【摘要】 为了能进一步提高数控磨床的磨削加工精度,运用多体系统运动学理论、神经网络方法和有限元分析 方法,分别对SMART-CNC超精密曲面数控磨床的空间几何误差、热变形误差以及力变形误差进行了建模和计 算,通过逆运动学迭代求解算法求解出可消除磨床几何误差和载荷误差的精密加工数控指令值,通过神经网络 算法所控制的微位移夹具机构,消除了磨床热变形误差的影响．实验结果表明,综合运用上述误差补偿方法后, 将该磨床的磨削误差由原来的0．225 μm减小到0．045μm,取得了比较明显的综合补偿效果．更多还原

【Abstract】 In order to enhance the grinding accuracy of CNC grinding machine, the kinematics of multi-body system, neural net and FEM are used to make geometric error, thermo error and load error prediction model. Geometric error and loading error can be reduced by use precision NC instruction which generated through an inverse kinematics solve method and the thermo error can be reduced through an special mini displacement mechanism which controlled by the thermo error prediction neural net. Experiment results show that the max size error of a high density horniness alloy reduced from 0.225 μm to 0.045 μm after use the above sythetic error compensation method.更多还原