• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.1038-1043
• /
• 1997

To adapt the neural network proess for the purpose of determination of optimal utting onditions (optimal cutting speed and feed rate), some selection strategies for the machining factors are necessary, which is considered planning cutting process. In this case, factors that have both nonlinearity and strong relationship must be selected. Although tool grade and chemical properties of workpiece material have strong effect to cutting speed, it's not easy to find a analytic relation between them. In this paper, a mathematical method for determining the optimal amount of cutting (depth of cut, feed rate) is presented by tool goemetry and heat generation during cutting process. And various tool grade and workpiece material groups ase classified based on its chemical properties. Thier chemical composition and hardness are used as input pattern for neural network learnig. The result of learning shows the relationship between tool grade and workpiece material and it is proved that it can be used as a sub-system for automatic process planning system.