Artificial Intelligence Approaches to Determine Graphite Nodularity in Ductile Iron
DOI:
https://doi.org/10.7494/jcme.2021.5.4.94Abstract
The complex metallurgical interrelationships in the production of ductile cast iron can lead to enormous differences in graphite formation and local microstructure by small variations during production. Artificial intelligence algorithms were used to describe graphite formation, which is influenced by a variety of metallurgical parameters. Moreover, complex physical relationships in the formation of graphite morphology are also controlled by boundary conditions of processing, the effect of which can hardly be assessed in everyday foundry operations. The influence of relevant input parameters can be predetermined using artificial intelligence based on conditions and patterns that occur simultaneously. By predicting the local graphite formation, measures to stabilise production were defined and thereby the accuracy of structure simulations improved. In course of this work, the most important dominating variables, from initial charging to final casting, were compiled and analysed with the help of statistical regression methods to predict the nodularity of graphite spheres. We compared the accuracy of the prediction by using Linear Regression, Gaussian Process Regression, Regression Trees, Boosted Trees, Support Vector Machines, Shallow Neural Networks and Deep Neural Networks. As input parameters we used 45 characteristics of the production process consisting of the basic information including the composition of the charge, the overheating time, the type of melting vessel, the type of the inoculant, the fading, and the solidification time. Additionally, the data of several thermal analysis, oxygen activity measurements and the final chemical analysis were included.
Initial programme designs using machine learning algorithms based on neural networks achieved encouraging results. To improve the degree of accuracy, this algorithm was subsequently adapted and refined for the nodularity of graphite.
Downloads
References
Bendsoe M.P. & Sigmund O. (2003). Topology optimization: theory, methods, and applications. Springer Science & Business Media, Heidelberg.
Eschenauer H., Koski J. & Osyczka A. (Eds.) (2012). Multicriteria design optimization: procedures and applications. Springer Science & Business Media, Heidelberg.
Cziegler A. (2020). Optimization of Green Sand Thermophysical Properties for Simulation of Cast Iron Processes. International MAGMA User Meeting 2020. 6–8 October 2020.
Bauer W. (2017) Graphitmorphologie: Entstehung, Einflussgrößen, Kontrolle. Österr.eichisches Gießerei-Institut, Leoben.
Benedetti M., Fontanari V. & Lusuardi D. (2019). Effect of graphite morphology on the fatigue and fracture resistance of ferritic ductile cast iron. Engineering Fracture Mechanics, 206, 427–441.
Doi: https://doi.org/10.1016/j.engfracmech.2018.12.019.
Liu Y., Xing J., Li Y., Wang S. & Tao D. (2018). Tomographical study of the effect of graphite on properties of cast iron. Steel research international, 89(8), 1800086. Doi: https://doi.org/10.1002/srin.201800086.
Herfurth K. (2016). Wachstum der Graphitkristalle im Gusseisen unter besonderer Berücksichtigung der Grenzflächenspannung. Gießerei-Praxis, 6, 254–260.
Stefanescu D., Alonso G., Larrañaga P., De la Fuente E. & Suarez R. (2018). A comparative study of graphite growth in cast iron and in analogous systems. International Journal of Metalcasting, 12(4), 722–752. Doi: https://doi.org/10.1007/s40962-017-0204-1.
Thankachan T., Prakash K.S. & Jothi S. (2021). Artificial neural network modeling to evaluate and predict the mechanical strength of duplex stainless steel during casting. Sādhanā, 46(4), 197. Doi: https://doi.org/10.1007/s12046-021-01742-w.
Agrawal A. & Choudhary A. (2016). Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science. APL Materials, 4(5), 053208. Doi: https://doi.org/10.1063/1.4946894.
Bührig-Polaczek A., Rudack M., Salentin F.G., Sandt M., Hartmann D., Brait M., Koppensteiner E. & Schumacher P. (2021). Digitalisierung in der Gießereitechnik. Teil 1: Herausforderungen und Chancen. Gießerei, 6/21, 37–43.
Çöl M., Ertunç H.M. & Yılmaz M. (2007). An artificial neural network model for toughness properties in microalloyed steel in consideration of industrial production conditions. Materials & Design, 28(2), 488–495. Doi: https://doi.org/10.1016
/j.matdes.2005.09.001.
