Probabilistic Engineering Mechanics最新文献

{"title":"An adaptive non-uniform discretization scheme in the frequency–wavenumber spectrum for efficient simulation of random fluctuating wind fields enhanced with NUFFT","authors":"Zheming Lu ,&nbsp;Jun Xu ,&nbsp;Yupeng Song ,&nbsp;Yang Li","doi":"10.1016/j.probengmech.2026.103950","DOIUrl":"10.1016/j.probengmech.2026.103950","url":null,"abstract":"<div><div>The wavenumber–frequency joint spectrum-based spectral representation method is an effective method for simulating multivariate random fluctuating wind fields. However, when using traditional uniform discretization scheme in the frequency and wavenumber domains, the simulation efficiency is unsatisfactory. In this paper, an adaptive wavenumber discretization scheme based on the characteristics of the frequency–wavenumber joint spectrum is proposed, which can obtain non-uniform wavenumber discrete points at all frequencies. In the proposed scheme, wavenumber–frequency joint spectrum is first standardized to obtain the equivalent probability density function. Then, the non-uniform wavenumber discrete points at all frequencies are obtained efficiently according to the equivalent probability distribution function. The non-uniform wavenumbers and wavenumber intervals are integrated into the wavenumber–frequency joint spectrum-based spectral representation method, and combine non-uniform fast fourier transform techniques to generate samples of multivariate random fluctuating wind speed fields. Numerical examples are carried out to verify the effectiveness and adaptability of the proposed scheme. Comparisons with the traditional uniform discretization scheme and other non-uniform sampling schemes indicate that the proposed method is more efficient in the simulation of fluctuating wind speed fields.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103950"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148177926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Static reliability analysis of structures with probability-interval hybrid uncertainties using direct probability integration method and genetic algorithms","authors":"Xiao-Xiao Liu,&nbsp;Dong Cheng,&nbsp;Xi Chen","doi":"10.1016/j.probengmech.2026.103922","DOIUrl":"10.1016/j.probengmech.2026.103922","url":null,"abstract":"<div><div>In modern engineering structural systems of increasing complexity, static reliability analysis considering stochastic and non-probabilistic hybrid uncertainties plays a crucial role in safety assessment. However, for large-scale and highly complex engineering systems, efficient and accurate analytical methods are still required to reliably estimate the upper and lower bounds of structural failure probabilities. This chapter introduces a static probability-interval hybrid reliability analysis method that integrates the Direct Probability Integral Method with a Genetic Algorithm (DPIM–GA) to better address practical engineering requirements. The proposed approach integrates the strengths of the DPIM and GA, enabling the simultaneous consideration of random variables and interval parameters to provide a comprehensive representation of system uncertainty. Employing the GA to conduct an extensive exploration of the solution space allows for the efficient identification of the global optimum or a near-optimal solution of the structural performance function, after which the DPIM is applied to obtain reliable analytical results. Through multiple engineering case studies, the effectiveness and high computational efficiency of the proposed method in handling high-dimensional and complex problems are validated. The results indicate that the DPIM–GA approach achieves satisfactory accuracy and reliability in practical applications, providing a valuable method for hybrid uncertainty reliability analysis of static systems.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103922"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Global sensitivity analysis through advanced ANCOVA decomposition for correlated inputs","authors":"Roman Kalbitz ,&nbsp;Andriy Prots ,&nbsp;Max Dittmann ,&nbsp;Matthias Voigt ,&nbsp;Ronald Mailach","doi":"10.1016/j.probengmech.2026.103915","DOIUrl":"10.1016/j.probengmech.2026.103915","url":null,"abstract":"<div><div>Sensitivity analysis methods have been shown to provide useful information on the influence of the variance of model inputs on the variance of the model response. Several methodologies have been devised to effectively capture these influences, particularly in the context of uncorrelated inputs. However, many real-world applications are hindered by the fact that inputs are correlated, a challenge that traditional methods often fail to address effectively. Consequently, a new methodology is proposed that can differentiate between structural and correlated influences for systems with quadratic main and interaction effects, using an advanced ANCOVA (analysis of covariance) approach. This approach is defined by a detailed analytical decomposition of the covariance matrix provided by the approximated model by means of a high dimensional model representation. This decomposition of the variance- and covariance-equations enables a novel allocation of all the derived terms, thus formulating new sensitivity measures. The validity of the proposed methodology is substantiated by means of a comprehensive numerical investigation, which involves the analysis of multiple test functions.