Research on Real Tooth Surface Deviation Correction of Spiral Bevel Gear Based on Truncated Singular Value Decomposition
According to structure of traditional milling machine, the tooth surface error identification model of forming processing spiral bevel gear is established by the 4×4 Denavit-Hartenberg homogeneous transformation matrix, the gear meshing theory and so on. In view of the present commonly used least squares method’s solution defects, the truncated singular value decomposition (TSVD) and the L curve method are proposed to solve the identification model and the solution method accuracy was verified by the example and experiment. The results show that the gear concave average error is at 0.01498mm before correction, and the average error will drop down to 0.00084mm after using this method correction, and the error corrects 94.4%. The gear convex average error is at 0.00846mm before correction and the average error will drop down to 0.00176mm after using this method correction and the error corrects 79.2%.
Daoguo Yang, Tianlong Gu, Huaiying Zhou, Jianmin Zeng and Zhengyi Jiang
S. H. Chen et al., "Research on Real Tooth Surface Deviation Correction of Spiral Bevel Gear Based on Truncated Singular Value Decomposition", Advanced Materials Research, Vols. 201-203, pp. 1562-1568, 2011