Simplified ResNet approach for data driven prediction of microstructure-fatigue relationship

Gebhardt, Christian; Trimborn, Torsten; Weber, Frank-Andreas; Bezold, Alexander; Broeckmann, Christoph; Herty, Michaël

doi:10.1016/j.mechmat.2020.103625

Cited by 25 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The predicted maximum relative errors for FNN and RBFNN are 19.79% and 31.25%, respectively. Other similar research predicting the fatigue strength as the output of the FNN can be found in previous works 239–249 …”

Section: Review Of Nn Applications In Fatiguementioning

confidence: 81%

Fatigue modeling using neural networks: A comprehensive review

Chen

Liu

2022

Fatigue Fract Eng Mat Struct

135

View full text Add to dashboard Cite

Neural network (NN) models have significantly impacted fatigue‐related engineering communities and are expected to increase rapidly due to the recent advancements in machine learning and artificial intelligence. A comprehensive review of fatigue modeling methods using NNs is lacking and will help to recognize past achievements and suggest future research directions. Thus, this paper presents a survey of 251 publications between 1990 and July 2021. The NN modeling in fatigue is classified into five applications: fatigue life prediction, fatigue crack, fatigue damage diagnosis, fatigue strength, and fatigue load. A wide range of NN architectures are employed in the literature and are summarized in this review. An overview of important considerations and current limitations for the application of NNs in fatigue is provided. Statistical analysis for the past and the current trend is provided with representative examples. Existing gaps and future research directions are also presented based on the reviewed articles.

show abstract

Section: Review Of Nn Applications In Fatiguementioning

confidence: 81%

Fatigue modeling using neural networks: A comprehensive review

Chen

Liu

2022

Fatigue Fract Eng Mat Struct

135

View full text Add to dashboard Cite

show abstract

“…Therefore, the effective processing of the fatigue performance database is the first task to realize accurate prediction. As shown in Figure 4 and Table 1, there are several approaches to establish a reliable fatigue performance database, which can be divided into three main categories: independent (experiment and simulation), [1,4,5,9,29, literature based, [13,[75][76][77][78][79][80][81][82][83][84][85] and data augmentation based. [86][87][88][89][90][91][92][93][94] Even an efficient data-driven method cannot construct an effective model via an insufficient database; thus, the establishment of a high-quality database is quite important.…”

Section: Establishment Methods For Fatigue Performance Databasementioning

confidence: 99%

A State‐of‐Art Review on Prediction Model for Fatigue Performance of Welded Joints via Data‐Driven Method

Feng

Zhao

et al. 2023

Adv Eng Mater

View full text Add to dashboard Cite

Fatigue fracture of welded joints is an important cause for engineering accidents. Due to the coexistence of so many influencing factors, the current prediction model of fatigue behavior cannot be used for all service conditions. The progress of each module (database establishment, data processing, data augmentation, dimension reduction, and comprehensive consideration of influencing factors) during the process of constructing a prediction model is summarized herein. Specifically, the state‐of‐art prediction models of fatigue strength, fatigue crack growth rate, fatigue life, and fatigue reliability are introduced. Moreover, suggestions for future work are given at the end of each section. Finally, according to the influential prediction methods, the establishment of fatigue performance prediction methods of welded joints via data‐driven method is summarized.

show abstract

“…Note that even the choice d = 1 is possible and, in addition, that networks of this type have been already proved to satisfy different formulations of the universal approximation theorem [31][32][33]. Further, they have been also applied to several (realworld) training problems [34,35].…”

Section: Neural Differential Equations and Mean-field Limitmentioning

confidence: 99%

Continuous limits of residual neural networks in case of large input data

Herty

Thünen

Trimborn³

et al. 2022

Communications in Applied and Industrial Mathematics

Self Cite

View full text Add to dashboard Cite

Residual deep neural networks (ResNets) are mathematically described as interacting particle systems. In the case of infinitely many layers the ResNet leads to a system of coupled system of ordinary differential equations known as neural differential equations. For large scale input data we derive a mean–field limit and show well–posedness of the resulting description. Further, we analyze the existence of solutions to the training process by using both a controllability and an optimal control point of view. Numerical investigations based on the solution of a formal optimality system illustrate the theoretical findings.

show abstract

Simplified ResNet approach for data driven prediction of microstructure-fatigue relationship

Cited by 25 publications

References 21 publications

Fatigue modeling using neural networks: A comprehensive review

Fatigue modeling using neural networks: A comprehensive review

A State‐of‐Art Review on Prediction Model for Fatigue Performance of Welded Joints via Data‐Driven Method

Continuous limits of residual neural networks in case of large input data

Contact Info

Product

Resources

About