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103915"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration-enhanced active learning method with composite parallel strategy for structural reliability analysis","authors":"He Zhang ,&nbsp;Jie Li ,&nbsp;Zhicheng Zhang ,&nbsp;Fanshu Kong ,&nbsp;Xianglong Zheng ,&nbsp;Fei Wang","doi":"10.1016/j.probengmech.2026.103952","DOIUrl":"10.1016/j.probengmech.2026.103952","url":null,"abstract":"<div><div>Accurate and efficient structural reliability analysis is essential for ensuring the safety, durability, and long-term performance of complex engineering systems throughout service life. However, existing methods often fall short in synergistically optimizing multiple learning criteria, constructing parallel computing schemes, and deeply integrating surrogate models within the analysis framework. To address these gaps, we propose an enhanced active learning reliability analysis approach. A novel composite learning function is developed to integrate multiple learning functions into a unified decision framework. This function improves the informativeness of selected samples. A sorting priority-based parallel computing strategy is then introduced to improve sample selection quality. This strategy selects multiple high-value samples in each iteration, significantly improving computational efficiency. To strengthen surrogate modeling, polynomial chaos–augmented Kriging (PC-Kriging) is employed to enhance global approximation accuracy. We further integrate this surrogate with subset simulation to better capture failure events. Performance evaluation on numerical benchmarks and a real engineering case shows that the proposed method maintains an average error below 1% and achieves an approximately 18-fold improvement in analysis efficiency compared with benchmarks. The proposed method combines high accuracy and efficiency with computational scalability, offering a practical tool for reliability assessment and engineering decision support.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103952"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148001186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Statistical analysis of structural eigenpairs with different non-Gaussian parameters using perturbation-Galerkin method","authors":"Yunchong Chen ,&nbsp;Bin Huang ,&nbsp;Heng Zhang ,&nbsp;Hongyou Cao","doi":"10.1016/j.probengmech.2026.103942","DOIUrl":"10.1016/j.probengmech.2026.103942","url":null,"abstract":"<div><div>The dynamic characteristic analysis is fundamental for evaluating structural safety. However, when structural parameters follow different non-Gaussian distributions, the efficient and accurate calculation of structural eigenpair remains challenging. To address this issue, a highly accurate calculation method is developed to solve the random eigenvalue problem with non-Gaussian parameters based on the stochastic perturbation-Galerkin method. Firstly, the eigenpair, including the eigenvalue and eigenvector, is expressed by the power series expansion with different non-Gaussian parameters, and the initial expansion coefficients are obtained by a high-order perturbation scheme. After redefining the power series expansion, a Galerkin projection is performed to minimize the residual in a statistical sense thereby improving the accuracy of the initial solution. Numerical examples demonstrate that, in the case of different non-Gaussian parameters, compared with the stochastic perturbation method (SPM) and the arbitrary polynomial chaos expansion method (APC), the proposed method has higher accuracy in calculating the first four statistic moments as well as probability density functions (PDF) of eigenvalues, where the PDFs show obvious non-Gaussian characteristics. On the other hand, the means and standard deviations of modal shapes obtained by the proposed method agree better with Monte Carlo simulation method (MCS) than those obtained by SPM.