ANN surface roughness prediction of AZ91D magnesium alloys in the turning process

Berat Barış Buldum; Aydın Şık; Ali Akdağlı; Mustafa Berkan Biçer; Kemal Aldaş; İskender Özkul

doi:10.3139/120.111088

Article

ANN surface roughness prediction of AZ91D magnesium alloys in the turning process

Berat Barış Buldum , Aydın Şık , Ali Akdağlı , Mustafa Berkan Biçer , Kemal Aldaş and İskender Özkul

Published/Copyright: October 2, 2017

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Materials Testing Volume 59 Issue 10

Abstract

This contribution presents an approach for the modeling and prediction of surface roughness in the turning of AZ91D magnesium alloys using an artificial neural network. The experiments were conducted with CCGT, DCGT and VCGT cutting tools under minimum quantity lubrication and dry machining conditions. AZ91D alloys were machined at different cutting speeds and feed rates, and the depth of cut was kept constant. 15 out of 18 experimental data points were used for the training of the artificial neural network model and the remaining 3 were used for the testing process. The average percentage error was calculated as 0.000815 % and 0.663 % for training and testing, respectively. The model and target results were found to have extremely low error rates.

Kurzfassung

Der vorliegende Beitrag stellt einen Ansatz zur Modellierung und Vorhersage der Oberflächenrauheit beim Drehen der Magnesiumlegierung AZ91D vor, wobei künstliche neuronale Netze (KNN – engl: Artificial Neural Network – ANN) zur Anwendung kamen. Die Experimente wurden mit Schneidwerkzeugen des Typs CCGT, DCGT und VCGT unter geringstem Schmiermittelzusatz und trockenen Bearbeitungsbedingungen durchgeführt. Die Legierung AZ91D wurde bei verschiedenen Schnittgeschwindigkeiten und Vorschubraten bei einer konstanten Schnitttiefe bearbeitet. Von den insgesamt 18 experimentellen Datenpunkten wurden 15 verwendet, um das KNN zu trainieren. Die restlichen drei verblieben zur Testüberprüfung. Der durchschnittliche prozentuale Fehler wurde mit 0.000815 % für das Trainieren bzw. mit 0.663 % für die Überprüfung berechnet. Es zeigte sich, dass die Ergebnisse extrem niedrige Fehlerraten aufwiesen.

*Correspondence Address, Assistant Prof. Dr. Iskender Özkul, Mechanical Engineering Department, Mersin University, Mersin 33343, Turkey, E-mail: iskender@mersin.edu.tr

Assistant Prof. Dr. Berat Barış Buldum obtained his PhD degree from the Institute of Science, Gazi University, Ankara, Turkey in 2013. His research interests are machinability of metals, light-metal cutting and design as well as construction. He is currently working at Mersin University, Department of Mechanical Engineering, Turkey.

Prof. Dr. Aydin Şık completed his Bachelor degree in the Faculty of Industrial Technical Education, Gazi University, Ankara, Turkey, his Master and PhD degrees at Institute of Science, Gazi University in 2002. His research interests are welding, machinability of metals, design and construction. Currently, he is working in the Department of Industrial Design of Gazi University.

Prof. Dr. Ali Akdağlı received his B. S., M. S. and PhD degrees in Electronic Engineering from Erciyes University, Kayseri, Turkey in 1995, 1997 and 2002, respectively. From 2003 to 2006, he was Assistant Professor in the Electronic Engineering Department at Erciyes University. He joined the same department at Mersin University, Turkey, where he is currently working as Professor. He has published more than 90 papers in journals and conference proceedings. His current research interests include evolutionary optimization techniques (genetic algorithm, ant colony optimization, differential evolution, particle swarm optimization and artificial bee colony algorithms), artificial neural networks and their applications to electromagnetic, wireless communication systems, microwave circuits, microstrip antennas and antenna pattern synthesis problems. He is an editorial board member of “Recent Patents on Electrical Engineering”, “International Journal of Computers” and “Journal of Computational Engineering”.

Research Assistant Mustafa Berkan Biçer completed his Bachelor Degree in Electrical-Electronic Engineering at Fırat University, Elazığ, Turkey, and his Master Degree at the Institute of Science, Mersin University, Turkey in 2012. His research interests are mathematical modeling. Currently, he is working in the department of Electrical and Electronic Engineering of Mersin University.

Prof. Kemal Aldaş is Professor in the Engineering Faculty of Aksaray University, Aksaray, Turkey. He obtained his PhD degree in Mechanical Engineering from Selçuk University, Konya, Turkey, in 1998. His research areas cover a wide range of mathematical modeling, materials sciences and fluid mechanics.

Assistant Prof. İskender Özkul obtained his PhD degree from Institute of Science, Aksaray University, Aksaray, Turkey in 2016. His research interests are materials science and mathematically modeling. He is currently working at Mersin University, Department of Mechanical Engineering in Mersin, Turkey.

