PSI - Issue 75

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

690 14

The statistical plot shows an excellent level of correlation. The difference between the Weibull parameters beta and eta of the FEA results shows a difference of less than 5%. All graphics show an overlap between the two populations within the 90% confidence interval. In conclusion, the outcome of these investigations shows how important it is to consider the right material properties when performing FEA simulations, particularly for stochastic fatigue life prediction. FEA fatigue simulations are able to provide exploitable results as long as the material cards are known with a high degree of confidence. Great effort is to be given to the collection of such data (the material cards), even though it might represent an initial cost to be budgeted, however the gain is clearly assessed (the ability to replace physical tests at least for the DV phase). A similar conclusion (even though the perspective of the study were different) is to be found in a previous investigation [Czerlunczakiewicz et al. (2023)]: simulations performed with material parameters too “generic” provide results that are neither accurate nor useful. The novelty of the present research lies in the iterative convergence of the FEA fatigue life simulation towards the experimental value by adjusting the simulation parameters to the value of the material being used, after its laboratory characterization. Indeed, the results clearly show that by tuning the FEA model, the output shows an excellent level of correlation when compared with the experimental data. 5 Further Works Further works are planned to complete the framework of FEA based stochastic simulations. - Vibration signal simplification: the FEA stochastic simulation is performed as frequency analysis method, not as transient calculation. For this purpose, the initial sine sweep on a random signal is converted to a pure random and pure sinus signal [Yang (2025)]. - Two-resonance model: for both FEA and experiment measurement, the model is now prototype having 2 modal shapes, as in [Yang (2025)] - Lower intensity input signal: to favor a longer time to failure and border experimental scatter. This could favor the correlation even in the case of not characterized materials. - Weibull 3 parameters: to refine the statistical analysis by comparing also the Weibull location (failure free period). - Industrial case study: the same methods are applied to a real automotive component undergoing a reliability vibration test provided by one of our customers. References Carnell, R., 2024. lhs: Latin Hypercube Samples, Retrieved 5th May 2025: https://cran.r-project.org/web/packages/lhs/lhs.pdf Czerlunczakiewicz, E., Majerczak, M. and Bonato, M., 2023. Variability Of Fatigue Simulation Predictions For Automotive Components, 2023 Annual Reliability and Maintainability Symposium (RAMS), Orlando, FL, USA. Halfpenny, A., Chabod, A., Czapski, P., Aldred, J., Munson, K., Bonato, M., 2019. Probabilistic Fatigue and Reliability Simulation, Procedia Structural Integrity, Volume 19, 2019, Pages 150-167. Bonato, M., and Delaux, D., 2015. Synthesis and validation of accelerated vibration durability tests, Annual Reliability and Maintainability Symposium (RAMS), Palm Harbor, FL, USA, 2015. Bonato, M., Duraipandi, A., Sridhar, K. and Leon, R., 2023. Equivalence of Vibration Signals for Fatigue Simulation. Effect of parameters on durability predictions, Procedia Structural Integrity, Volume 57, 2024, Pages 799-809. Bonato, M., and Pierrat, L., 2024. Statistical Estimation of the Minimal Resistance Threshold. Case Study of Automotive Heat Exchangers, Annual Reliability and Maintainability Symposium (RAMS), Albuquerque, NM, USA. Strzelecki, P., 2021. Determination of fatigue life for low probability of failure for different stress levels using 3-parameter Weibull distribution, Int. J. Fat. 145, 106080. Weibull, W., 1939. The Statistical Theory of the Strength of Materials". Ingeniors Vetenskaps Academy Handlingar (151). Stockholm: Generalstabens Litografiska Anstalts Förlag: 1 – 45. Yang, Z., Sridhar, K., and Lingareddy, M., 2025. A Comparative Study of Fatigue Damage from Simultaneous Swept Sine and Random Acceleration for Structures with Distinct Dynamic Response Characteristics," SAE Technical Paper 2025-01-8232.