Prediction of the Temperature of Liquid Aluminum and the Dissolved Hydrogen Content in Liquid Aluminum with a Machine Learning Approach
Abstract
:1. Introduction
2. Materials and Methods
3. Machine Learning Modeling
3.1. Data Acquisition
3.2. Data Acquisition
3.3. Machine Learning Model
4. Results and Discussion
4.1. Melt Temperature of the Liquid Aluminum
4.2. Dissolved Hydrogen Content in the Liquid Aluminum
4.3. Comparison of ML Model to Numerical Model
5. Conclusions
Author Contributions
Funding
Conflicts of Interest
References
- Puga, H.; Barbosa, J.; Gabriel, J.; Seabra, E.; Ribeiro, S.; Prokic, M. Evaluation of ultrasonic aluminium degassing by piezoelectric sensor. J. Mater. Process. Technol. 2011, 211, 1026–1033. [Google Scholar] [CrossRef]
- Dong, X.; Yang, H.; Zhu, X.; Ji, S. High strength and ductility aluminium alloy processed by high pressure die casting. J. Alloys Compd. 2019, 773, 86–96. [Google Scholar] [CrossRef]
- Bejaxhin, A.B.H.; Paulraj, G.; Prabhakar, M. Inspection of casting defects and grain boundary strengthening on stressed Al6061 specimen by NDT method and SEM micrographs. J. Mater. Res. Technol. 2019, 8, 2674–2684. [Google Scholar] [CrossRef]
- Fritzsche, A.; Hilgenberg, K.; Teichmann, F.; Pries, H.; Dilger, K.; Rethmeier, M. Improved degassing in laser beam welding of aluminum die casting by an electromagnetic field. J. Mater. Process. Technol. 2018, 253, 51–56. [Google Scholar] [CrossRef]
- Xu, H.; Han, Q.; Meek, T.T. Effects of ultrasonic vibration on degassing of aluminum alloys. Mater. Sci. Eng. A 2008, 473, 96–104. [Google Scholar] [CrossRef]
- Xiong, B.; Lin, X.; Wang, Z.; Yan, Q.; Yu, H. Microstructures and mechanical properties of vacuum counter-pressure casting A357 alloys solidified under grade-pressurising: Effects of melt temperature. Mater. Sci. Eng. A 2014, 611, 9–14. [Google Scholar] [CrossRef]
- Wang, F.; Wang, N.; Yu, F.; Wang, X.; Cui, J. Study on micro-structure, solid solubility and tensile properties of 5A90 Al–Li alloy cast by low-frequency electromagnetic casting processing. J. Alloys Compd. 2020, 820, 153318. [Google Scholar] [CrossRef]
- Liu, Y.; Jie, W.; Gao, Z.; Zheng, Y. Investigation on the formation of microporosity in aluminum alloys. J. Alloys Compd. 2015, 629, 221–229. [Google Scholar] [CrossRef]
- Yolshina, L.A.; Kvashinchev, A.G. Chemical interaction of liquid aluminum with metal oxides in molten salts. Mater. Des. 2016, 105, 124–132. [Google Scholar] [CrossRef]
- Monroe, R. Porosity in Castings. AFS Transcations 2005, 5, 1–28. [Google Scholar] [CrossRef]
- Mitrasinovic, A.; Robles Hernández, F.C.; Djurdjevic, M.; Sokolowski, J.H. On-line prediction of the melt hydrogen and casting porosity level in 319 aluminum alloy using thermal analysis. Mater. Sci. Eng. A 2006, 428, 41–46. [Google Scholar] [CrossRef]
- Zhao, L.; Pan, Y.; Liao, H.; Wang, Q. Degassing of aluminum alloys during re-melting. Mater. Lett. 2012, 66, 328–331. [Google Scholar] [CrossRef]
