Fatigue Life Prediction of 6082 Aluminum Alloy with Machine Learning

Yıl 2025, Cilt: 6 Sayı: 1, 15 - 31, 19.06.2025

Resul Ünal , Recai Kuş , Mustafa Acarer

https://doi.org/10.55546/jmm.1606483

Öz

Aluminum alloys are one of the most preferred materials in industry. One of the most important properties of aluminum is its low density. In this way, it has taken its place as an important engineering material in many sectors, including the automotive sector. One of the aluminum alloys used in the automotive industry is the 6082 series aluminum alloy. Machine parts subjected to repetitive loads develop and accumulate micro cracks over time and these cracks cause sudden breakage. It is of great importance to understand this phenomenon known as fatigue in materials and to determine the number of fracture cycles by performing fatigue tests. In this study, during the fatigue tests of 6082 aluminum alloy, current and voltage values were collected from the specimen using DCPD (Direct Current Potential Drop) technique, while the applied force and displacement data were recorded simultaneously. These data were then given as input to 5 different machine learning algorithms: decision trees, extra trees, random forest, XGBoost and KNN (K-Nearest Neighbor) and the number of cycles were estimated. Among the tested models, the decision trees machine learning model performed the best based on R-square (R2) and mean absolute percentage error (MAPE) values.

Anahtar Kelimeler

Fatigue , DCPD , 6082 Aluminum alloy , Machine learning , Prediction

Makine Öğrenmesi ile 6082 Alüminyum Alaşımının Yorulma Ömrü Tahmini

Öz

Alüminyum alaşımlar endüstride oldukça sık tercih edilen malzemelerdendir. Alüminyumun en önemli özelliklerinden biri yoğunluklarının düşük olmasıdır. Bu sayede otomotiv sektörü de dahil olmak üzere birçok sektörde önemli bir mühendislik malzemesi olarak yerini almıştır. Otomotiv sektöründe kullanılan alüminyum alaşımlardan bir tanesi de 6082 serisi alüminyum alaşımıdır. Tekrarlı yüklere maruz kalan makine parçalarında zamanla mikro çatlaklar oluşarak birikir ve bu çatlaklar ani kırılmalara sebep olur. Malzemelerde yorulma olarak bilinen bu olgunun anlaşılması ve yorulma deneylerinin yapılarak kırılma çevrim sayılarının belirlenmesi büyük önem taşır. Bu çalışmada 6082 alüminyum alaşımının yorulma testleri sırasında DCPD (Direct Current Potential Drop) tekniğiyle numune üzerinden akım ve gerilim değerleri toplanırken, uygulanan kuvvet ve meydana gelen deplasman verileri de eş zamanlı olarak kaydedilmiştir. Daha sonra bu veriler karar ağacı, ekstra ağaçlar, rastgele orman, XGBoost (Aşırı Gradyan Arttırma) ve KNN (K-En Yakın Komşu) olmak üzere 5 farklı makine öğrenmesi algoritmasına girdi olarak verilmiş ve çevrim sayıları tahmin edilmiştir. Test edilen modeller arasında R-kare (R2) ve ortalama mutlak yüzde hata (MAPE) değerleri baz alındığında en iyi performansı karar ağacı ve ekstra ağaçlar makine öğrenmesi modelleri göstermiştir.

Anahtar Kelimeler

Yorulma , DCPD , 6082 Alüminyum alaşımı , Makine öğrenmesi , Tahmin

Teşekkür

Yazarlar Selçuk Üniversitesi Bilimsel Araştırma Projeleri Koordinatörlüğüne maddi desteklerinden dolayı teşekkür ederler. Ayrıca yazarlar bu çalışmada gerçekleştirilen yorulma deneylerindeki katkılarından ve yardımlarından dolayı AYD ARGE ekibine teşekkür ederler.

