Effect of Training Data Ratio and Normalizing on Fatigue Lifetime Prediction of Aluminum Alloys with Machine Learning

Document Type : Original Article

Authors

Faculty of Mechanical Engineering, Semnan University, Semnan, Iran

Abstract

It is critical to evaluate the estimation of the fatigue lifetimes for the piston aluminum alloys, particularly in the automotive industry. This paper investigates the effect of different normalization methods on the performance of the fatigue lifetime estimation using Extreme Gradient Boosting (XGBoost), as a supervised machine learning method. For this purpose, the dataset used in this study includes various physical and experimental inputs related to an aluminum alloy and the corresponding fatigue lifetime outputs. Furthermore, before fitting the XGBoost model, different fatigue lifetime preprocessing methods were utilized and evaluated using metrics such as Root Mean Square Error (RMSE), Determination Coefficient (R2), and Scatter Band (SB). The results indicate that modeling fatigue lifetime with logarithmic values as a preprocessing method excels when XGBoost is trained with 100% of the data. However, other normalization methods demonstrate superior accuracy in estimating test data with a 20% test and 80% train set split.

Graphical Abstract

Effect of Training Data Ratio and Normalizing on Fatigue Lifetime Prediction of Aluminum Alloys with Machine Learning

Keywords

Main Subjects

References

  1. Farrahi G, Faghidian S, Smith D. Reconstruction of residual stresses in autofrettaged thick-walled tubes from limited measurements. International Journal of Pressure Vessels and Piping. 2009;86(11):777-84. 10.1016/j.ijpvp.2009.03.010
  2. Ismaiel A. Wind turbine blade dynamics simulation under the effect of atmospheric turbulence. Emerging Science Journal. 2023;7(1):162-76. 10.28991/ESJ-2023-07-01-012
  3. Ali Faghidian S. Analytical approach for inverse reconstruction of eigenstrains and residual stresses in autofrettaged spherical pressure vessels. Journal of Pressure Vessel Technology. 2017;139(4):041202. 10.1115/1.4035980
  4. Farrahi G, Faghidian S, Smith D. An inverse method for reconstruction of the residual stress field in welded plates. 2010. 10.1115/1.4001268
  5. Azadi M, Farrahi G, Winter G, Eichlseder W. The effect of various parameters on out-of-phase thermo-mechanical fatigue lifetime of A356. 0 cast aluminum alloy. International Journal of Engineering, Transactions C: Aspects. 2013;26(12):1461-70. 10.5829/idosi.ije.2013.26.12c.06
  6. Azadi M, Rezanezhad S, Zolfaghari M. Effects of various ageing heat treatments on microstructural features and hardness of piston aluminum alloy. International Journal of Engineering, Transactions A: Basics. 2019;32(1):92-8. 10.5829/ije.2019.32.01a.12
  7. Akhtar M, Qamar SZ, Muhammad M, Nadeem A. Optimum heat treatment of aluminum alloy used in manufacturing of automotive piston components. Materials and Manufacturing Processes. 2018;33(16):1874-80. 10.1080/10426914.2018.1512128
  8. Azadi M, Parast MSA. Data analysis of high-cycle fatigue testing on piston aluminum-silicon alloys under various conditions: Wear, lubrication, corrosion, nano-particles, heat-treating, and stress. Data in brief. 2022;41:107984. 10.1016/j.dib.2022.107984
  9. Choi D-K. Data-driven materials modeling with XGBoost algorithm and statistical inference analysis for prediction of fatigue strength of steels. International Journal of Precision Engineering and Manufacturing. 2019;20:129-38. 10.1007/s12541-019-00048-6
  10. Matin M, Azadi M. A Novel Machine Learning-Based Model for Predicting of Transient Fatigue Lifetime in Piston Aluminum Alloys. Available at SSRN 4598611. 10.2139/ssrn.4598611
  11. Matin M, Azadi M. Machine learning-based modeling for estimating bending fatigue lifetimes in AlSi12CuNiMg aluminum alloy of engine pistons under different inputs: Fretting force, corrosion time, lubrication, heat-treating, nano-particles, and stress. Corrosion Time, Lubrication, Heat-Treating, Nano-Particles, and Stress. 2023. 10.2139/ssrn.4549376
  12. Munkhdalai L, Munkhdalai T, Park KH, Lee HG, Li M, Ryu KH. Mixture of activation functions with extended min-max normalization for forex market prediction. IEEE Access. 2019;7:183680-91. 10.1109/ACCESS.2019.2959789
  13. Zhou W, Liu A, Wu L, Chen X. A L1 normalization enhanced dynamic window method for SSVEP-based BCIs. Journal of Neuroscience Methods. 2022;380:109688. 10.1016/j.jneumeth.2022.109688
  14. Dai Z, Chen W, Huang X, Li B, Zhu L, He L, et al., editors. Cnn descriptor improvement based on l2-normalization and feature pooling for patch classification. 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO); 2018: IEEE. 10.1109/ROBIO.2018.8665330
