论文信息 - Plastic Boss Design Using Knowledge Extraction Method

Plastic Boss Design Using Knowledge Extraction Method

Complex product design often relies on design experience and numerous repeated equipment operations. In order to speed up the design process on a complicated product, a methodology of knowledge extraction (KE) is proposed. Moreover, a case study in designing a shaped boss using the KE technique in conjunction with a Back-Propagation Network (BPN) method as well as a genetic algorithm (GA) will be introduced. The results indicate that the prediction error between learned and examined data is found to be within 8%. Moreover, the error between the GA’s solution and the specific target is also found to be within 5%. Therefore, the bi-directional prediction scheme constructed in this project is deemed to be reliable. Consequently, knowledge extraction can provide a rapid and economical way to design and shape a complicated product.

[1] D. Parker,et al. Separable helper factors support B cell proliferation and maturation to Ig secretion. , 1982, Journal of immunology.

[2] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[3] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[4] Kenneth Alan De Jong,et al. An analysis of the behavior of a class of genetic adaptive systems. , 1975 .

[5] Paul J. Werbos,et al. Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[6] A. Patrikar,et al. Nonlinear system identification and adaptive control using polynomial networks , 1996 .

[7] Tadashi Kondo. The learning algorithms of the GMDH neural network and their application to the medical image recognition , 1998, Proceedings of the 37th SICE Annual Conference. International Session Papers.