A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Probabilistic sensitivity matrices under stressor uncertainty
https://orcid.org/0000-0001-6772-1468
Abstract Quantitative risk analysis critically relies on precise numerical simulations to furnish optimal estimations. These numerical models encompass complex constitutive material correlations, multi-fidelity assumptions within model generation, and the encompassing variability of applied loading conditions. Traditional sensitivity analysis methodologies, being contingent upon the specific stressors applied, bear the potential to yield deceptive deductions or flawed prioritization strategies. In this paper, we introduce an extensive array of probabilistic sensitivity analysis techniques, coupled with post-processing of their outputs to rigorously quantify the inherent uncertainties. To establish a cohesive framework, we present an integrative formulation, subsequently applied to a reinforced concrete bridge component. Additionally, we present a pioneering technique, involving intensifying artificial acceleration in conjunction with the sensitivity matrix concept, ushering in a novel path-dependent sensitivity metric. Comparative analyses against other probabilistic sensitivity methodologies underscore the distinctive attributes of our proposed approach, spanning accuracy, computational efficiency, comprehensiveness, and result stability. Our proposed technique exhibits remarkable efficiency, culminating in a computational time reduction of approximately 90%.
Highlights A path-dependent sensitivity matrix is introduced in seismic analysis. The results of the sensitivity assessment depend on the stressor type. The importance of material randomness is altered with the loading intensity.
Probabilistic sensitivity matrices under stressor uncertainty
https://orcid.org/0000-0001-6772-1468
Abstract Quantitative risk analysis critically relies on precise numerical simulations to furnish optimal estimations. These numerical models encompass complex constitutive material correlations, multi-fidelity assumptions within model generation, and the encompassing variability of applied loading conditions. Traditional sensitivity analysis methodologies, being contingent upon the specific stressors applied, bear the potential to yield deceptive deductions or flawed prioritization strategies. In this paper, we introduce an extensive array of probabilistic sensitivity analysis techniques, coupled with post-processing of their outputs to rigorously quantify the inherent uncertainties. To establish a cohesive framework, we present an integrative formulation, subsequently applied to a reinforced concrete bridge component. Additionally, we present a pioneering technique, involving intensifying artificial acceleration in conjunction with the sensitivity matrix concept, ushering in a novel path-dependent sensitivity metric. Comparative analyses against other probabilistic sensitivity methodologies underscore the distinctive attributes of our proposed approach, spanning accuracy, computational efficiency, comprehensiveness, and result stability. Our proposed technique exhibits remarkable efficiency, culminating in a computational time reduction of approximately 90%.
Highlights A path-dependent sensitivity matrix is introduced in seismic analysis. The results of the sensitivity assessment depend on the stressor type. The importance of material randomness is altered with the loading intensity.
Probabilistic sensitivity matrices under stressor uncertainty
https://orcid.org/0000-0001-6772-1468
Abstract Quantitative risk analysis critically relies on precise numerical simulations to furnish optimal estimations. These numerical models encompass complex constitutive material correlations, multi-fidelity assumptions within model generation, and the encompassing variability of applied loading conditions. Traditional sensitivity analysis methodologies, being contingent upon the specific stressors applied, bear the potential to yield deceptive deductions or flawed prioritization strategies. In this paper, we introduce an extensive array of probabilistic sensitivity analysis techniques, coupled with post-processing of their outputs to rigorously quantify the inherent uncertainties. To establish a cohesive framework, we present an integrative formulation, subsequently applied to a reinforced concrete bridge component. Additionally, we present a pioneering technique, involving intensifying artificial acceleration in conjunction with the sensitivity matrix concept, ushering in a novel path-dependent sensitivity metric. Comparative analyses against other probabilistic sensitivity methodologies underscore the distinctive attributes of our proposed approach, spanning accuracy, computational efficiency, comprehensiveness, and result stability. Our proposed technique exhibits remarkable efficiency, culminating in a computational time reduction of approximately 90%.
Highlights A path-dependent sensitivity matrix is introduced in seismic analysis. The results of the sensitivity assessment depend on the stressor type. The importance of material randomness is altered with the loading intensity.
Probabilistic sensitivity matrices under stressor uncertainty
Hariri-Ardebili, Mohammad Amin (author) / Segura, Christopher L. Jr. (author) / Sattar, Siamak (author)
2024-02-16
Article (Journal)
Electronic Resource
English
Probabilistic sensitivity matrices under stressor uncertainty
| Elsevier | 2024
| British Library Conference Proceedings | 2007
Impact of Role Stressor Research
| British Library Conference Proceedings | 1994
Development of a Postpartum Stressor Measure
| British Library Online Contents | 2015