PSI - Issue 75

Marco Bonato et al. / Procedia Structural Integrity 75 (2025) 677–690 / Structural Integrity Procedia (2025) / Structural Integrity Procedia (2025)

678

2

2

resonances and are made from multiple components. The output of the study is to (1) identify the factors considered during stochastic fatigue simulations and (2) find the best way to correlate FEA simulation to real vibration tests. Indeed, given the constraints faced nowadays by the automotive industry, the ultimate goal is the scale-up of the method at industrial level for a routine faster validation (zero test mindset). © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2025 organizers Keywords: Vibration, Fatigue, Simulation, Finite Element Analysis, Reliability, Correlation, Automotive. Introduction During the design validation of automotive components, Finite Element Analysis (FEA) structural simulation plays a crucial role to identify the design flaw and the weakness of the product. Such techniques have highly evolved in the last decades. If initially FEA simulation served the purpose of justification of the subsequent physical tests on prototypes, they later evolved into a more pro-active support, allowing the optimization of the design. Finally, with the advent of digital twins, simulation results are part of a more inclusive model-based engineering design that puts together the correlation between simulation modelling and measurements In a highly competitive industrial environment, automotive car makers rely more and more on virtual prototypes to validate their new vehicle platform before launching real parts. This approach is firstly cost effective, since simulation iterations are cheaper than physical tests to be performed initially on prototypes, then on pre-series part, and finally on assembled systems. Secondly, structural simulations used for design validation are significantly faster than the lengthy reliability tests typically conducted in testing facilities, for example, a typical Mono-axial vibration reliability test has a duration between 8 and 30 hours per axial orientation. The objective of most automotive OEMs cascade to the supply chain. In particular, tier1 suppliers face an increasing challenge to follow this “Zero Physical Tests” approach for reliability and fatigue. In this optic, FEA simulations are adopted to reach for the design freeze milestone, and as milestone and, as mentioned, ideally to replace physical tests (initially for the design validation phase, DV, but also targeting the product validation phase, PV). The correlation of FEA models plays a crucial role in this framework. If the failure mechanism is related to materials fatigue (a cumulative effect), such simulation should provide the results as expected useful life duration, or total cumulative damage at the “hot zone” of the component under validation. Unfortunately, typical FEA simulations are capable of providing only results based on the “nominal design”, i.e. any possible source of variability that affects the expected cumulative damage (and therefore the time-to-failure) is not included in the model. In this paper, we investigate the effect of such intrinsic variabilities on the predicted scatter of the fatigue time-to failure results. The case study illustrated will focus on vibration loadings, since vibration tests are longer and the scatter of the time-to-failure has high dispersion. resonances and are made from multiple components. The output of the study is to (1) identify the factors considered during stochastic fatigue simulations and (2) find the best way to correlate FEA simulation to real vibration tests. Indeed, given the constraints faced nowadays by the automotive industry, the ultimate goal is the scale-up of the method at industrial level for a routine faster validation (zero test mindset). © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the Fatigue Design 2025 organizers Keywords: Vibration, Fatigue, Simulation, Finite Element Analysis, Reliability, Correlation, Automotive. 1. Introduction During the design validation of automotive components, Finite Element Analysis (FEA) structural simulation plays a crucial role to identify the design flaw and the weakness of the product. Such techniques have highly evolved in the last decades. If initially FEA simulation served the purpose of justification of the subsequent physical tests on prototypes, they later evolved into a more pro-active support, allowing the optimization of the design. Finally, with the advent of digital twins, simulation results are part of a more inclusive model-based engineering design that puts together the correlation between simulation modelling and measurements acquired during the physical test phase. 1.1 Product Validation in Automotive In a highly competitive industrial environment, automotive car makers rely more and more on virtual prototypes to validate their new vehicle platform before launching real parts. This approach is firstly cost effective, since simulation iterations are cheaper than physical tests to be performed initially on prototypes, then on pre-series part, and finally on assembled systems. Secondly, structural simulations used for design validation are significantly faster than the lengthy reliability tests typically conducted in testing facilities, for example, a typical Mono-axial vibration reliability test has a duration between 8 and 30 hours per axial orientation. The objective of most automotive OEMs cascade to the supply chain. In particular, tier1 suppliers face an increasing challenge to follow this “Zero Physical Tests” approach for reliability and fatigue. In this optic, FEA simulations are adopted to reach for the design freeze milestone, and as milestone and, as mentioned, ideally to replace physical tests (initially for the design validation phase, DV, but also targeting the product validation phase, PV). The correlation of FEA models plays a crucial role in this framework. If the failure mechanism is related to materials fatigue (a cumulative effect), such simulation should provide the results as expected useful life duration, or total cumulative damage at the “hot zone” of the component under validation. Unfortunately, typical FEA simulations are capable of providing only results based on the “nominal design”, i.e. any possible source of variability that affects the expected cumulative damage (and therefore the time-to-failure) is not included in the model. In this paper, we investigate the effect of such intrinsic variabilities on the predicted scatter of the fatigue time-to failure results. The case study illustrated will focus on vibration loadings, since vibration tests are longer and the scatter of the time-to-failure has high dispersion. © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under the responsibility of Dr Fabien Lefebvre with at least 2 reviewers per paper 1. acquired during the physical test phase. 1.1 Product Validation in Automotive

Nomenclature ALT Nomenclature ALT

Accelerated Life Tests Design for Reliability Design Validation Finite Element Analysis Accelerated Life Tests Design for Reliability Design Validation Finite Element Analysis

DfR DV FEA LHS DfR DV FEA LHS

Latin Hypercube Sampling MTD Modal Transient Dynamic OEM Original Equipment Manufacturer PDF Probability Density Function Latin Hypercube Sampling MTD Modal Transient Dynamic OEM Original Equipment Manufacturer PDF Probability Density Function