A Support-Vector-Machine Method for Precisely Evaluating Planar Straightness Error

Abstract:

To present a precise and efficient algorithm to solve planar straightness error problems, a machine-learning approach to evaluate planar straightness error was presented in this paper. According to the similarity between the least envelope zone and the support vector regression model, the SVR-based method was developed to solve the problem of straightness error. The evaluation method was compared to some existing techniques. According to the results from three datasets, it is shown that the SVR-based method can provide precise and exact values of planar straightness error.

• Periodical: Applied Mechanics and Materials (Volumes 220-223)
• Pages: 1611-1614
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.220-223.1611
• Online since: November 2012
• Authors: Jiang Meng
• Keywords: Least Envelope Zone, Straightness Error Evaluation, Support Vector Machine (SVM)

References

[1] Y. Wang. In: Winter Annual Meeting of the American Society of Mechanical Engineers, Anaheim, California (1992).

[2] G. L. Samuel, M. S. Shunmugam. Computer-Aided Design, Vol. 31, No. 3, pp.829-843 (1999)

[3] S. H. Cheraghi, G. Jiang, J. S. Ahmad. Precision Engineering, Vol. 27, pp.195-204 (2003)

[4] H. Lai, W. Jywe, C. Chen, et al. Precision Engineering, Vol.24, No. 4, pp.310-319 (2000)

[5] J. B. Gou, Y. X. Chu, Z. X. Li. Precision Engineering, Vol. 23, pp.79-93 (1999)

[6] V. Vapnik. The Nature of Statistical Learning Theory. Springer Verlag. (1995)

[7] V. Vapnik, S. Golowich, A. Smola. In: Neural Information Processing Systems, Vol. 9. MIT Press, Cambridge, MA. (1997)

[8] J. Huang. Computer-Aided Design, Vol. 35, pp.15-25 (2003)

[9] M. C. Chen, S. S. Fan. Computers & Industrial Engineering, Vol. 43, pp.437-45 (2002)

[10] J. Meng, K. An. In: International Conference on Computing, Control and Industrial Engineering, (2010)