Abstract

High-speed machining is a new trend in the manufacturing industry, especially in the mould and aerospace industries. Compared to conventional CNC machining techniques, high-speed machining has two obvious advantages: high feed rates and high part accuracy, which requires some new requirements on machining parameters. One requirement is on the material removal rate (MRR). In non-high speed machining, the MRR is not necessary to be kept constant. However, a constant MRR is a basic requisite in high-speed machining (HSM) processes, because the tool load is very sensitive to the MRR in high-speed machining. Quick tool wear, unexpected breakage, and deflection of tool occur frequently if the MRR is higher. Generally, the mould and aerospace industries have many parts including free surface or complex curves represented by NURBS (Non-Uniform Rational B-Splines). In this work, research is conducted to ensure the MRR be constant in machining 2-D NURBS curve profiles. Another peculiarity of the high-speeding machining is that its toolpath must be smooth without sharp direction changes. Overcut or gouging may occur for momentum of the tool is too big in this machining. If there are many sharp connections in linear toolpath, the feed rate cannot reach to the specified high value, because of frequent starts and stops of the servomotor, shortening the motor life. In this research, effective NURBS toolpaths are generated to machine complex curve profiles, like NURBS or B-Spline contours. In this thesis, an intelligent approach to determining the machining parameters for NURBS profiles is established. First, a 2-D geometric model of machining NURBS profiles is given. Second, a cutting force model is established based on the artificial neural-network. Third, appropriate feed rates are gotten from the predicted cutting force of machining. Then new NURBS segments are proposed to represent the previous curve in the specified tolerance. Finally, simulation and NC programs are given to demonstrate the advantages of this research