Research On Tool Path Planning Algorithm For Free-form Surface Machining

Free-form surfaces are widely used in the shipbuilding, automotive, mould, aeronautic andastronautic areas and many other industries for their special functional or aesthetic feature. Comparedwith the maturate free-form surface modeling, free-form surface machining is dropped behind for along time. As a bridge built between the free-form surface designing and machining, NC program offree-form surface is becoming more and more important with the development of multi-axismachining center. As key technology of NC program, tool path planning of free-form surfacemachining is a research hotspot and difficulty in this field.According to the research progress at home and abroad in the field of tool path planning forfree-form surface machining, key technologies of tool path planning for free-form surface machiningare studied and implemented aiming at improving the machining efficiency and quality. Severalalgorithms are proposed, including the algorithm of automatic optimum of multi-cutters combinationfor free-form surface machining, the algorithm of variable forward step with constant chord error intool path planning, the tool path planning algorithm based on constant scallop curve(CSC) forfree-form surface machining, and the algorithm of tool path planning for spline curve interpolationbased on the selection of key cutter contact (CC) points.A new algorithm of automatic optimum of multi-cutters combination is proposed. After a novelnon-linear automatic fitness function is designed according to the feature of the principal curvature,and new selection rules of the units to be copied are given, the maximum principal curvature of thefree-form surface is calculated by the new proposed improved genetic algorithm(IGA). The biggestcutter to machining the free-form surface with no curvature interference is selected according to theglobal maximum principal curvature. New interference area boundary tracing algorithm is proposed toclassify the free-form surface into different areas. Tool paths are generated in different areas withdifferent cutter, so the function can be constructed according to the total tool path length and the cuttersize. The optimal combination of multi-cutters are solved by using the genetic algorithm (GA). Themaximum of machining efficiency and avoidance of curvature interference are realized.A new algorithm of variable forward step with constant chord error in tool path planning isproposed. Calculation method of the chord error in forward step is proposed. Golden section methodis applied to calculate the maximum chord error and the parameter in every variable forward step.New method of chord error verification based on the sign of the curvature radius is presented to avoidthe wrong verification with mid-point verification method. Full description of the algorithm ofvariable forward step with constant chord error is given at last.A new kind of CSC algorithm for free-form surface machining is proposed. The constant scallopcurve is calculated according to the current CC path and the next CC path is calculated according tothe constant scallop curve. This processes are repeated until the free-form surface is covered entirelywith tool paths. Newton-Raphson iterative algorithm to calculate the projection point and the Newtoniterative algorithm to calculate the constant scallop point and the cutter location point are proposed.Calculation method of the initial angle to calculate the side step according to the curvature radius andthe cutter size is presented to improve the calculation efficiency compared that of bisection method.Full description of CSC algorithm is given at last. A new algorithm of tool path planning for spline curve interpolation based on the selection of keycutter contact (CC) points is proposed. The initial key points are selected according to the discretecurvature extrema, and the new key points are selected by the analysis of shape index. Number of thenew key points to be added is identified according to the proposed multi-point adjusting algorithmbased on the calculated deviation. The knot vectors are constructed according to the selected keypoints and the corresponding parameters, and least-square method is employed to approximate the CCpoints with least number of control points within the tolerance requirement by the new proposedalgorithm. A new kind of double NURBS curve interpolation G code is defined, and the method togenerate the double NURBS curve interpolation G code from the tool path directly is proposed.