| 题名 | 高精度平面子孔径拼接算法研究 |
| 作者 | 朱鹏辉 |
| 学位类别 | 硕士 |
| 答辩日期 | 2016 |
| 授予单位 | 中国科学院上海光学精密机械研究所 |
| 导师 | 唐锋 |
| 关键词 | 大口径平面 干涉测量 子孔径拼接 累积误差 伪剪切干涉 |
| 其他题名 | Research on High Accuracy Sub-aperture Stitching Algorithm for Large Flat Optics |
| 中文摘要 | 随着现代光学技术的发展，大口径平面光学元件在极大规模集成电路制造、天文望远镜、惯性约束聚变和同步辐射等系统中的应用越来越广泛。大口径平面光学元件的面形精度直接影响光学系统的整体性能。为了实现高精度大口径平面光学元件加工与制造，需要高精度的检测手段。 子孔径拼接方法采用小口径干涉仪主机，能够有效降低检测成本，且测量口径易拓展，能够适应未来大口径平面光学元件的发展要求，但拼接检测系统固有的拼接累积误差导致拼接检测精度受限，影响了拼接检测系统的推广和应用。针对该问题，本文围绕提高拼接检测系统精度及拼接检测精度的评价开展工作。 一、对拼接累积误差产生的原因进行系统分析，提出标准镜面形局部斜率差和直流偏差是导致拼接累积误差的根本原因。推导了拼接累积误差与局部斜率差、直流偏差之间的定量关系式。基于数值仿真，对标准镜面形第4~37项Zernike像差对拼接累积误差的贡献进行了分析。每项Zernike像差在拼接重合区域存在不同的x、y方向局部斜率差和局部直流偏差，拼接方向的局部斜率差产生的拼接累积误差最大，拼接垂直方向的局部斜率差次之，直流偏差不产生拼接累积误差。并且只有二阶像差的拼接结果不存在拼接痕迹；高阶像差对拼接的影响可以转化为二阶像差对拼接的影响。 二、提出了一种减小平面子孔径拼接累积误差的新方法，该方法根据最小二乘计算的局部斜率差利用第4项和第6项Zernike像差合成一个假设的准参考镜面形，再从每个子孔径面形测量结果中减去，最后拼接合成全口径面形。通过数值仿真验证了该方法的可行性。对450mm×60mm的平面镜进行了8个子孔径的拼接检测，去除准参考镜面形前后，拼接测量结果与Zygo公司24英寸口径干涉仪检测结果的偏差从λ/7 PV减小至λ/100 PV。所拟合的准参考镜面形误差为λ/50 PV，与标准镜的面形误差为同一量级，其对每个子孔径测量结果的影响可以忽略。实验结果表明，所提出的方法能够有效控制拼接累积误差，提高拼接检测精度，并且不会导致拼接痕迹。 三、提出了一种基于伪剪切干涉方法标定大口径被测平面镜绝对面形的技术方案。采用大口径干涉仪获得被测平面镜在x、y方向平移前后的3组面形测量结果，进行差分运算得到伪剪切数据，基于最小二乘波前重建算法获得被测平面镜的绝对面形。通过数值仿真验证了算法的有效性。采用小口径干涉仪进行了原理性实验验证，实验结果表明，基于伪剪切干涉方法的平面面形绝对标定技术对干涉仪的测量重复性有很高要求，目前的实验条件很难满足。 |
| 英文摘要 | With the development of modern optical technology, large flat optics are more and more widely applied in giga scale integration, astronomical telescopes, inertial confinement fusion, synchrotron radiation and other systems. Surface figure accuracy of large flat optics directly affect the overall performance of the optical system. We need high-precision testing methods to manufacture large high performance flat mirrors. Sub-aperture stitching only needs a small-aperture standard mirror. The cost can be reduced evidently. The measured aperture can be extended easily so that this method can be able to adapt to the requirements of the development of large-aperture flats in the future. But the test accuracy is limited by the inherent accumulated error of the sub-aperture stitching system. Therefore the promotion and application of the sub-aperture stitching system is difficult. Aiming at the problem, this dissertation focuses on how to improve and evaluate the test accuracy of the sub-aperture stitching system. 1. After systematic analysis on why the accumulated error generates, it is proposed that the slope and piston difference of the reference flat in the overlapping area lead to error accumulation. The formulas between accumulated error and the slope and piston difference is derived. Based on numerical simulation, the effect of 4th to 37th Zernike aberrations on accumulated error is analyzed. Different slope and piston difference exist in the stitching overlap region of every Zernike aberration. The accumulated error because of the slope difference along the stitching direction is much larger than the accumulated error generated by the slope difference perpendicular to the stitching direction. The piston difference doesn't result in the accumulated error. There is no stitching edge only in the stitching results of second order aberrations. The effect of high order aberrations can be represented by the effect of the second order aberrations on stitching. 2. A new method to remove the accumulated error is proposed. Based on the slope difference evaluated by the least square method, the 4th and 6th Zernike aberrations are used to synthesize a virtual reference surface to represent the real figure of the reference flat. Sub-apertures with the virtual reference surface figure removed are stitched together to obtain a full aperture surface figure. This method is proved to be feasible by numerical simulation. A flat mirror with aperture of 450mm×60mm is tested by 8 sub-apertures. The deviation between the stitching result and the test result of a Zygo 24 inch interferometer is reduced from λ/7 PV to λ/100 PV after removing the virtual reference surface figure. Compared with the real reference, the effect of the virtual reference surface figure with only 0.02λ PV on sub-apertures can also be ignored. The accumulated error is controlled effectively and the measurement accuracy is evidently improved. 3. We propose to absolutely calibrate the tested large flat mirror based on pseudo-shear interferometry. Three figure results of the flat mirror before and after translating are tested by a large-aperture interferometer. The pseudo-shear data is calculated by the differential operation. The absolute shape of the tested flat mirror is evaluated by phase reconstruction algorithm based on the least square method. The algorithm is proved to be effective by numerical simulation. An experiment to testify the scheme is conducted by a small-aperture interferometer. The experimental results indicate that absolutely calibrating the flat mirror based on pseudo-shear interferometry requires high measurement repeatability. The current experimental conditions can hardly meet the requirement. |
| 语种 | 中文 |
| 内容类型 | 学位论文 |
| 源URL | [http://ir.siom.ac.cn/handle/181231/16990]  |
| 专题 | 上海光学精密机械研究所_学位论文 |
| 推荐引用方式 GB/T 7714 | 朱鹏辉. 高精度平面子孔径拼接算法研究[D]. 中国科学院上海光学精密机械研究所. 2016. |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论