Prediction of Surface Roughness Profiles for Milling Process with Fractal Parameters Based on LS-SVM

Abstract:

Least squares support vector machines (LS-SVM) were developed for the analysis and prediction of the relationship between the cutting conditions and the corresponding fractal parameters of machined surfaces in face milling operation. These models can help manufacturers to determine the appropriate cutting conditions, in order to achieve specific surface roughness profile geometry, and hence achieve the desired tribological performance (e.g. friction and wear) between the contacting surfaces. The input parameters of the LS-SVM are the cutting parameters: rotational speed, feed, depth of milling. The output parameters of the LS-SVM are the corresponding calculated fractal parameters: fractal dimension D and vertical scaling parameter G. The LS-SVM were utilized successfully for training and predicting the fractal parameters D and G in face milling operations. Moreover, Weierstrass-Mandelbrot（W–M ）fractal function was integrated with the LS-SVM in order to generate an artificially fractal predicted profiles at different milling conditions. The predicted profiles were found statistically similar to the actual measured profiles of test specimens and there is a relationship between the scale-independent fractal coefficients(D and G).

• Periodical: Advanced Materials Research (Volumes 97-101)
• Pages: 1186-1193
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.97-101.1186
• Online since: March 2010
• Authors: Ben Gan, Yi Jian Huang, Gui Xia Zheng
• Keywords: Fractal Parameter, Least Squares Support Vector Machines (Ls-SVM), Milling Process, Rough Surface Profile

References

[1] G.A. Stark and K.S. Moon :J. Manuf. Sci. Eng. Trans. ASME. Vol. 121(1999) , pp.251-256.

[2] E. Santner, D. Klaffke, K. Meine, Ch. Polaczyk, and D. Spaltmann : J. Wear. Vol. 261(2006), p . 101-106.

DOI: 10.1016/j.wear.2005.09.028

[3] M. Antler: J. Wear. Vol. 214 (1998). pp.1-9.

[4] S. Ge and K. Tonder: J. Tribology. Vol. 17 (1997), pp.73-80 (in Chinese).

[5] G. Chen and S. Ge: gy. Tribology. Vol. 18 n2(1998), pp.179-184 (in Chinese).

[6] H. B Liu, D.P. Wan, and D.J. Hu: J. Key. Eng. Mat. Vol. 373-374(2008), pp.762-765.

[7] G.Y. Zhou, M.C. Leu and D. Blackmore: J. Wear. Vol. 170(1993) , pp.1-14.

[8] L. He and J. Zhu: J. Wear. Vol. 208 (1997), pp.17-24.

[9] C.A. Brown and G. Savary: J. Wear. Vol. 141(1991), pp.221-226.

[10] C. Tricot, P. Ferlans and G. Baran:J. Wear. Vol. 172 (1994), pp.127-133.

[11] G. Shirong and C. Gouan: J. Wear. Vol. 231(1999), P249-255.

[12] W. Dehui: J. Computer Integrated Manufacturing Systems, CIMS. Vol. 13(2007), pp.1137-1141.

[13] D.R. Salgado , F.J. Alonso , I. Cambero and A. Marcelo : Int. J. Adv. Manuf. Technol. Vol. 43(2009), pp.40-51.

[14] H. Dong, W. Dehui, S. Haitao:C. Proc. SPIE. Int. Soc. Opt. Eng. Vol. 6280 II(2006), Third International Symposium on Precision Mechanical Measurements.

[15] I.A. El-Sonbaty, U.A. Khashaba, A.I. Selmy and A.I. Ali: J. Mater. Process. Technol. Vol. 200( 2008), pp.271-278.

[16] Z. Hua, G. Shirong, H. Xiaolong, Z. Dekuen and L. Jinlong: J. Wear. Vol. 255 (2003) pp.309-314.

[17] D.L. Jordan, R.C. Hollins and E. Jakeman: J. Wear. Vol. 109(1986), pp.127-134.

[18] A. Majumdar and C. L Tien.: J. Wear. Vol. 136(1990), pp.313-327.

[19] L. Yuan, Z. Yi and X. Xiaodong: J. Proc. SPIE. Int. Soc. Opt. Eng. Vol. 6279 (2007).

[20] J. Lundberg: Tribol. Int. Vol. 28(1995), pp.317-322.

[21] B. Bhushan: Handbook of Micro-�ano Tribology (CRC Press, USA 1999).

[22] C. Cortes and V.N. Vapnik: Support Vector networks. Machine Learning. Vol. 20(1995), p.273.

[23] V. Vapnik: Statistical Learning Theory( Wiley, New York, 1998).

[24] J.A.K. Suykens and J. Vandewalle: J. Neural Process. Letters. Vol. 9(1999), pp.293-300.

[25] M.T. Gencoglu and M. Uyar: J. Expert Sys. Appl. Vol. 36(2009), p . 10789-10798.