This paper describes the development of a neural network system for grinding wheel selection. The system employs a back-propagation network with one hidden layer and was trained using data from reference handbooks. It is shown that a neural network is capable of learning the relationship between the wheel and the grinding process without a requirement for rules or equations. It was further found that a relatively small number of training examples allows the system to produce reliable recommendations for a much greater number of combinations of grinding conditions. The system was developed on a PC using the C++ programming language.
NakajimaT.A new standard for proper selection of grinding wheels in plunge grinding operation. Ann. CIRP, 1978, 27(1), 249–253.
2.
Machinability Data Centre. Machining Data Handbook, 3rd edition, Vol. 2, 1980 (Metcut Research Associates Inc., USA).
3.
KingR. I.HahnR. S.Handbook of Modern Grinding Technology, 1986 (Chapman and Hall, New York).
4.
MidhaP. S.ZhuC. B.TrmalG. J.An expert system for wheel selection for cylindrical grinding operations. In Proceedings of Sixth International Conference on Computer Aided Production Engineering, London, 1990, pp. 445–450.
5.
ZhuC. B.MidhaP. S.TrmalG. J.Development of knowledge base abrasive wheel selection system using a proprietary expert system shell. In Proceedings of Fourth International Congress on Condition Monitoring and Diagnostic Engineering Manufacturing, Selis, France, 1992, pp. 318–325.
6.
Grinding Data Book, 1992 (Universal Grinding Wheel Company Limited, UK).