- Lapham, D.P.; Schwandt, C.; Hills, M.P.; Kumar, R.V.; Fray, D.J. The detection of hydrogen in molten aluminium. Ionics (Kiel) 2002, 8, 391–401. [Google Scholar] [CrossRef]
- El-Sayed, M.A.; Hassanin, H.; Essa, K. Bifilm defects and porosity in Al cast alloys. Int. J. Adv. Manuf. Technol. 2016, 86, 1173–1179. [Google Scholar] [CrossRef] [Green Version]
- Kumar, S.; Gupta, A.K.; Chandna, P. Optimization of Process Parameters of Pressure Die Casting using Taguchi Methodology. World Acad. Sci. Eng. Technol. 2012, 6, 590–594. [Google Scholar]
- Wang, J.-M.; Yan, H.-J.; Zhou, J.-M.; Li, S.-X.; Gui, G.-C. Optimization of parameters for an aluminum melting furnace using the Taguchi approach. Appl. Therm. Eng. 2012, 33, 33–43. [Google Scholar] [CrossRef]
- Quintana, I.; Azpilgain, Z.; Pardo, D.; Hurtado, I. Numerical Modeling of Cold Crucible Induction Melting. In Proceedings of the 2011 COMSOL conference, Stuttgart, Germany, 26–28 October 2011. [Google Scholar]
- Nieckele, A.O.; Naccache, M.F.; Gomes, M.S.P. Numerical Modeling of an Industrial Aluminum Melting Furnace. J. Energy Resour. Technol. 2004, 126, 72–81. [Google Scholar] [CrossRef]
- Buliński, P.; Smolka, J.; Golak, S.; Przyłucki, R.; Palacz, M.; Siwiec, G.; Melka, B.; Blacha, L. Numerical modelling of multiphase flow and heat transfer within an induction skull melting furnace. Int. J. Heat Mass Transf. 2018, 126, 980–992. [Google Scholar] [CrossRef]
- Brůna, M.; Sládek, A. Hydrogen analysis and effect of filtration on final quality of castings from aluminium alloy AlSi7Mg0.3. Arch. Foundry Eng. 2011, 11, 5–10. [Google Scholar]
- Lee, C.; So, T.; Shin, K. Effect of gas bubbling filtration treatment on microporosity variation in A356 aluminium alloy. Acta Metall. Sin. 2016, 29, 638–646. [Google Scholar] [CrossRef] [Green Version]
- Shih, T.-S.; Wen, K.-Y. Effects of Degassing and Fluxing on the Quality of Al-7%Si and A356.2 Alloys. Mater. Trans. 2005, 46, 263–271. [Google Scholar] [CrossRef] [Green Version]
- Arnberg, L.; Di Sabatino, M.; Dispinar, D.; Nordmark, A.; Akhtar, S. Degassing, hydrogen and porosity phenomena in A356. Mater. Sci. Eng. A 2010, 527, 3719–3725. [Google Scholar]
- Haghayeghi, R.; Bahai, H.; Kapranos, P. Effect of ultrasonic argon degassing on dissolved hydrogen in aluminium alloy. Mater. Lett. 2012, 82, 230–232. [Google Scholar] [CrossRef]
- Kim, S.B.; Cho, Y.H.; Lee, J.M.; Jung, J.G.; Lim, S.G. The effect of ultrasonic melt treatment on the microstructure and mechanical properties of Al-7Si-0.35Mg casting alloys. J. Korean Inst. Met. Mater. 2017, 55, 240–246. [Google Scholar]
- Uludağ, M.; Çetin, R.; Dispinar, D.; Tiryakioğlu, M. The effects of degassing, grain refinement & Sr-addition on melt quality-hot tear sensitivity relationships in cast A380 aluminum alloy. Eng. Fail. Anal. 2018, 90, 90–102. [Google Scholar]
