Identification Surfaces Family
In mechanical systems, the dimensional shape and the position deviations of the surfaces of the components represent very important attributes of quality assessment. This is why technical specifications include a large number of requirements regarding these attributes. At present the verification of these requirements is based on the measurement of the coordinates of the points belonging to the surface of the component. After the points of the coordinates are obtained, the numerical model of the surface is fitted. Finally, the numerical models are used to evaluate the actual dimensions of the features, to compare these dimensions with the model dimensions, and to check for the tolerances. Because of this there emerge some uncertainties regarding the dimensions, such as the distance between two planes which are not actual parallel. This is why there arises the need for grouping the component surfaces into families, for obtaining cloud point coordinates for each surface, and for the coherent modeling of a family instead of the individual modeling of each surface. The quality of the junction between two assemblies of components is given by the compatibility degree between surfaces belonging to one piece and the conjugate surfaces belonging to the other pieces which form the junction.
In this paper there are proposed two methods for geometric feature family identification (using a genetic algorithm and using neural networks) for a better evaluation of the deviations of the surfaces.
KeywordsGenetic Algorithm Point Cloud Reference System Topological Structure Support Vector Regression
- 2.Barcenas CC, Griffin PM (2001) Geometric tolerance verification using superquadrics. IIE Trans 33:1109–1120Google Scholar
- 3.Baskar N, Saravanan R, Asokan P, Prabhaharan G (2004) Ants colony algorithm approach for multi-objective optimization of surface grinding operations. Int J Adv Manuf Technol 23:311–317, DOI 10.1007 0/s 00170–002-1533–6Google Scholar
- 4.Li YY, Li LX, Wen QY, Yang YX (2006) Data fitting via chaotic ant swarm. Springer-Verlag Berlin Heidelberg, ICNC 2006, Part II, LNCS 4222, pp 180–183Google Scholar
- 8.Prakasvudhisarn C, Trafalis TB, Raman S (2003) Support vector regression for determination of minimum zone. Trans ASME 125:736–739Google Scholar
- 11.Tucker T, Kurfess T (2002) Newton methods for parametric surface registration. Part I: Theory. Elsevier Science, Comput-Aid Des 35:107–114Google Scholar
- 12.Tucker T, Kurfess T (2002) Newton methods for parametric surface registration. Part II: Experimental validation. Elsevier Science, Comput-Aid Des 35:115–120Google Scholar