  15. Gómez-Escalonilla V, Martínez-Santos P, Martín-Loeches M. Preprocessing approaches in machine-learning-based groundwater potential mapping: an application to the Koulikoro and Bamako regions, Mali. Hydrology and Earth System Sciences. 2022;26(2):221-43. 10.5194/hess-26-221-2022
  16. Lv S, Liu C, Chen D, Zheng J, You Z, You L. Normalization of fatigue characteristics for asphalt mixtures under different stress states. Construction and Building Materials. 2018;177:33-42. 10.1016/j.conbuildmat.2018.05.109
  17. Lv S, Wang P, Fan X, Cabrera MB, Hu L, Peng X, et al. Normalized comparative study on fatigue characteristics of different pavement materials. Construction and Building Materials. 2021;271:121907. 10.1016/j.conbuildmat.2020.121907
  18. Medar R, Rajpurohit VS, Rashmi B, editors. Impact of training and testing data splits on accuracy of time series forecasting in machine learning. 2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA); 2017: IEEE. 10.1109/ICCUBEA.2017.8463779
  19. Ramezan CA, Warner TA, Maxwell AE, Price BS. Effects of training set size on supervised machine-learning land-cover classification of large-area high-resolution remotely sensed data. Remote Sensing. 2021;13(3):368. 10.3390/rs13030368
  20. Meng Y, Yang N, Qian Z, Zhang G. What makes an online review more helpful: an interpretation framework using XGBoost and SHAP values. Journal of Theoretical and Applied Electronic Commerce Research. 2020;16(3):466-90. 10.3390/jtaer16030029
  21. Md AQ, Kulkarni S, Joshua CJ, Vaichole T, Mohan S, Iwendi C. Enhanced preprocessing approach using ensemble machine learning algorithms for detecting liver disease. Biomedicines. 2023;11(2):581. 10.3390/biomedicines11020581
  22. Chiu W-Y, Chen B-S. Mobile location estimation in urban areas using mixed Manhattan/Euclidean norm and convex optimization. IEEE transactions on Wireless Communications. 2009;8(1):414-23. 10.1109/T-WC.2009.080156
  23. Azadi M, Shahsavand A, Parast MSA. Analyzing experimental data from reciprocating wear testing on piston aluminum alloys, with and without clay nano-particle reinforcement. Data in Brief. 2022;45:108766. 10.1016/j.dib.2022.108766
  24. Nasiri H, Azadi M, Dadashi A. Interpretable extreme gradient boosting machine learning model for fatigue lifetimes in 3D-printed polylactic acid biomaterials. Available at SSRN 4364418. 2023. 10.2139/ssrn.4364418
  25. Eesa AS, Arabo WK. A normalization methods for backpropagation: a comparative study. Science Journal of University of Zakho. 2017;5(4):319-23. 10.25271/2017.5.4.381
  26. Singh D, Singh B. Investigating the impact of data normalization on classification performance. Applied Soft Computing. 2020;97:105524. 10.1016/j.asoc.2019.105524
  27. Surono S, Afitian MYF, Setyawan A, Arofah DKE, Thobirin A. Comparison of CNN Classification Model using Machine Learning with Bayesian Optimizer. HighTech and Innovation Journal. 2023;4(3):531-42. 10.28991/HIJ-2023-04-03-05
  28. Kurdthongmee W. Comprehensive Evaluation of Deep Neural Network Architectures for Parawood Pith Estimation. HighTech and Innovation Journal. 2023;4(3):543-59. 10.28991/HIJ-2023-04-03-06
  29. He L, Wang Z, Akebono H, Sugeta A. Machine learning-based predictions of fatigue life and fatigue limit for steels. Journal of Materials Science & Technology. 2021;90:9-19. 10.1016/j.jmst.2021.02.021
  30. Althnian A, AlSaeed D, Al-Baity H, Samha A, Dris AB, Alzakari N, et al. Impact of dataset size on classification performance: an empirical evaluation in the medical domain. Applied Sciences. 2021;11(2):796. 10.3390/app11020796
  31. Khameneh MJ, Azadi M. Evaluation of high-cycle bending fatigue and fracture behaviors in EN-GJS700-2 ductile cast iron of crankshafts. Engineering Failure Analysis. 2018;85:189-200. 10.1016/j.engfailanal.2017.12.017
  32. Cruz F, Castelli M. Learning Curves Prediction for a Transformers-Based Model. Available at SSRN 4305463. 2023. 10.28991/ESJ-2023-07-05-03
  33. Zhang G, Shi Y, Yin P, Liu F, Fang Y, Li X, et al. A machine learning model based on ultrasound image features to assess the risk of sentinel lymph node metastasis in breast cancer patients: Applications of scikit-learn and SHAP. Frontiers in Oncology. 2022;12:944569. 10.3389/fonc.2022.944569
  34. Giola C, Danti P, Magnani S. Learning curves: A novel approach for robustness improvement of load forecasting. Engineering Proceedings. 2021;5(1):38. 10.3390/engproc2021005038
International Journal of Engineering
Volume 37, Issue 7
TRANSACTIONS A: Basics
July 2024
Pages 1296-1305

Files

Share

How to cite

Statistics