- Tzanakis, I.; Lebon, G.S.B.; Eskin, D.G.; Pericleous, K. Investigation of the factors influencing cavitation intensity during the ultrasonic treatment of molten aluminium. Mater. Des. 2016, 90, 979–983. [Google Scholar] [CrossRef]
- Saternus, M. Influence of impeller shape on the gas bubbles dispersion in aluminium refining process. J. Achiev. Mater. Manuf. Eng. 2012, 55, 285–290. [Google Scholar]
- Warke, V.S.; Shankar, S.; Makhlouf, M.M. Mathematical modeling and computer simulation of molten aluminum cleansing by the rotating impeller degasser: Part II. Removal of hydrogen gas and solid particles. J. Mater. Process. Technol. 2005, 168, 119–126. [Google Scholar] [CrossRef]
- Elton, D.C.; Boukouvalas, Z.; Butrico, M.S.; Fuge, M.D.; Chung, P.W. Applying machine learning techniques to predict the properties of energetic materials. Sci. Rep. 2018, 8, 9059. [Google Scholar] [CrossRef]
- Liu, Y.; Zhao, T.; Ju, W.; Shi, S.; Shi, S.; Shi, S. Materials discovery and design using machine learning. J. Mater. 2017, 3, 159–177. [Google Scholar] [CrossRef]
- Pei, Z.; Yin, J. Machine learning as a contributor to physics: Understanding Mg alloys. Mater. Des. 2019, 172, 107759. [Google Scholar] [CrossRef]
- Wang, X. Based on Large-scale Data with Random Forest. IEEE/CAA J. Autom. Sin. 2017, 4, 770–774. [Google Scholar] [CrossRef]
- Wang, H.; Xu, A.; Tian, N.; He, F.; He, D. Hybrid Model of Molten Steel Temperature Prediction Based on Ladle Heat Status and Artificial Neural Network. J. Iron Steel Res. Int. 2014, 21, 181–190. [Google Scholar]
- Ghosh, I.; Das, S.K.; Chakraborty, N. An artificial neural network model to characterize porosity defects during solidification of A356 aluminum alloy. Neural Comput. Appl. 2014, 25, 653–662. [Google Scholar] [CrossRef]
- Shafyei, A.; Anijdan, S.H.M.; Bahrami, A. Prediction of porosity percent in Al-Si casting alloys using ANN. Mater. Sci. Eng. A 2006, 431, 206–210. [Google Scholar] [CrossRef]
- Lee, C.-H.; Lin, C.-R.; Chen, M.-S. Sliding window filtering: An efficient method for incremental mining on a time-variant database. Inf. Syst. 2005, 30, 227–244. [Google Scholar] [CrossRef]
- Moraes, D.A.O.; Oliveira, F.L.P.; Duczmal, L.H.; Cruz, F.R.B. Comparing the inertial effect of MEWMA and multivariate sliding window schemes with confidence control charts. Int. J. Adv. Manuf. Technol. 2016, 84, 1457–1470. [Google Scholar] [CrossRef]
- Delbart, C.; Valdes, D.; Barbecot, F.; Tognelli, A.; Richon, P.; Couchoux, L. Temporal variability of karst aquifer response time established by the sliding-windows cross-correlation method. J. Hydrol. 2014, 511, 580–588. [Google Scholar] [CrossRef]
- Matlab and Statistics and Machine Learning Toolbox (R2019a); The MathWorks, Inc.: Natick, MA, USA, 2019.
- Matlab Statistics and Machine Learning Toolbox, User’s Guide (R2019a); The MathWorks, Inc.: Natick, MA, USA, 2019.