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103942"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147945393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Step-stress likelihood inference for the modified Lindley distribution","authors":"Ayman Baklizi","doi":"10.1016/j.probengmech.2026.103931","DOIUrl":"10.1016/j.probengmech.2026.103931","url":null,"abstract":"<div><div>We develop likelihood-based inference for a two-level step-stress accelerated life test (SSALT) under the Modified Lindley (MoL) lifetime model. The distributional parameter is allowed to change at a pre-specified stress-change time through a log-link, with <span><math><mrow><msub><mrow><mi>θ</mi></mrow><mrow><mn>1</mn></mrow></msub><mo>=</mo><mo>exp</mo><mrow><mo>(</mo><msub><mrow><mi>β</mi></mrow><mrow><mn>0</mn></mrow></msub><mo>)</mo></mrow></mrow></math></span> before the change and <span><math><mrow><msub><mrow><mi>θ</mi></mrow><mrow><mn>2</mn></mrow></msub><mo>=</mo><mo>exp</mo><mrow><mo>(</mo><msub><mrow><mi>β</mi></mrow><mrow><mn>0</mn></mrow></msub><mo>+</mo><msub><mrow><mi>β</mi></mrow><mrow><mn>1</mn></mrow></msub><mo>)</mo></mrow></mrow></math></span> after the change. Under Type-I hybrid censoring, we derive the corresponding log-likelihood, obtain the maximum likelihood estimators, and construct Wald and profile likelihood-ratio confidence intervals for the model parameters and for decision-relevant functionals, including the mean life, quantiles, and mission-time reliability. We also introduce a Monte Carlo calibration-and-design layer for selecting <span><math><mrow><mo>(</mo><mi>n</mi><mo>,</mo><mi>τ</mi><mo>,</mo><msub><mrow><mi>t</mi></mrow><mrow><mi>end</mi></mrow></msub><mo>)</mo></mrow></math></span> to meet operating-risk requirements at two planning points while minimizing an operational objective such as expected test duration or total cost. The proposed framework clarifies the role of the Modified Lindley model in SSALT, gives explicit conditions supporting MLE-based inference, uses Type-I hybrid censoring to control test duration while retaining failure information, and includes a preliminary examination of the observed failure times before the hybrid-censored analysis.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103931"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147713829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prescribed-performance-based adaptive control for Markovian jump stochastic nonlinear systems with actuator backlash-like hysteresis and unmodeled dynamics","authors":"Mohamed Kharrat","doi":"10.1016/j.probengmech.2026.103927","DOIUrl":"10.1016/j.probengmech.2026.103927","url":null,"abstract":"<div><div>This study investigates the problem of adaptive fuzzy control with prescribed performance for a class of stochastic nonlinear systems characterized by Markovian jumps, unmodeled dynamics, and unknown backlash-like hysteresis. The system nonlinearities and the stochastic switching behavior are considered entirely unknown. To address these uncertainties, fuzzy logic systems are employed to approximate the unknown continuous nonlinear functions, while a dynamic auxiliary signal is introduced to mitigate the influence of unmodeled dynamics and facilitate the controller design. By embedding a prescribed-performance framework into the control design, the proposed approach ensures that the tracking error is strictly restricted within a predefined performance bound. Rigorous stochastic stability analysis demonstrates that all closed-loop signals remain bounded in probability and that the tracking error converges while satisfying the specified performance constraints. The effectiveness and practical feasibility of the developed control scheme are further verified through numerical example and an application to a single-link robotic system.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103927"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147713857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive framework for seismic reliability analysis of multi-story structures via modern deep-learning model","authors":"Thanh-Tung Pham,&nbsp;Manh-Tung Vo,&nbsp;Viet-Hung Dang,&nbsp;Truong-Thang Nguyen","doi":"10.1016/j.probengmech.2026.103928","DOIUrl":"10.1016/j.probengmech.2026.103928","url":null,"abstract":"<div><div>This study investigates a novel, adaptive framework utilizing <strong>q</strong>uantile regression for the <strong>D</strong>ynamic <strong>R</strong>eliability <strong>A</strong>nalysis (qDRA) of structures subjected to ground motions, a highly challenging and time-intensive task involving a two-stage computational procedure consisting of dynamic analysis and reliability analysis. The novelty of the proposed method is threefold. First, it designs and implements a hybrid deep learning surrogate model to predict structural dynamic responses given ground motions and random variable inputs. Next, a quantile regression technique is integrated for uncertainty quantification, enabling the selection of relevant candidates to update the model based on estimated uncertainty. In the third step, an adaptive learning process is conducted to progressively improve the prediction model's performance using these selected candidates. It is noteworthy that the prediction model can provide entire time history responses of structures, hence multiple failure criteria could be considered, such as maximum displacement, inter-story drift, and soft safety thresholds with different numbers of exceedances. The method's applicability and efficiency are highlighted via two case studies involving a planar frame with viscous dampers and a 3D three-story building structure with up to 68 random variables. The obtained results reaffirm that the proposed method can provide dynamic reliability results with comparable accuracy to the conventional Monte Carlo simulation, 30 times faster. It also outperforms fourteen other deep learning/machine learning-based counterparts in terms of both accuracy and efficiency.