论文部分内容阅读
为了寻找一种适用于长程面形仪(Long Trace Profiler,LTP)测量数据的曲线拟合算法,提高同步辐射用光学反射镜表面几何参数拟合的准确性,并合理可靠地评价同步辐射用光学反射镜表面质量,本文对镜面曲率半径、测量环境噪声和镜面测量姿态这三个可能影响镜面几何参数拟合精度的因素进行了数值模拟,采用基于代数距离的最小二乘法、基于几何距离的最小二乘法和遗传算法对模拟面形进行了拟合对比。结果表明,基于几何距离的最小二乘法对以上三个因素均不敏感,最适合长程面形仪测量数据的拟合。该方法能够给上海同步辐射光源反射镜的面形检测提供一定的参考价值。
In order to find a curve fitting algorithm suitable for long trace profiler (LTP) measurement data, the accuracy of surface geometry parameters fitting for synchrotron radiation mirrors is improved and a reasonable and reliable evaluation of synchrotron radiation In this paper, three factors that may influence the fitting accuracy of mirror geometry parameters, such as mirror radius of curvature, measurement ambient noise and mirror measurement attitude, are numerically simulated. The least square method based on algebraic distance is used, and the minimum based on geometric distance Square method and genetic algorithm to simulate the shape of the comparison and comparison. The results show that the least squares method based on geometric distance is not sensitive to the above three factors and is most suitable for the fitting of the long profile profilometry data. The method can provide certain reference value to the surface shape detection of Shanghai synchrotron radiation light source reflector.