Kaynakça

• Abdullatef M. S., Alzubaidi F. N., Al-Tamimi A., Mahmood Y. A., Fatigue Life Estimation of High Strength 2090-T83 Aluminum Alloy Under Pure Torsion Loading Using Various Machine Learning Techniques, Fluid Dynamics and Materials Processing, 19(8), 2083–2107, 2023.
• Arunachalam S., Fawaz S., Test Method for Corrosion Pit-To-Fatigue Crack Transition From A Corner of Hole in 7075-T651 Aluminum Alloy, International Journal of Fatigue, 91, 50–58, 2016.
• ASTM, Specification for Aluminum and Aluminum-Alloy Extruded Bars, Rods, Wire, Profiles, and Tubes, Test, 2021.
• ASTM, Standard Practice for Conducting Force Controlled Constant Amplitude Axial Fatigue Tests of Metallic Materials, Test, 03, 4–8, 2002.
• Bhardwaj H. K., Shukla M., Low-Cycle Fatigue Life Prediction of Austenitic Stainless Steel Alloys: A Data-Driven Approach with Identification of Key Features, International Journal of Fatigue, 187, 108454, 2024.
• Birol Y., Gokcil E., Ali M., Akdi S., A Processing of High Strength EN AW 6082 Forgings Without a Solution Heat Treatment, Materials Science & Engineering:A, 674, 25–32, 2016.
• Breiman L., Random forests, Machine Learning, 45(1), 5–32, 2001.
• Cai Q., Mendis C. L., Chang I. T. H., Fan Z., Microstructure Evolution and Mechanical Properties of New Die-Cast Al-Si-Mg-Mn Alloys, Materials & Design, 187, 108394, 2020.
• Černý I., The use of DCPD Method for Measurement of Growth of Cracks in Large Components at Normal and Elevated Temperatures, Engineering Fracture Mechanics, 71(4–6), 837–848, 2004.
• Chen J., Liu Y., Fatigue Modeling Using Neural Networks: A Comprehensive Review, Fatigue and Fracture of Engineering Materials and Structures, 45(4), 945–979, 2022.
• Chen J., Zhao F., Sun Y., Yin Y., Improved XGBoost Model Based On Genetic Algorithm, International Journal of Computer Applications in Technology, 62(3), 240, 2020.
• Çavdar F., Günen A., Sert M., Borlanmış AISI H11 Takım Çeliğinin Kaplama Özellikleri ve Korozyon Oranının Makine Öğrenmesi Temelli Modellenmesi, Çukurova Üniversitesi Mühendislik Fakültesi Dergisi, 39(3), 625–638, 2024.
• Doré M. J., Maddox S. J., Accelerated Fatigue Crack Growth in 6082 T651 Aluminium Alloy Subjected to Periodic Underloads, Procedia Engineering, 66, 313–322, 2013.
• Doremus L., Nadot Y., Henaff G., Mary C., Pierret S., Calibration of The Potential Drop Method for Monitoring Small Crack Growth from Surface Anomalies - Crack Front Marking Technique and Finite Element Simulations, International Journal of Fatigue, 70, 178–185, 2015.
• Efeoğlu E., Kablosuz Sinyal Gücünü Kullanarak İç Mekan Kullanıcı Lokalizasyonu için Karar Ağacı Algoritmalarının Karşılaştırılması, Acta Infologica, 163 - 173, 2022.
• Elmousalami H. H., Artificial Intelligence and Parametric Construction Cost Estimate Modeling: State-of-the-Art Review, Journal of Construction Engineering and Management, 146(1) 2020.
• European Aluminium, Aluminium Usage in Cars Surges As Automotive Industry Shifts Towards Electrification, (May) www.european-aluminium.eu.
• Funk M., Bär J., DCPD Based Detection of the Transition from Short to Long Crack Propagation in Fatigue Experiments on The Aluminum Alloy 7475 T761, Procedia Structural Integrity, 17, 183–189, 2019.
• Geurts P., Ernst D., Wehenkel L., Extremely Randomized Trees. Machine Learning, 63(1), 3–42, 2006.
• Gitter R., Aluminium Materials for Structural Engineering – Essential Properties and Selection of Materials, Structural Engineering International, 16(4), 294–300, 2006.
• Guo J., Zan X., Wang L., Lei L., Ou C., Bai S., A Random Forest Regression with Bayesian Optimization-Based Method for Fatigue Strength Prediction of Ferrous Alloys, Engineering Fracture Mechanics, 293, 109714, 2023.
• Guo Y. B., Liu K. G., Lan Z. Y., Du X. K., Fatigue Life Prediction and System Development for 6082-T6 Aluminum Alloy Under Variable Amplitude Loading, International Conference on Advanced Sensing and Smart Manufacturing (ASSM 2022), SPIE 2022.
• He L., Wang Z. L., Akebono H., Sugeta A., Machine Learning-Based Predictions of Fatigue Life and Fatigue Limit for Steels, Journal of Materials Science & Technology, 90, 9–19, 2021.
• Jíŝa D., Liŝkutín P., Kruml T., Polák J., Small Fatigue Crack Growth in Aluminium Alloy EN-AW 6082/T6, International Journal of Fatigue, 32(12), 1913–1920, 2010.
• Kalita B., Abhiraaj R. C., Jayaganthan R., Fatigue Life and Crack Growth Rate Prediction of Additively Manufactured 17-4 PH Stainless Steel Using Machine Learning, Procedia Structural Integrity, 56, 105–110, 2024.
• Karolczuk A., Skibicki D., Pejkowski Ł., Gaussian Process for Machine Learning-Based Fatigue Life Prediction Model Under Multiaxial Stress–Strain Conditions, Materials, 15(21) 2022.
