety-sensitive engineering domains. Traditional Deep Ensemble (DE) methods, while easy to implement, frequently suffer from poorly calibrated uncertainty estimates and limited predictive accuracy due to reliance on fixed architectures with varied weight initializations. To address these issues, we introduce a workflow that combines Bayesian Optimization (BO) and DE. The workflow is modular, scalable, and integrates parallel BO initialized with Sobol sequences to individually optimize the hyperparameters of each ensemble member. This method enhances ensemble diversity, improves predictive accuracy, and provides reliable uncertainty estimates.</div><div>We evaluate the proposed BODE approach in a sodium fast reactor thermal stratification modeling case study, where we used a densely connected convolutional neural network to predict turbulent viscosity during the reactor transient with consideration of data noise. We benchmark its performance against several optimization approaches, including baseline deep ensemble, evolutionary algorithm-optimized ensemble, ensemble formed via random search combined with greedy selection, and a BO ensemble using random initialization. Our results demonstrate superior performance of the developed BODE approach. In noise-free scenarios, BODE notably reduces incorrect aleatoric uncertainty and significantly enhances predictive accuracy. Under conditions of 5% and 10% Gaussian noise, BODE adaptively quantifies uncertainty proportional to data noise, achieving up to an 80% reduction in root mean square error compared to baseline methods and producing well-calibrated prediction intervals.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111353"},"PeriodicalIF":9.4,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144339016","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}

{"title":"Frequency-domain analysis and dynamic reliability assessment of random vibration for non-classically damped linear structure under non-Gaussian random excitations","authors":"Xiangqian Sheng ,&nbsp;Kuahai Yu ,&nbsp;Wenliang Fan ,&nbsp;Shihong Xin","doi":"10.1016/j.ress.2025.111363","DOIUrl":"10.1016/j.ress.2025.111363","url":null,"abstract":"<div><div>Frequency domain analysis is the important component in the random vibration analysis. However, frequency domain analysis for the non-classically damped linear structure under non-Gaussian random excitations remains a challenge. Thus, this paper establishes a unified computational framework of higher-order moment spectra of response, and performs reliability assessment based on moment spectra of response. Firstly, the theoretical expressions of the higher-order moment spectra of response are deduced by the complex mode superposition method and the generalized impulse response function. Secondly, the expressions of the higher-order moment spectra of response are reconstructed with the help of responses for the harmonic excitation. Subsequently, the dynamic reliability is estimated based on the approximation joint probability density function which is constructed through the unified Hermite polynomial model and Gaussian Copula function. Finally, two numerical examples are investigated to verify the accuracy and efficiency of the calculation method of response the higher-order moment spectra and the dynamic reliability.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111363"},"PeriodicalIF":9.4,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144313179","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}

{"title":"Measured displacement data-driven efficient interpretation and real-time risk assessment method for the service performance of arch dams with cracks","authors":"Bo Xu ,&nbsp;Zeyuan Chen ,&nbsp;Huaizhi Su ,&nbsp;Hu Zhang ,&nbsp;Pengcheng Xu ,&nbsp;Shaowei Wang","doi":"10.1016/j.ress.2025.111377","DOIUrl":"10.1016/j.ress.2025.111377","url":null,"abstract":"<div><div>Current methods for predicting arch dam displacement rarely consider the impact of cracks on displacement and the interpretability of factors, nor do they reasonably assess the risk probability of arch dam. To address these issues, first, clustering partitions are conducted based on the Ward criterion, and the comprehensive displacement is obtained through the Criteria Importance Through Intercriteria Correlation (CRITIC) method. Secondly, considering the impact of cracks, a displacement monitoring model HSCT is constructed, and feature selection for the HSCT model factors is performed using the Max-Relevance and Min-Redundancy (mRMR), while Kernel Principal Component Analysis (KPCA) is utilized for feature extraction of crack factors. Furthermore, to enhance interpretability, an attention mechanism is incorporated into the Convolutional Neural Network with Long Short-Term Memory (CNN-LSTM) model, establishing a CNN-LSTM-Attention model to predict comprehensive displacement and visualize the importance of influencing factors. Finally, Kernel Density Estimation (KDE) is applied to the residuals of the comprehensive displacement, and a multivariate Copula function is used to construct the joint distribution to calculate the overall risk rate. The results indicate that the proposed methods and models are reasonable and feasible, providing scientific basis and technical support for the health diagnosis of hydraulic structures.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111377"},"PeriodicalIF":9.4,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144338984","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}