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103928"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147713858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on the random detuning vibration characteristics of the pre-twisted multi -blade disk coupled system based on Gaussian regression surrogate model","authors":"Xinbo He ,&nbsp;Wenkai Feng ,&nbsp;Ju Su ,&nbsp;Dazhi Zhang ,&nbsp;Tianyu Zhao","doi":"10.1016/j.probengmech.2026.103933","DOIUrl":"10.1016/j.probengmech.2026.103933","url":null,"abstract":"<div><div>With the performance upgrade of high-end power equipment such as aero engines and gas turbines, the rotor of the disk-blade coupling system, as a core component, often suffers from blade mistuning in practical applications. This will significantly alter the modal characteristics of the disk-blade coupling system and trigger a series of hazards. To explore the influence of mistuning on the disk-blade coupling system, this paper adopts the component mode synthesis method, combines Kirchhoff plate theory with Lagrange's equation to establish a disk-blade coupling model. The accuracy of the model was verified through commercial finite element analysis and experimental validation. The random detuning of blade stiffness was introduced, and the influence laws of key parameters on detuning vibration characteristics were studied based on machine learning. The research can provide theoretical support and technical reference for the mistuning fault-tolerant design, vibration control and life prediction of disk-blade coupling rotors.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103933"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147748145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Graph-theory-driven configuration synthesis and evidence-based reliability optimization for lightweight morphing wings","authors":"Jianghong Yu ,&nbsp;Renrui Liu ,&nbsp;Jiachang Tang ,&nbsp;Penglin Dong ,&nbsp;Qishui Yao","doi":"10.1016/j.probengmech.2026.103939","DOIUrl":"10.1016/j.probengmech.2026.103939","url":null,"abstract":"<div><div>To address the challenges of high complexity in configuration synthesis of multi-bar mechanisms for morphing wings, the oversight of parameter correlation in reliability optimization, and the difficulty in balancing lightweight design with structural reliability, this paper proposes a graph theory-driven configuration synthesis method for 8-bar morphing wing mechanisms and a reliability optimization approach integrating Copula functions with evidence theory. Based on graph theory, isomorphism discrimination between kinematic chains and topological graphs is implemented, overcoming the challenge of nonlinear polynomial equations encountered in traditional analytical methods for 8-bar mechanisms. A novel morphing wing mechanism configuration is designed to satisfy requirements for large deformation and high stiffness. Local sensitivity analysis is employed to quantify the quantitative relationships between structural dimensions and stiffness, total deformation, and mass, enabling the selection of optimal optimization parameters. The Copula function is introduced into evidence theory to construct a joint basic probability assignment (JBPA) that accounts for structural parameter correlations. Addressing the issue that the independence assumption in evidence theory may lead to significant errors in reliability analysis. Compared with traditional graph theory-based configuration synthesis methods, the proposed method enables systematic screening and innovative design of 8-bar mechanism topologies, eliminating configuration redundancy and motion interference. Unlike conventional structural topology optimization, this method achieves lightweight design under reliability constraints, better aligning with the uncertain operating conditions of practical engineering. Finally, optimization results show that at a 95% target reliability, the structural mass is reduced by 16.43% compared with the original mechanism, with the maximum deformation and equivalent stress both meeting safety limits.</div></div>","PeriodicalId":54583,"journal":{"name":"Probabilistic Engineering Mechanics","volume":"84 ","pages":"Article 103939"},"PeriodicalIF":3.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147858245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}