- Kutner, M.H.; Nachtsheim, C.J.; Neter, J.; Li, W. Applied Linear Statistical Models, 5th ed.; McGraw-Hill/Irwin: New York, NY, USA, 2005. [Google Scholar]
- Seber, G.A.F.; Lee, A.J. Linear Regression Analysis, 2nd ed.; John Wiley and Sons, Inc.: New York, NY, USA, 2003. [Google Scholar]
- García, Á.; Anjos, O.; Iglesias, C.; Pereira, H.; Martínez, J.; Taboada, J. Prediction of mechanical strength of cork under compression using machine learning techniques. Mater. Des. 2015, 82, 304–311. [Google Scholar] [CrossRef]
- Breiman, L.; Friedman, J.; Stone, C.J.; Olshen, R.A. Classification and Regression Trees; Chapman and Hall/CRC: New York, NY, USA, 1984. [Google Scholar]
- Kim, D.; Jeon, J.; Kim, D. Predictive Modeling of Pavement Damage Using Machine Learning and Big Data Processing. J. Korean Soc. Hazard Mitig. 2019, 19, 95–107. [Google Scholar] [CrossRef] [Green Version]
- Rasmussen, C.E.; Williams, C.K.I. Gaussian Processes for Machine Learning; The MIT Press: Miami, FL, USA, 2006. [Google Scholar]
- Cortes, C.; Vapnik, V. Support-Vector Networks. Mach. Learn. 1995, 20, 273–297. [Google Scholar] [CrossRef]
- Vapnik, V. The Nature of Statistical Learning Theory, 2nd ed.; Springer-Verlag: New York, NY, USA, 2000. [Google Scholar]
- Breiman, L. Bagging predictors. Mach. Learn. 1996, 24, 123–140. [Google Scholar] [CrossRef] [Green Version]
- Hastie, T.; Tibshirani, F.; Jerome, R. The Elements of Statistical Learning Data Mining, Inference, and Prediction, 2nd ed.; Springer-Verlag: New York, NY, USA, 2009. [Google Scholar]
- Mohamed, H.; AbdelazimNegm; Salah, M.; Nadaoka, K.; Zahran, M. Assessment of proposed approaches for bathymetry calculations using multispectral satellite images in shallow coastal/lake areas: A comparison of five models. Arab. J. Geosci. 2017, 10, 1–17. [Google Scholar] [CrossRef]
- Zheng, M.; Chen, S.; Liu, Y. A Web-based Application for Sales Predication. Proc. Int. Conf. Softw. Eng. Res. Pract. 2018, 18, 218–222. [Google Scholar]
- Withers, C.S.; Nadarajah, S. Unbiased estimates for linear regression with roundoff error. Probab. Math. Stat. 2011, 31, 177–182. [Google Scholar]
Target Property | No. | Features | Experimental Condition | |
---|---|---|---|---|
To Construct Model | To Test Model | |||
Temperature of the liquid aluminum | 1 | Electric power (kW) | 15, 20, 25, 30 | 27 |
2 | Ingot weight (kg) | 4, 6, 7 | 6 | |
3 | Time for electric power (sec) | 5, 10, 15, 20, 25, 30, 40, 50, 60 | 10, 20, 30, 40, 50 | |
4 | Initial melt temperature (°C) | Measured value during process | ||
Dissolved hydrogen content | 1 | Gas flow rate (cc/min) | 2, 4, 6, 7 | 6 |
2 | Gas bubbling filtration (GBF) rpm | 1510 | 1510 | |
3 | Melt temperature (°C) | 700, 730, 760, 770, 790, 800 | 740 | |
4 | Treatment time in GBF (min) | 0.5, 1, 1.5, 2, 2.5, 3, 5, 7.5, 10, 12.5 | 0.5, 1, 1.5, 2, 2.5, 3, 5, 7.5, 10, 12.5 | |
5 | Initial hydrogen density | Measured value during process |
Kernel Name | Kernel Function |
---|---|
Rational Quadratic | |
Exponential | |
Squared Exponential | |
Matern 5/2 |
Kernel Name | Kernel Functions |
---|---|
Linear | |
Gaussian | |
Polynomial | where is in the set |