Bashar A. (2019). Survey on evolving deep learning neural network architectures. Journal of Artificial Intelligence and Capsule Networks, 1(2), 73–82. Doi: https://doi.org/10.36548/jaicn.
2.003.
LeCun Y., Bengio Y. & Hinton G. (2015). Deep learning. Nature, 521(7553), 436–444. Doi: https://doi.org/10.1038/nature14539.
Glavas Z., Lisjak D. & Unkic F. (2007). The application of artificial neural network in the prediction of the as-cast impact toughness of spheroidal graphite cast iron. Kovové Materiály, 45(1), 41–49.
Liliac M.M., Yamamoto K. & Ogi K. (2001). A neural network prediction for the graphite nucleation and morphology in modified cast irons. International Journal of Cast Metals Research, 14(3), 157–164.
Amini S. & Abbaschian R. (2013). Nucleation and growth kinetics of graphene layers from a molten phase. Carbon, 51, 110–123.
Doi: https://doi.org/10.1016/j.carbon.2012.08.019.
Stefanescu D.M., Alonso G., Larrañaga P., De la Fuente E. & Suarez R. (2016). On the crystallization of graphite from liquid iron-carbon-silicon melts. Acta Materialia, 107, 102–126. Doi: https://doi.org/10.1016/j.actamat.2016.01.047.
Ghahremani nahr J., Nozari H. & Sadeghi M.E. (2021). Artificial intelligence and Machine Learning for Real-world problems
(A survey). International Journal of Innovation in Engineering, 1(3), 38–47.
Fragassa C., Babic M., Bergmann C.P. & Minak G. (2019). Predicting the tensile behaviour of cast alloys by a pattern recognition analysis on experimental data. Metals, 9(5), 557. Doi: https://doi.org/10.3390/met9050557.
Perzyk M. & Kochański A.W. (2001). Prediction of ductile cast iron quality by artificial neural networks. Journal of Materials Processing Technology, 109(3), 305–307.
Swain S.K. (2008). Effect of Chemistry and Processing Variables on the Mechanical Properties of Thin Wall Ductile Iron Castings. National Institute of Technology: Rourkela [Doctoral dissertation].
Lim K.Y.H., Zheng P. & Chen C.H. (2020). A state-of-the-art survey of Digital Twin: techniques, engineering product lifecycle management and business innovation perspectives. Journal of Intelligent Manufacturing, 31(6), 1313–1337. Doi: https://doi.org/
1007/s10845-019-01512-w.
Nielsen F. (1993). Gieß- und Anschnitttechnik Grundlagen und Anwendung einer Methode. Giesserei-Verlag GmbH, Dusseldorf.
AMGuss. Professionelle Analyse der Mikrostruktur von Gusseisen. GFaI Gesellschaft zur Förderung angewandter Informatik, 30 November 2021, Retrieved from https://www.gfai.de/entwicklungen/bildverarbeitung/amguss [accessed 7.10.2021].
GFaI. Gesellschaft zur Förderung Angewandter Informatik, Am Guss, Version 1.1.60.
ISO/DIS 945-4:2016(E) working draft (2016). Microstructure of cast irons – Part 4: test method for determining nodularity in spheroidal graphite cast irons.
Friess J., Sonntag U., Steller I. & Bührig-Polaczek A. (2020). From Individual Graphite Assignment to an Improved Digital Image Analysis of Ductile Iron. International Journal of Metalcasting, 14(4), 1090–1104.
Mampaey F., Habets D., Plessers J. & Seutens F. (2008). The use of oxygen activity measurement to determine optimal properties of ductile iron during production. Giessereiforschung, 60(1), 2–19.
Biessmann F., Rukat T., Schmidt P., Naidu P., Schelter S., Taptunov A., Lange D. & Salinas D. (2019). DataWig: Missing Value Imputation for Tables. Journal of Machine Learning Research, 20 (175), 1–6.
Dara S. & Tumma P. (2018). Feature Extraction by Using Deep Learning: A Survey. In: 2018 Second International Conference on Electronics, Communication and Aerospace Technology (ICECA),
–31 March 2018, RVS Technical Campus, Coimbatore, India, 1795–1801.
