Prediction of surface roughness for turning operation of aluminium alloys using artificial neural networks

Rockey Kumar, Sudipto Chaki


Present paper attempted to predict surface roughness obtained from turning process of Al 6063 aluminium alloy using different algorithms of Back Propagation Neural Network (BPNN) models. A full factorial experimentation (33) has been conducted with controllable process parameters as cutting speed (rpm), feed (mm/min) and depth of cut (mm) respectively to generate the training dataset. Average surface roughness value, Ra is considered as output parameters. 28 networks have been trained and tested using BPNN with Levenberg Marquardt (LM) algorithm and BPNN with Bayesian Regularization. Among all, 3-4-1 network trained and tested using BPNN with LM is found to produce best prediction capability with minimum value of Maximum absolute % prediction error of 4.21% and considered as best network for prediction of surface roughness.


Turning, Surface Roughness, Back Propagation Neural Network (BPNN), BPNN with Levenberg Marquardt algorithm, BPNN with Bayesian Regularization

Full Text:



Gomez-Parra, A., Puerta, F.J., Rosales, E.I., Gonzalez-Matrigal, J.M., Marcos, M., “Study of the influence of cutting parameters on the Ultimate Tensile Strength (UTS) of UNS A92024 alloy dry turned bars”, Proceedings of MESIC, 1999,pp.796-803.

.Stahl, J-E., Schultheiss, F., Hagglund. S., “Analytical and Experimental Determination of the Ra Surface Roughness during turning”,proceedings of CSI, 2011, pp.349-356.

Schubert, A. and Nestler. A., “Enhancement of Surface Integrity in Turning of Particle Reinforced Aluminium Matrix Composites by Tool Design”, 2011, proceedings of CSI, pp.300-305.

Rajasekaran, T., Palanikumar, K., Arunachalam. S.,“Investigation on the Turning parameters for surface roughness using Taguchi Analysis”, Proceedings of NUiCONE, 2012, pp.781-790.

Sathees kumar, N., Shetty, A, Shetty,A, Ananth K, Shetty,H., “Effect of spindle speed and feed rate on surface roughness of Carbon Steels in CNC turning”, proceedings of ICMOC, 2012, pp.691-697.

Mohamad S.S, Nafrizuan M.Y. and Ahmad R. Y., “Taguchi Method Approach on Effect of Lubrication Condition on Surface Roughness in Milling Operation”, Proceedings of MUCET, 2012, pp.594-599.

Wang,T, Xie, L.J., Wang, X.B., Jiao, L., Shen, J.W., Xu, H., Nie. F.M., “Surface integrity of high speed milling of Al/SiC/65p Aluminium matrix composite”,14th CIRP Conference on Modeling of Machining Operations (CIRP CMMO), 2013, pp.475-480.

Hornik, K, Tinchcombe, M & White, H,“Multilayer Feed Forward Networks are Universal Approximators”, IEEE Transactions on Neural Networks, 1989, 2, pp.359-366.

Hagan, MT & Menhaj, MB, ‘Training feed forward networks with the Marquardt Algorithm’, IEEE Transactions on Neural Networks, 1994, 5(6), pp.989–993.

Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

                                                                                                                                       ISSN: 2319-5606

                               Copyright © Blue Ocean Publication ( All Rights Reserved.