• Kumar V., Singh I. V., Mishra B. K., Jayaganthan R., Crack Growth Behavior of 6082 Al Alloys Under Mixed Mode-I Loading, In: Kumar A., Kumar Singla Y., Maughan M.R. (Ed.), Fracture Behavior of Nanocomposites and Reinforced Laminate Structures (pp. 207–237), Springer Nature Switzerland: Cham 2024.
• Li S., Zhu Q., Lu Z., Yan H., Zhu C., Li P., Fatigue Life Prediction of AA2524 Thin Plate Strengthened using Compound Laser Heating and Laser Shot Peening Method. Theoretical and Applied Fracture Mechanics, 129, 104178, 2024.
• Lian Z., Li M., Lu W., Fatigue Life Prediction of Aluminum Alloy Via Knowledge-Based Machine Learning, International Journal of Fatigue, 157, 106716, 2022.
• Malek B., Mabru C., Chaussumier M., Fatigue Behavior of 2618-T851 Aluminum Alloy under Uniaxial And Multiaxial Loadings, International Journal of Fatigue, 131, 105322, 2020.
• Mann T., Härkegård G., Stärk K., Short Fatigue Crack Growth in Aluminium Alloy 6082-T6, International Journal of Fatigue, 29(9–11), 1820–1826, 2007.
• Mirzaei A. H., Haghi P., Shokrieh M. M., Prediction of Fatigue Life of Laminated Composites by Integrating Artificial Neural Network Model and Non-Dominated Sorting Genetic Algorithm, International Journal of Fatigue, 188, 108528, 2024.
• Ni C., Xue H., Wang S., Yang F., Zhao K., Effect of Deformation on Crack Extension Measurement for Compact Tension Specimen with the DCPD Technique, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 238(5), 1618–1628, 2024.
• Ojo S. A., Shrestha S., Manigandan K., Morscher G. N., Gyekenyesi A. L., Scott-Emuakpor O. E., Application of Small Geometry Specimens to Determine the Fatigue Crack Growth Anisotropy of Ti–6Al–4V Additively Manufactured for Repair, Results in Materials, 15 2022.
• Olguner S., Bozdana A. T., Prediction of Lankford Coefficients for AA1050 and AA5754 Aluminum Sheets Using Uniaxial Tensile Tests and Cup Drawing Experiments, Lecture Notes in Mechanical Engineering, 438–446, 2020.
• Pandey A., Jain A., Comparative Analysis of KNN Algorithm using Various Normalization Techniques, International Journal of Computer Network and Information Security, 9(11), 36–42, 2017.
• Paris P., Erdogan F., A Critical Analysis of Crack Propagation Laws, Journal of Basic Engineering, 85(4), 528–533, 1963.
• Peng Y., Zhang Y., Zhang L., Yao L., Guo X., Prediction of Corrosion Fatigue Crack Growth Rate in Aluminum Alloys Based on İncremental Learning Strategy, International Journal of Fatigue, 187, 108481, 2024.
• Pokharel A., Keesler-Evans J., Tempke R., Musho T., A Machine Learning Model for Predicting Progressive Crack Extension Based on Experimental Data Obtained Using DCPD Measurement Technique, Journal of Materials Research and Technology, 24, 5687–5701, 2023.
• Sagar S., Singh N. K., Maurya N. S., Compressıon, Flexure and Wear Behavıours of Aluminium 6082-T6 Alloy, U.P.B. Scientific Bulletin, Series D: Mechanical Engineering, 85(3), 193–204, 2023.
• Sarkar A., Aktunali M., Marthe Arbo S., Holmestad J., Mario Viespoli L., Nyhus B., Ringen G., Razavi N., A Study on the Influence Of Impurity Content on Fatigue Endurance In A 6082 Al-Alloy, International Journal of Fatigue, 186, 108406, 2024.
• Silva F. S., Pinho A. C. M., The Effect of Temperature on Crack Behavior In An 7175 Aluminum Alloy Under Mode I + Steady Mode III, European Structural Integrity Society, 29(C), 247–256, 2002.
• Song M., Liu J., Chen H., Hu Y., Shi Z., Yin H., Xia J., Berto F., Li R., Effects and Optimization of Biomimetic Laser Shock Peening on Residual Fatigue Life İmprovement of Aluminum Alloy Used In Aircraft Skin. Theoretical and Applied Fracture Mechanics, 117, 103155, 2022.
• Vavouliotis A., Paipetis A., Kostopoulos V., On the Fatigue Life Prediction of CFRP Laminates using The Electrical Resistance Change Method, Composites Science and Technology, 71(5), 630–642, 2011.
• Vecchiato L., Campagnolo A., Meneghetti G., Numerical Calibration and Experimental Validation of The Direct Current Potential Drop (DCPD) Method for Fracture Mechanics Fatigue Testing of Single-Edge-Crack Round Bars, International Journal of Fatigue, 150(February) 2021.
• Winter L., Hockauf K., Winter S., Lampke T., Equal-Channel Angular Pressing Influencing the Mean Stress Sensitivity In The High Cycle Fatigue Regime of the 6082 Aluminum Alloy, Materials Science and Engineering: A, 795, 140014, 2020.
• Yılmaz N. F., Çakır M. V., Yılmaz M., Saplama Kaynak Bağlantılarının Çekme Dayanımının ANFIS ile Modellenmesi, Çukurova Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi, 31(ÖS1), 79–88, 2016.