PSI - Issue 54

Available online at www.sciencedirect.com Available online at www.sciencedirect.com Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 54 (2024) 521–535 Structural Integrity Procedia 00 (2023) 000–000 Structural Integrity Procedia 00 (2023) 000–000

www.elsevier.com / locate / procedia www.elsevier.com / locate / procedia

© 2023 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 ICSI 2023 organizers Abstract Material fatigue describes the failure due to cyclic loading. To ensure structural integrity, i.e. to design structures against material fatigue, a fatigue assessment is essential. Such an assessment can be conducted following either a statistical or a sampling-based ’non-statistical’ approach, often denoted frequency- resp. time-domain approach. Currently, the highly-e ffi cient statistical approach uses power spectral densities to characterize loading and stresses, while the sampling-based approach implies the computationally costly processing of time-domain realizations. However, real-world applications often involve non-stationary vibration loading, which commonly causes significant discrepancies between these approaches, leaving no alternative to the sampling-based approach. To bridge this gap, we propose a statistical approach that qualifies for non-stationary loading. This employs the non-stationarity matrix to characterize non-stationary loading and to statistically calculate response kurtosis for linear structures, which then serves as input for a machine learning (ML) model that predicts its e ff ect on fatigue damage. Herein, we detail the process of generating training data and defining an appropriate model. We compare the predicted fatigue damage and computational e ff ort to those of the established approaches. Our findings suggest that the integration of the non-stationarity matrix for structural dynamics in connection with an appropriate ML model can significantly enhance the prediction of fatigue damage under non-stationary loading conditions while retaining computational e ffi ciency. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of the scientific committee of the ICSI 2023 organizers. Keywords: non-stationary random vibration; non-stationarity matrix; neural network; spectral damage estimator International Conference on Structural Integrity 2023 (ICSI 2023) Data-driven spectral damage estimator for non-stationary vibration loading Arvid Trapp a, ∗ , David Fra¨ulin a , Marcin Hinz a , Peter Wolfsteiner a a Munich University of Applied Sciences, Department of Mechanical, Automotive and Aeronautical Engineering, Dachauer Strasse 98b, 80335 Munich, Germany Abstract Material fatigue describes the failure due to cyclic loading. To ensure structural integrity, i.e. to design structures against material fatigue, a fatigue assessment is essential. Such an assessment can be conducted following either a statistical or a sampling-based ’non-statistical’ approach, often denoted frequency- resp. time-domain approach. Currently, the highly-e ffi cient statistical approach uses power spectral densities to characterize loading and stresses, while the sampling-based approach implies the computationally costly processing of time-domain realizations. However, real-world applications often involve non-stationary vibration loading, which commonly causes significant discrepancies between these approaches, leaving no alternative to the sampling-based approach. To bridge this gap, we propose a statistical approach that qualifies for non-stationary loading. This employs the non-stationarity matrix to characterize non-stationary loading and to statistically calculate response kurtosis for linear structures, which then serves as input for a machine learning (ML) model that predicts its e ff ect on fatigue damage. Herein, we detail the process of generating training data and defining an appropriate model. We compare the predicted fatigue damage and computational e ff ort to those of the established approaches. Our findings suggest that the integration of the non-stationarity matrix for structural dynamics in connection with an appropriate ML model can significantly enhance the prediction of fatigue damage under non-stationary loading conditions while retaining computational e ffi ciency. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of the scientific committee of the ICSI 2023 organizers. Keywords: non-stationary random vibration; non-stationarity matrix; neural network; spectral damage estimator International Conference on Structural Integrity 2023 (ICSI 2023) Data-driven spectral damage estimator for non-stationary vibration loading Arvid Trapp a, ∗ , David Fra¨ulin a , Marcin Hinz a , Peter Wolfsteiner a a Munich University of Applied Sciences, Department of Mechanical, Automotive and Aeronautical Engineering, Dachauer Strasse 98b, 80335 Munich, Germany

Nomenclature Nomenclature

ANN ANN DK DK

artificial neural network artificial neural network

Dirlik Dirlik

E-mail address: arvid.trapp@hm.edu E-mail address: arvid.trapp@hm.edu

2452-3216 © 2023 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 ICSI 2023 organizers 10.1016/j.prostr.2024.01.115 ∗ Corresponding author 2210-7843 © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of the scientific committee of the ICSI 2023 organizers. ∗ Corresponding author 2210-7843 © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of the scientific committee of the ICSI 2023 organizers.