Yogatama D. & Mann G. (2014). Efficient transfer learning method for automatic hyperparameter tuning. In: Artificial Intelligence and Statistics. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33, 1077–1085.
Weisberg S. (2005). Applied linear regression, Vol. 528, John Wiley & Sons, Hoboken.
Schulz E., Speekenbrink M. & Krause A. (2018). A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions. Journal of Mathematical Psychology, 85, 1–16. Doi: https://doi.org/10.1016/j.jmp.2018.03.001.
Li Y. (2006). Predicting materials properties and behavior using classification and regression trees. Materials Science and Engineering: A, 433(1–2), 261–268.
Ponomareva N., Radpour S., Hendry G., Haykal S., Colthurst T.,
Mitrichev P. & Grushetsky A. (2017). TF boosted trees: A scalable tensorflow based framework for gradient boosting. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, Cham, 423–427.
Ding S., Qi B. & Tan H. (2011). An overview on theory and algorithm of support vector machines. Journal of University of Electronic Science and Technology of China, 40(1), 2–10. Doi: dx.chinadoi.cn/10.3969/j.issn.1001-0548.2011.01.001.
Bhadeshia H., Dimitriu R., Forsik S., Pak J. & Ryu J. (2009). Performance of neural networks in materials science. Materials Science and Technology, 25(4), 504–510. Doi: https://doi.org/
1179/174328408X311053.
Goodfellow I., Bengio Y. & Courville A. (2016). Deep learning. MIT Press.
Schmidhuber J. (2015). Deep learning in neural networks: An overview. Neural networks, 61, 85–117. Doi: https://doi.org/
1016/j.neunet.2014.09.003.
Kim P. (2017). Convolutional neural network. In MATLAB deep learning. Apress, Berkeley, CA, 121–147.
Gulli A. & Pal S. (2017) Deep learning with Keras. Packt Publishing Ltd, Birmingham.
Dillon J.V., Langmore I., Tran D., Brevdo E., Vasudevan S., Moore D.,
Patton B., Alemi A., Hoffman M. & Saurous R.A. (2017). TensorFlow distributions. Retrieved from: https://arxiv.org/pdf/1711.
pdf [accessed 7.10.2021].
Chicco D., Warrens M.J. & Jurman G. (2021). The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science, 7, e623. Doi: https://doi.org/10.7717/peerj-cs.623.
Singh K. & Xie M. (2008). Bootstrap: a statistical method. Rutgers, The State University of New Jersey, USA. Retrieved from: http://www.stat.rutgers.edu/home/mxie/RCPapers/bootstrap. pdf, 1–14 [unpublished manuscript]
Wang F., Sahana M., Pahlevanzadeh B., Pal S.C., Shit P.K., Piran M.J.,
Janizadeh S., Bandh S.S. & Mosavi A. (2021). Applying different resampling strategies in machine learning models to predict head-cut gully erosion susceptibility. Alexandria Engineering Journal, 60(6), 5813–5829. Doi: https://doi.org/10.1016/j.aej.
04.026.
Eck D.J. (2018). Bootstrapping for multivariate linear regression models. Statistics & Probability Letters, 134, 141–149. Doi: https://doi.org/10.1016/j.spl.2017.11.001.
Su Y. & Mwanakatwe P.K. (2021). The comparison study of the model selection criteria on the Tobit regression model based on the bootstrap sample augmentation mechanisms. Journal of Statistical Computation and Simulation, 91(7), 1415–1440. Doi: https://doi.org/10.1080/00949655.2020.1856848.
Lundberg S.M. & Lee S.-I. (2017). A unified approach to interpreting model predictions. In: Proceedings of the 31st International Conference on Neural Information Processing Systems, 4768– 4777.
Lacaze J. (2017). Trace elements and graphite shape degeneracy in nodular graphite cast irons. International Journal of Metalcasting, 11(1), 44–51. Doi: https://doi.org/10.1007/s40962-016-
-6.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2021 Maximilian Brait, Eduard Koppensteiner, Gerhard Schindelbacher, Jiehua Li, Peter Schumacher
This work is licensed under a Creative Commons Attribution 4.0 International License.