{"title":"A novel hybrid deep learning methodology for real-time wellhead pressure forecasting and risk warning during shale gas hydraulic fracturing","authors":"Liangjie Gou ,&nbsp;Zhaozhong Yang ,&nbsp;Chao Min ,&nbsp;Duo Yi ,&nbsp;Liangping Yi ,&nbsp;Xiaogang Li","doi":"10.1016/j.ress.2025.111309","DOIUrl":"10.1016/j.ress.2025.111309","url":null,"abstract":"<div><div>Accurate real-time forecasting of wellhead pressure significantly impacts risk warning and optimization of fracturing parameters. However, the complexity and non-stationary of data limit the accuracy of traditional deep learning (DL). We propose a novel hybrid DL method to enhance risk warning capabilities. The proposed method integrates the complex forecasting process into four modules. Firstly, the VMD-Fuzzy entropy module classifies intrinsic mode functions (IMFs) obtained from variational mode decomposition to significantly reduce feature redundancy. Then the Attention-GNN automatically learns latent features between multiple variables to automatically update the graph structure and incorporate controllable future input features. Additionally, the temporal–spatial feature extraction module captures spatial and temporal correlations to improve accuracy. The uncertainty quantification module employs a backtrack loss function and multi-head attention to enhance the capturing capability for critical data features. The method is verified using fracturing data from a shale gas block in Sichuan, China. The average root mean square error (RMSE), average maximum allowable error (MAE) and average R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> of the target area are 1.31 (MPa), 1.27 (MPa) and 0.94, respectively, which are significantly better than the traditional DL. In addition, the data of 4 overpressure well stages were used for example verification, and the corresponding traffic light risk warning system was developed. The verification results prove that the proposed method can effectively improve the warning timeliness, and provide an effective technical way to achieve intelligent and efficient hydraulic fracturing.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111309"},"PeriodicalIF":9.4,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144329535","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}

{"title":"Dynamic vulnerability analysis of multi-modal public transport network using generalized travel costs from a multi-layer perspective","authors":"Ziqi Wang ,&nbsp;Yulong Pei ,&nbsp;Jianhua Zhang ,&nbsp;Zhixiang Gao","doi":"10.1016/j.ress.2025.111375","DOIUrl":"10.1016/j.ress.2025.111375","url":null,"abstract":"<div><div>The proper functioning of urban multi-modal public transport networks (MPTNs) is essential for sustainable urban development. However, as these networks become increasingly complex, their dynamic vulnerability to disturbances also rises. This study proposes a cascading failure model based on localized dynamic flow redistribution, aimed at mitigating and controlling the dynamic vulnerability of MPTNs. Firstly, we construct a multi-layered MPTN weighted by both generalized cost and traffic flow attributes, considering the heterogeneity and interdependence among various routes and modes. Building on this structure, we develop a localized user equilibrium traffic redistribution model that accounts for passenger congestion effects, enabling the analysis of dynamic vulnerability from both structural and functional perspectives. The proposed methodology is applied to a case study of the MPTN in Harbin. Simulation results reveal that the propagation of cascading failures in MPTNs is strongly associated with the geographical locations and importance of stations. Increasing station capacity effectively reduces the scale of cascading failure propagation, thereby alleviating network vulnerability. Moreover, dynamic vulnerability analysis shows that network connectivity and generalized travel efficiency deteriorate nonlinearly over time. Failures at critical stations disproportionately accelerate the dynamic vulnerability evolution, leading to nonlinear and compound degradation of network performance, including connectivity loss, increased travel costs, and service efficiency deterioration. This study provides valuable insights for enhancing the resilience of MPTNs, particularly in complex urban environments.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111375"},"PeriodicalIF":9.4,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144339017","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}

{"title":"Operational readiness-oriented condition-based maintenance and spare parts optimization for multi-state systems","authors":"Shihan Tan ,&nbsp;Qiwei Hu ,&nbsp;Chiming Guo ,&nbsp;Dunxiang Zhu ,&nbsp;Enzhi Dong ,&nbsp;Fang Zhang","doi":"10.1016/j.ress.2025.111367","DOIUrl":"10.1016/j.ress.2025.111367","url":null,"abstract":"<div><div>In many military scenarios, engineered systems are required to remain satisfied with operational readiness to respond to unexpected tasks. However, the degradation caused by daily usage inherently decreases the operational readiness of these systems. Condition-based maintenance is an efficient strategy that can recover the system operational readiness by restoring the system condition. On the other hand, the activities of maintenance are often constrained by spare parts ordering. Most existing research only pays attention on the daily work and ignores the requirement of operational readiness. In this paper, a novel reinforcement learning (RL) based condition-based maintenance and spare parts optimization method for multiple unit multi-state systems (MSS) is proposed, aimed at minimizing long-term cost rate considering the requirement of operational readiness and daily work. The resulting joint decision-making problem is formulated as a discrete-time discrete-state Markov decision process (MDP) and a customized architecture of value iteration algorithm embedded with a stratified sampling Monte Carlo (SSMC) method is introduced. A real case of armored vehicles in a military base is provided to prove the effectiveness of our method. From comparative experiments and sensitivity analysis of serval examples, several interesting suggestions are presented.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111367"},"PeriodicalIF":9.4,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144365992","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}

