Intelligent Optimization for Sculptured Surface CNC Tool-paths

This paper suggests the application of genetic algorithms for the intelligent generation of optimum sculptured surface CNC machining tool-paths. Two robust full quadratic mathematical models are developed relating the physical relation among machining surface deviation and resulting cutting time; quality objectives which are treated as conflicting ones. The independent variables are the tool inclination angles -lead and tilt- in the case of 5-axis machining and step-over engagement among subsequent XY passes; using a toroidal cutter. A Box-Behnken response surface design was established to prepare and conduct simulation experiments in a cutting-edge manufacturing software using a benchmark multivariable sculptured surface and a special multi-axis tool-path strategy. The genetic algorithm utilizes both models expressed as a common Pareto-based fitness function so that multi-objective optimization is achieved, yet; arriving at one optimum solution to ease the efforts of end-users and numerical control programmers. The methodology is validated by utilizing the genetic algorithm's recommendations for the settings of the machining parameters and the optimum tool-path simulation is performed to verify the operation.