References

1 H.Oktem, T.Erzurumlu, F.Erzincanli: Prediction of minimum surface roughness in end milling mold parts using neural network and genetic algorithm, Materials & Design27 (2006), No. 9, pp. 735–74410.1016/j.matdes.2005.01.010Search in Google Scholar

2 S. K.Pal, D.Chakraborty: Surface roughness prediction in turning using artificial neural network, Neural Computing & Applications14 (2005), No. 4, pp. 319–32410.1007/s00521-005-0468-xSearch in Google Scholar

3 B. S.Patel, M. H.Pal: Optimization of machining parameters for surface roughness in milling operation, International Journal of Applied Engineering Research7 (2012), No. 11, pp. 2129–2133Search in Google Scholar

4 M.Nalbant, H.Gökkaya, İ.Toktaş, G.Sur: The experimental investigation of the effects of uncoated, PVD- and CVD-coated cemented carbide inserts and cutting parameters on surface roughness in CNC turning and its prediction using artificial neural networks, Robotics and Computer-Integrated Manufacturing25 (2009), No. 1, pp. 211–22310.1016/j.rcim.2007.11.004Search in Google Scholar

5 K.Aldaş, İ.Ozkul, M.Eskil: Prediction of surface roughness in longitudinal turning process by a genetic learning algorithm, Materials Testing56 (2014), No. 5, pp. 375–38010.3139/120.110570Search in Google Scholar

6 K.Aldas, I.Ozkul, A.Akkurt: Modelling surface roughness in WEDM process using ANFIS method, Journal of the Balkan Tribological Association20 (2014), No. 4, pp. 548–558Search in Google Scholar

7 K.Aldaş, I.Ozkul, A.Taşkesen, Y.Kayır: Investigation of drilling parameters on thrust force on AZ91 magnesium alloy by genetic expression programming, Düzce University Journal of Science & Technology2 (2014), No. 1, pp. 169–177 (in Turkish)Search in Google Scholar

8 N.Deshpande, M.Fofana: Nonlinear regenerative chatter in turning, Robotics and Computer-Integrated Manufacturing17 (2001), No. 1, pp. 107–11210.1016/S0736-5845(00)00043-0Search in Google Scholar

9 S. M.Ali, N. R.Dhar: Modeling of tool wear and surface roughness under MQL condition – A neural approach, Canadian Journal on Artificial Intelligence, Machine Learning & Pattern Recognition1 (2010), No. 2, pp. 7–25Search in Google Scholar

10 L.Dobrzański, M.Król, T.Tański: Application a neural networks in crystallization process of Mg-Al-Zn alloys, Archives of Computational Materials Science and Surface Engineering149 (2010), pp. 149–156Search in Google Scholar

11 K. U.Kainer: Magnesium-Alloys and Technology, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany (2003) 10.1002/3527602046.ch2Search in Google Scholar

12 G.Padmanaban, V.Balasubramanian: Optimization of laser beam welding process parameters to attain maximum tensile strength in AZ31B magnesium alloy, Optics & Laser Technology42 (2010), No. 8, pp. 1253–126010.1016/j.optlastec.2010.03.019Search in Google Scholar

13 S.Fintová, L.Kunz: Fatigue properties of magnesium alloy AZ91 processed by severe plastic deformation, Journal of the Mechanical Behavior of Biomedical Materials42 (2015), pp. 219–22810.1016/j.jmbbm.2014.11.019Search in Google Scholar PubMed

14 R.Lucci, R. L.Padilla, S.Cantero, R.Bariles, C.Oldani: Refining of AZ91 magnesium alloy obtained in machining chips recycling, Procedia Materials Science8 (2015), pp. 886–89310.1016/j.mspro.2015.04.149Search in Google Scholar

15 A.Boby, A.Srinivasan, U. T. S.Pillai, B. C.Pai: Mechanical characterization and corrosion behavior of newly designed Sn and Y added AZ91 alloy, Materials & Design88 (2015), pp. 871–87910.1016/j.matdes.2015.09.010Search in Google Scholar

16 T.Yu-Bo, Z.Su-Ling, L.Jing-Yi: Modeling resonant frequency of microstrip antenna based on neural network ensemble, International Journal of Numerical Modelling: Electronic Networks, Devices and Fields24 (2011), No. 1, pp. 78–8810.1002/jnm.761Search in Google Scholar

17 A.Kayabasi, A.Akdagli, D.Pham: A novel method of support vector machine to compute the resonant frequency of annular ring compact microstrip antennas, Cogent Engineering2 (2015), No. 1, pp. 1–1410.1080/23311916.2014.981944Search in Google Scholar

18 J. P.Davim, V.Gaitonde, S.Karnik: Investigations into the effect of cutting conditions on surface roughness in turning of free machining steel by ANN models, Journal of Materials Processing Technology205 (2008), No. 1, pp. 16–2310.1016/j.jmatprotec.2007.11.082Search in Google Scholar

Published Online: 2017-10-02

Published in Print: 2017-10-04

You are currently not able to access this content.

Articles in the same Issue

https://doi.org/10.3139/120.111088