{"title":"Development and validation of a novel method to predict flame behavior in tank fires based on CFD modeling and machine learning","authors":"Zhenqi Hu ,&nbsp;Jinlong Zhao ,&nbsp;Shaohua Zhang ,&nbsp;Hanchao Ma ,&nbsp;Jianping Zhang","doi":"10.1016/j.ress.2025.111368","DOIUrl":"10.1016/j.ress.2025.111368","url":null,"abstract":"<div><div>Ensuring storage tank farm safety involves systematic engineering. Tank fire with a large ullage height is a common type of accident and poses a serious threat to tank farms due to the air restrictions by ullage height. This study investigates the impact of ullage height on flame morphology, air entrainment, and burning behaviors through experiments and computational fluid dynamics (CFD) simulations. Results showed that ullage height of the tank significantly affect burning rate, flame morphology and air entrainment. Three burning regimes were identified as ullage height changes. Experimental and simulation data were then used in a machine learning (ML) model, which combines particle swarm optimization (PSO) and back-propagation neural networks (BPNN) to predict the mass burning rate and internal flow field. The input datasets included the tank diameter, ullage height, experimental mass burning rate, and the internal flow field predicted by the CFD model. The predicted results by the ML model agree well with the experimental and numerical data. It was shown that the larger number of the training datasets, the more accurate predictions. The new model provides a fast and efficient way to predict the burning behaviors and supports risk assessment for tank fire accidents with limited experimental and numerical inputs.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111368"},"PeriodicalIF":9.4,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144329531","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}

{"title":"A new random vibration response analysis method for laminates: Geometric nonlinearity and uncertainty are both involved for higher consistency with reality","authors":"Yuan Liu ,&nbsp;Xuan Zhang ,&nbsp;Xibin Cao ,&nbsp;Jinsheng Guo ,&nbsp;Zhongxi Shao ,&nbsp;Qingyang Deng ,&nbsp;Pengbo Fu ,&nbsp;Yaodong Hou","doi":"10.1016/j.ress.2025.111343","DOIUrl":"10.1016/j.ress.2025.111343","url":null,"abstract":"<div><div>To enable high-precision analysis and reliable design of laminates, a novel method is proposed for solving the random vibration response while accounting for geometric nonlinearity and structural uncertainty. The power spectral densities (PSDs) of deflection, velocity, and acceleration for a SSSS plate (indicating that all edges are simply supported) are derived using statistical linearization. In particular, the number of unknowns in the displacement field model of an SSSS-2 plate (where SSSS-2 denotes a simply supported plate with a free mid-surface) is reduced from five to three compared to conventional algorithms. This simplification reduces the complexity of the nonlinear equations and significantly improves computational efficiency. Furthermore, a novel framework was proposed, featuring a multiscale feature extraction, fusion, and learning network (MFEFLN). This network consists of three multiscale feature extraction blocks, one multiscale feature concatenation block, and one high-level feature fusion block. A dedicated network system was developed to analyze the influence of the mean values and tolerance zones of uncertain structural parameters on the random vibration responses. When predicting the same number of random response PSDs, the MFEFLN-based procedure demonstrates greater efficiency than direct Monte Carlo simulation (MCS) and superior accuracy compared to BP, GAN, LSTM, 2D CNN, and ADCNN methods. This research is beneficial for the design optimization and reliability guarantee of laminated structures by providing high-precision analysis results of the dynamic performance by covering the geometric nonlinearity and uncertainty that exist in actual products.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111343"},"PeriodicalIF":9.4,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144313173","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}

{"title":"Safety assurance of Machine Learning for autonomous systems","authors":"Colin Paterson,&nbsp;Richard Hawkins,&nbsp;Chiara Picardi,&nbsp;Yan Jia,&nbsp;Radu Calinescu,&nbsp;Ibrahim Habli","doi":"10.1016/j.ress.2025.111311","DOIUrl":"10.1016/j.ress.2025.111311","url":null,"abstract":"<div><div>Machine Learning (ML) components are increasingly incorporated into systems, with different degrees of autonomy, where model performance is reported as meeting, or exceeding, the capabilities of human experts. This promises to transform products and services, in diverse domains such as healthcare, transport and manufacturing, to better serve underrepresented groups, reduce costs, and increase delivery effectiveness, especially where expert resources are scarce. The greatest potential for transformative impact lies in the development of autonomous systems for safety-critical applications where their acceptance, and subsequent deployment, is reliant on establishing justified confidence in system safety. Creating a compelling safety case for ML is challenging however, particularly since the ML development lifecycle is significantly different to that employed for traditional software systems. Typically ML development involves replacing detailed software specifications with representative data sets from which models of behaviour is learnt. Indeed, ML’s strength lies in tackling problems which are challenging for traditional software development practices. This shift in development practices introduces challenges to established assurance processes which are crucial to developing the compelling safety case required for regulation and societal acceptance. In this paper we introduce the first methodology for the Assurance of Machine Learning for use in Autonomous Systems (AMLAS). The AMLAS process describes how to systematically and attractively integrate safety assurance into the development of ML components and how to generate the evidence base for explicitly justifying the acceptable safety of these components when integrated into autonomous system applications. We describe the use of AMLAS by considering how a safety case may be constructed for an object detector for use in the perception pipeline of an autonomous driving application. We further discuss how AMLAS has been applied in several domains including healthcare, automotive and aerospace as well as supporting policy and industry guidance for defence, healthcare and automotive.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"264 ","pages":"Article 111311"},"PeriodicalIF":9.4,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144335663","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}