Gaussian process regression based remaining fatigue life prediction for metallic materials under two-step loading

Gao, Jingjing; Wang, Cunjun; Zhou, Xu; Wang, Jun; Song, Yan; Wang, Zhen

doi:10.1016/j.ijfatigue.2022.106730

Cited by 31 publications

(9 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such datasets, suitable for comparison of fatigue damage accumulation models, are scarce. The state‐of‐the‐art verification of fatigue damage accumulation models is based on block loading spectra, of which two‐level block loading is the most commonly used 27 . The most extensive suitable dataset published in literature was developed by Manson et al in 1967 28 .…”

Section: Introductionmentioning

confidence: 99%

“…The state-of-the-art verification of fatigue damage accumulation models is based on block loading spectra, of which two-level block loading is the most commonly used. 27 The most extensive suitable dataset published in literature was developed by Manson et al in 1967. 28 Their work was also published in the book ASTM STP 415, 29 but the tabular data were only published in the NASA Technical note.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Method for statistical evaluation of cumulative damage models applied to block loading

Hectors

Hecke

Waele

2022

Fatigue Fract Eng Mat Struct

View full text Add to dashboard Cite

There is an abundance of nonlinear fatigue damage accumulation models but a lack of verification and comparison to experimental datasets. First, an extensive two-level block loading experimental fatigue dataset is reprocessed according to current standard practice. Next, four nonlinear damage accumulation models and the Palmgren-Miner linear damage rule are critically compared using a range of statistical metrics. For the considered dataset, the Palmgren-Miner rule consistently performs worst and the damage curve approach is found to perform significantly better than the other nonlinear models. This study shows that the current practice of performance verification of new fatigue damage accumulation models in literature is too limited which enables cherry-picking of verification datasets to improve perceived performance.Future fatigue damage accumulation models should be verified much more rigorously to both readily available and new experimental datasets.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Method for statistical evaluation of cumulative damage models applied to block loading

Hectors

Hecke

Waele

2022

Fatigue Fract Eng Mat Struct

View full text Add to dashboard Cite

show abstract

“…Equation (27) estimates the output values for the regions R lm , thus Equation ( 26) can be written as…”

Section: Application Of Machine Learningmentioning

confidence: 99%

“…Machine learning techniques are effective tools to improve the speed of predictive models, as shown recently in various fields of science. [8][9][10][11][12][13][14][15][16][17][18][19][20] Data-driven machine learning methods were employed successfully to predict fatigue life, [21][22][23][24][25][26][27][28][29][30][31] fatigue crack growth, [32][33][34][35] creep, [36][37][38][39][40] and creepfatigue 41,42 of metallic alloys, but less research has explored the probabilistic modeling of creep and fatigue. Previous work includes [43][44][45] applying a distributed collaborative wavelet neural network regression (DCWNNR), a distributed-coordinated neural network metamodel (DCNNM), and a decomposed collaborative time-variant Kriging surrogate (DCTKS) model approach to make probabilistic assessments of low cycle fatigue (LCF) and creep-fatigue life of turbine disks.…”

Section: Introductionmentioning

confidence: 99%

Data‐driven prediction of the probability of creep–fatigue crack initiation in 316H stainless steel

Chavoshi

Tagarielli

2022

Fatigue Fract Eng Mat Struct

View full text Add to dashboard Cite

Stainless steel components in advanced gas‐cooled reactors (AGRs) are susceptible to creep–fatigue cracking at high temperatures. Quantifying the probability of creep–fatigue crack initiation requires probabilistic numerical simulations; these are complex and computationally intensive. Here, we present a data‐driven approach to develop fast probabilistic surrogate models of creep–fatigue crack initiation in 316H stainless steel. We perform a set of Monte Carlo simulations based on the R5V2/3 high temperature assessment procedure and determine the sensitivity of the probability of crack initiation to loads and operating conditions. The data are used to train different supervised machine learning models considering Bayesian hyperparameter optimization. We discuss the relative performance of such models and show that a gradient tree boosting algorithm results in surrogate models with the highest accuracy.

show abstract

“…This again implicitly shows that the amount of suitable experimental data in the literature is limited. The work of Gan et al [22] inspired Gao et al [23] to further investigate remaining fatigue life prediction under two-step loading, comparing a number of machine learning models to conventional fatigue damage accumulation models. They found that all the machine learning models performed better than the conventional models.…”

Section: Introductionmentioning

confidence: 99%

Open-Access Experiment Dataset for Fatigue Damage Accumulation and Life Prediction Models

Hectors

Vanspeybrouck

Plets

et al. 2023

Metals

View full text Add to dashboard Cite

This work addresses the lack of focus on verification and comparison of existing fatigue damage accumulation and life prediction models on the basis of large and well-documented experiment datasets. Sixty-four constant amplitude, 54 two-level block loading, and 27 three-level block loading valid experiments were performed in order to generate an open-access, high-quality dataset that can be used as a benchmark for existing models. In the future, more experiments of various specimen geometries and loading conditions will be added. The obtained dataset was used for a study comparing five (non)linear fatigue damage and life prediction models. It is shown how the performance of several (non)linear damage models is strongly dependent on the considered material dataset and loading sequence. Therefore, it is important to verify models with a broad set of independent datasets, as many existing models show significant bias to certain datasets.

show abstract

Gaussian process regression based remaining fatigue life prediction for metallic materials under two-step loading

Cited by 31 publications

References 41 publications

Method for statistical evaluation of cumulative damage models applied to block loading

Method for statistical evaluation of cumulative damage models applied to block loading

Data‐driven prediction of the probability of creep–fatigue crack initiation in 316H stainless steel

Open-Access Experiment Dataset for Fatigue Damage Accumulation and Life Prediction Models

Contact Info

Product

Resources

About