30 m望远镜三镜是世界上最大的平面镜, 由于其复杂的功能, 又被称作大型科学可控反射镜(Giant Steering Science Mirror, GSSM)。为了更好地分析与抑制GSSM在实现光线中继功能时的抖动, 需要对抖动有精确的测量。由于GSSM抖动的要求特殊且严格, 在测量之前, 需要对抖动测量的误差进行仔细地分析与理解, 才可以更好地完成检测任务。首先, 文中针对使用编码器进行抖动测量的情况, 在不同的测量方案以及相对距离下, 通过蒙特卡洛方法可以得到由激光跟踪仪进行标定的误差: 俯仰轴线的定位误差, 在轴上测量的期望为1 μm, 小于轴外测量的3 μm; 方位轴线的定位误差在激光跟踪仪偏离两米的情况下为4.6 μm。之后对加速度计测量抖动的误差进行了考虑, 首先推导了使用功率谱方法的完备性条件; 之后使用累积功率谱对于333B32的标定结果进行处理, 得到当采样频率为2 048 Hz, 0.05 Hz以外的频段, 其精度为0.6 μm 。抖动的测量以及误差分析, 不仅是GSSM建设过程中十分重要的环节, 同时, 它可以为大口径望远镜建设提供宝贵的统计学先验知识, 并对于系统工程的发展是一种很好的推动。

Thirty meter telescope tertiary mirror is one of the largest flat mirror in this word. It is also called giant steering science mirror(GSSM). In order to better analyze and constrain the jitter of GSSM, the measurement procedure shall be carefully processed. The analysis and understanding of jitter testing was necessary, for the unique and tight requirements of jitter testing. The first section was about jitter testing with high accuracy encoder. And the error was simulated by Monte Carlo method for the profile and relative location between the tracker and the GSSM was discussed. The tilt axis location error for the co-axis case was 1 μm versus the off-axis case 3 μm. When it came to the rotational axis, the location error was 4.6 μm, and the next consideration was the accelerometers used in the jitter test. For 333B32, at 2 048 Hz sampling frequency, the error was 0.6 μm above 0.05 Hz. The jitter measurement provides useful insight for the jitter of the system, the error analysis consequently specifies the statistic property of the jitter. This will also give guide to the system engineering.