Predicting length of fatigue cracks by means of machine learning algorithms in the small-data regime

In this paper several statistical learning algorithms are used to predict the maximal length of fatigue cracks based on a sample composed of 31 observations. The small-data regime is still a problem for many professionals, especially in the areas where failures occur rarely. The analyzed object is a high-pressure Nozzle of a heavy-duty gas turbine. Operating parameters of the engines are used for the regression analysis. The following algorithms are used in this work: multiple linear and polynomial regression, random forest, kernel-based methods, AdaBoost and extreme gradient boosting and artificial neural networks. A substantial part of the paper provides advice on the effective selection of features. The paper explains how to process the dataset in order to reduce uncertainty; thus, simplifying the analysis of the results. The proposed loss and cost functions are custom and promote solutions accurately predicting the longest cracks. The obtained results confirm that some of the algorithms can accurately predict maximal lengths of the fatigue cracks, even if the sample is small.

[1]  Liyang Xie,et al.  Curved fatigue crack growth prediction under variable amplitude loading by artificial neural network , 2021 .

[2]  Wolfgang Ludwig,et al.  Using machine learning and a data-driven approach to identify the small fatigue crack driving force in polycrystalline materials , 2018, npj Computational Materials.

[3]  Joel Nothman,et al.  SciPy 1.0-Fundamental Algorithms for Scientific Computing in Python , 2019, ArXiv.

[4]  S. M. Beden,et al.  Review of fatigue crack propagation models for metallic components , 2009 .

[5]  John Yen,et al.  Introduction , 2004, CACM.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  N. A. Fellows,et al.  Artificial neural network for random fatigue loading analysis including the effect of mean stress , 2018, International Journal of Fatigue.

[8]  Mariel Alfaro-Ponce,et al.  Fatigue damage effect approach by artificial neural network , 2019, International Journal of Fatigue.

[9]  Wes McKinney,et al.  Data Structures for Statistical Computing in Python , 2010, SciPy.

[10]  Jan Deeken,et al.  Numerically Efficient Fatigue Life Prediction of Rocket Combustion Chambers using Artificial Neural Networks , 2019 .

[11]  R. Khanna,et al.  Support Vector Regression , 2015 .

[12]  Douglas M. Jones,et al.  Haemoconcentration, not decreased blood temperature, increases blood viscosity during cold water immersion. , 2020, Diving and hyperbaric medicine.

[13]  Matteo Iannitelli,et al.  A Hybrid Model for on-line Detection of Gas Turbine Lean Blowout Events , 2018 .

[14]  R. B. Kalombo,et al.  Fatigue life estimation of an all aluminium alloy 1055 MCM conductor for different mean stresses using an artificial neural network , 2020 .

[15]  Carmine Allegorico,et al.  A Data-Driven Approach for on-line Gas Turbine Combustion Monitoring using Classification Models , 2014 .

[16]  E. Newport,et al.  Science Current Directions in Psychological Statistical Learning : from Acquiring Specific Items to Forming General Rules on Behalf Of: Association for Psychological Science , 2022 .

[17]  Maciej Ł. Pawełczyk,et al.  Industrial gas turbine operating parameters monitoring and data-driven prediction , 2020, Eksploatacja i Niezawodnosc - Maintenance and Reliability.

[18]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[19]  P. C. Paris,et al.  A Critical Analysis of Crack Propagation Laws , 1963 .

[20]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[21]  Trevor Hastie,et al.  Linear Model Selection and Regularization , 2021, Springer Texts in Statistics.

[22]  Fausto Carlevaro,et al.  Use of Operating Parameters, Digital Replicas and Models for Condition Monitoring and Improved Equipment Health , 2018 .

[23]  Ya-Ju Fan,et al.  Regression with small data sets: a case study using code surrogates in additive manufacturing , 2018, Knowledge and Information Systems.

[24]  Rob Kitchin,et al.  Small data in the era of big data , 2015 .

[25]  David Nowell,et al.  A machine learning approach to the prediction of fretting fatigue life , 2020, Tribology International.

[26]  S. Cioncolini,et al.  Machine Learning in Gas Turbines , 2018, Mechanical Engineering.

[27]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[28]  Julian J. Faraway,et al.  When small data beats big data , 2018 .

[29]  Can Berk Kalayci,et al.  Soft computing methods for fatigue life estimation: A review of the current state and future trends , 2020 .

[30]  F. Viana,et al.  Physics-Informed Neural Networks for Bias Compensation in Corrosion-Fatigue , 2020, AIAA Scitech 2020 Forum.

[31]  K. Jarrod Millman,et al.  Array programming with NumPy , 2020, Nat..

[32]  Paris Perdikaris,et al.  Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations , 2019, J. Comput. Phys..

[33]  Robert B. Abernethy,et al.  The new Weibull handbook : reliability & statistical analysis for predicting life, safety, risk, support costs, failures, and forecasting warranty claims, substantiation and accelerated testing, using Weibull, Log normal, Crow-AMSAA, probit, and Kaplan-Meier models , 2006 .

[34]  Christos E. Athanasiou,et al.  A machine learning approach to fracture mechanics problems , 2020, Acta Materialia.

[35]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[36]  M. Karagiannopoulos,et al.  Feature Selection for Regression Problems , 2007 .

[37]  H. Lemu,et al.  A review of fatigue crack propagation modelling techniques using FEM and XFEM , 2017 .

[38]  Hua Li,et al.  Machine learning based fatigue life prediction with effects of additive manufacturing process parameters for printed SS 316L , 2021, International Journal of Fatigue.

[39]  Yun Xu,et al.  On Splitting Training and Validation Set: A Comparative Study of Cross-Validation, Bootstrap and Systematic Sampling for Estimating the Generalization Performance of Supervised Learning , 2018, Journal of Analysis and Testing.

[40]  Hao Wu,et al.  Estimation of remaining fatigue life under two-step loading based on kernel-extreme learning machine , 2021, International Journal of Fatigue.

[41]  Harris Drucker,et al.  Improving Regressors using Boosting Techniques , 1997, ICML.

[42]  Pierre Dardenne,et al.  Validation and verification of regression in small data sets , 1998 .