【摘要】 为了解决在测量平面线轮廓度中由于存在被测轮廓与其测量基准间存在位置误差而影响评定精度的问题,提出了一种基于遗传算法和自适应的计算平面线轮廓度误差的新方法.该方法满足最小条件原理,它利用样条插值函数拟合理论轮廓,并在评定过程中能自动地实现被测轮廓与理论轮廓之间的适应性调整,从而能够分离并消除被测轮廓与其测量基准之间的位置误差对轮廓误差评定结果的影响,在遗传优化中获得全局最优解.实例计算验证了这一结果.这种算法简单明确,具有精度高、收敛速度快、易于计算机程序实现、易于推广应用等特点.更多还原

【Abstract】 For solving the problem why position error between measured profile and its measurement reference influences evaluation precision in measuring the plane line profile error, a new method for calculating the plane line profile error based on genetic algorithm and autospotting has been proposed. It used spline function to simulate ideal profile and used genetic optimum algorithm to obtain the best optimum solution in overall. Its distinguishing characteristic is that the position of measured profile can be adjusted automatically to adapt to an ideal profile during the process of calculation. As a result, the position error between the measured profile and its measurement reference is separated and eliminated from the profile error. Example counting verified this result. This method is simple and clear, it has features of high precision and fast convergent speed and popularizing application easily and so on.更多还原