Statistics | Linear Regression | Regression Tree | Gaussian Process Regression | Support Vector Machine | Ensembles | Experimental |
---|---|---|---|---|---|---|
Mean | 744.9 | 744.8 | 744.1 | 743.3 | 741.6 | 744.5 |
Median | 743.2 | 744.1 | 742.7 | 742.4 | 742 | 743.1 |
Min | 705.6 | 702.2 | 703.7 | 713.1 | 712.5 | 704.9 |
Max | 787.5 | 790.2 | 787.4 | 773.9 | 765.8 | 787.1 |
Standard Deviation | 19.5 | 19.6 | 20.2 | 14.2 | 16.8 | 19.2 |
Metric | Linear Regression | Regression Tree | Gaussian Process Regression | Support Vector Machine | Ensembles |
---|---|---|---|---|---|
Root Mean Squared Error | 0.5284 | 2.1979 | 0.1609 | 1.5926 | 4.4455 |
Coefficient of Determination | 0.9997 | 0.995 | 1 | 0.9978 | 0.98 |
Mean Squared Error | 0.2792 | 4.8306 | 0.0259 | 2.5365 | 19.763 |
Mean Absolute Error | 0.392 | 1.3105 | 0.1092 | 1.3536 | 2.745 |
Estimated Parameters | Linear Regression | Regression Tree | Gaussian Process Regression | Support Vector Machine | Ensembles | Experimental |
---|---|---|---|---|---|---|
μ | 744.9 | 744.7 | 744.1 | 743.3 | 741.6 | 744.5 |
19.5 | 19.6 | 20.2 | 14.2 | 16.8 | 19.2 |
Statistics | Linear Regression | Regression Tree | Gaussian Process Regression | Support Vector Machine | Ensembles | Experimental |
---|---|---|---|---|---|---|
Mean | 0.1194 | 0.1194 | 0.1194 | 0.1192 | 0.1193 | 0.1194 |
Median | 0.1146 | 0.1148 | 0.1149 | 0.115 | 0.1154 | 0.115 |
Min | 0.0864 | 0.1 | 0.0999 | 0.0998 | 0.1 | 0.1 |
Max | 0.1843 | 0.2075 | 0.2175 | 0.1872 | 0.1884 | 0.22 |
Standard Deviation | 0.0132 | 0.0142 | 0.0142 | 0.0134 | 0.0127 | 0.0143 |
Statistic | Linear Regression | Regression Tree | Gaussian Process Regression | Support Vector Machine | Ensembles |
---|---|---|---|---|---|
Root Mean Squared Error | 0.0055 | 0.0029 | 0.0013 | 0.0029 | 0.0044 |
Coefficient of Determination | 0.8548 | 0.9851 | 0.9979 | 0.9781 | 0.9396 |
Mean Squared Error | 3.00 × 10−5 | 8.38 × 10−6 | 1.77 × 10−6 | 8.62 × 10−6 | 1.94 × 10−5 |
Mean Absolute Error | 0.0032 | 0.0014 | 0.0007 | 0.0011 | 0.0026 |
Estimated Parameters | Linear Regression | Regression Tree | GPR | SVM | Ensembles | Experimental |
---|---|---|---|---|---|---|
μ | −2.154 | −2.147 | −2.140 | −2.131 | −2.154 | −2.197 |
0.0720 | 0.0968 | 0.0923 | 0.0741 | 0.0911 | 0.1005 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Kim, M.-J.; Yun, J.P.; Yang, J.-B.-R.; Choi, S.-J.; Kim, D. Prediction of the Temperature of Liquid Aluminum and the Dissolved Hydrogen Content in Liquid Aluminum with a Machine Learning Approach. Metals 2020, 10, 330. https://doi.org/10.3390/met10030330
Kim M-J, Yun JP, Yang J-B-R, Choi S-J, Kim D. Prediction of the Temperature of Liquid Aluminum and the Dissolved Hydrogen Content in Liquid Aluminum with a Machine Learning Approach. Metals. 2020; 10(3):330. https://doi.org/10.3390/met10030330
Chicago/Turabian StyleKim, Moon-Jo, Jong Pil Yun, Ji-Ba-Reum Yang, Seung-Jun Choi, and DongEung Kim. 2020. "Prediction of the Temperature of Liquid Aluminum and the Dissolved Hydrogen Content in Liquid Aluminum with a Machine Learning Approach" Metals 10, no. 3: 330. https://doi.org/10.3390/met10030330