自适应光学测试与系统优化研究

发布时间：2019-02-23 12:44

【摘要】：自适应光学(Adaptive Optics,AO)是一种实时探测并补偿波前畸变的技术,在天文成像、激光光束控制、自由空间激光通信、视网膜成像、显微成像、空间光学等领域具有广泛的应用。尽管自适应光学在很多领域已经取得了很大的成功,然而自适应光学在控制方法、非共光路像差校正等方面仍然存在着可以进一步提高的地方。开展自适应光学测试与系统优化的研究,对测量、分析和提高自适应光学系统的性能具有重要意义。自适应光学测试与系统优化涵盖的内容很多,本文结合工程应用中的实际需求,选择波前处理器测试方法、自适应光学系统性能测试方法、利用波前处理器校正非共光路像差的方法,以及评价与改善区域重构方法的闭环控制稳定性等四项内容进行研究,开展了以下研究工作:1、介绍了自适应光学原理和自适应光学系统的组成;总结并分析了自适应光学的重构算法;阐明了自适应光学系统中的误差来源和抑制误差的方法;推导并实现了一种基于离散傅里叶变换的Zernike多项式快速算法。2、对于基于FPGA的波前处理器硬件调试过程中的中间计算结果不易监测、调试比较困难的问题,设计并实现了一种利用上位机软件实现波前处理器测试的方案。这种方法能够分步测试波前处理器的中间计算结果,也能够避免硬件调试中的不当输出对精密的波前校正器带来损害。首先,对波前处理器的功能、组成和工作流程进行了分析,确定了波前处理器的测试步骤。然后,通过判断波前处理器的上传数据来测试波前处理器的工作模式、系统参数是否正确设置;通过比较软件计算值和波前处理器上传值来测试波前处理器的波前斜率计算过程、波前重构过程和波前控制过程。最后,将测试方法应用于97单元自适应光学波前处理器的测试,结果表明,波前处理器的硬件调试效率得到较大提高。经过测试后的波前处理器能够在自适应光学实验系统中正常工作,进行连续校正后系统的残余波前像差的rms和pv分别为0.034波长和0.392波长。3、对于基于fpga的波前处理器不易校正非共光路像差的问题,提出了一种适合波前处理器的非共光路像差校正方法。首先,讨论了非共光路像差的产生原因和运用相位差异技术检测非共光路像差的方法。然后,根据波前处理器的工作流程,推导了将非共光路像差折算到shack-hartmann波前探测器(sh-wfs)参考点偏移量的算法,编写了实现算法的主控计算机软件模块。最后,在望远镜光路中利用光源为目标开展实验,采用这种方法进行非共光路像差校正后,目标能量集中度提高了17.6%,证明了这种方法的可行性。4、为了满足自适应光学系统性能测试的需求,采用校正后的白光光纤光源的图像计算strehlratio(sr)并将sr作为性能测试的评价指标,设计并实现了在一种实验室内对自适应光学系统进行性能测试的方法。该方法采用光学传递函数积分法计算sr,避免了sr计算公式中测量目标与理想目标能量不一致的问题;为了模拟影响自适应光学系统校正效果的实际因素,该方法通过改变光源的亮度模拟不同的sh-wfs图像信噪比,分别通过改变湍流模拟器的位置和转速模拟不同的fried大气相干长度和greenwood频率。最后,将本文设计的性能测试方法用于97单元自适应光学实验系统性能测试。性能测试结果表明,对于中等的sh-wfs图像信噪比,自适应光学实验系统可以在fried大气相干长度大于5cm而且greenwood频率低于60hz的条件下较好地进行闭环校正。5、对于区域重构法的闭环控制稳定性劣于模式重构法的问题,采用lyapunov稳定性理论对稳定性的评价与改善开展了研究。基于lyapunov稳定性理论的误差传播因子能够涵盖积分增益和响应矩阵两方面对稳定性的影响,因此,采用误差传播因子作为闭环控制稳定性的评价判据。采用southwell对应方法确定sh-wfs子孔径和变形镜促动器之间的位置对应关系来避免fried对应方法可能引起的waffle模式,提高稳定性。同时,采用响应矩阵的奇异值滤波方法来提高稳定性,并提出采用误差传播因子评价响应矩阵的奇异值滤波对稳定性的影响。通过计算误差传播因子,分析了不同的积分增益和滤除响应矩阵不同个数的奇异值对闭环控制稳定性的影响。分析结果证明响应矩阵的奇异值滤波可以改善闭环控制的稳定性,分析结果也说明当积分增益很高时可以通过适当多滤除一些响应矩阵的奇异值来保持稳定性。最后,通过实验测量了滤除响应矩阵的13个最小的奇异值之后的系统性能,测量结果表明自适应光学系统的波前校正能力仍然能够比较好地发挥出来,滤除响应矩阵的小部分奇异值对系统性能影响不大。
[Abstract]:Adaptive Optics (AO) is a technique for real-time detection and compensation of pre-wave distortion. It is widely used in the fields of astronomical imaging, laser beam control, free space laser communication, retinal imaging, microimaging, and space optics. Although the adaptive optics have made great success in many fields, the adaptive optics still have a further improvement in the control method, the non-co-optical path aberration correction, and the like. The research of self-adaptive optics test and system optimization is of great significance for measuring, analyzing and improving the performance of the adaptive optical system. The self-adaptive optical test and system optimization cover a lot of content. In this paper, the method of pre-processor test method, self-adaptive optical system performance test method, and the method of correcting non-co-optical path aberration using the pre-wave processor are selected based on the actual demand in the engineering application. The following research work is carried out: 1. The composition of the adaptive optics and the adaptive optics system is introduced, and the self-adaptive optics reconstruction algorithm is summarized and analyzed. A fast algorithm of Zernike polynomial based on discrete Fourier transform is derived and a fast algorithm of Zernike polynomial based on discrete Fourier transform is derived and a fast algorithm of Zernike polynomial based on discrete Fourier transform is derived. In order to solve the problem of difficult debugging, a scheme is designed and implemented to realize the pre-wave processor test by using the upper computer software. The method can test the middle calculation result of the wave front processor in step, and can avoid the damage to the precision wave front corrector due to improper output in the hardware debugging. First, the function, composition and workflow of the pre-wave processor are analyzed, and the test procedure of the pre-wave processor is determined. then, the working mode of the pre-wave processor is tested by judging the uploading data of the pre-wave processor, the system parameter is set correctly, the pre-wave slope calculation process of the pre-wave processor is tested by comparing the software calculation value and the pre-wave processor upload value, The pre-wave reconstruction process and the pre-wave control process. Finally, the test method is applied to the test of the 97-unit adaptive optical pre-processor, and the result shows that the hardware debugging efficiency of the pre-wave processor is greatly improved. after the pre-tested wave front processor can work normally in the self-adaptive optical experimental system, the rms and pv of the residual wave front aberration of the system after continuous correction are respectively 0.034 wavelength and 0.392 wavelength. A non-co-optical path aberration correction method suitable for a pre-wave processor is proposed. First, the cause of the non-co-optical path aberration and the method of using the phase difference technique to detect the aberration of the non-co-optical path are discussed. Then, according to the working flow of the wave front processor, the algorithm of converting the non-co-optical path image difference to the reference point offset of the shack-hartmann wave front detector (sh-wfs) is derived, and a main control computer software module for implementing the algorithm is developed. in that end, the experiment is carried out using the light source in the optical path of the telescope, and after the aberration correction of the non-co-optical path is carried out by adopting the method, the target energy concentration degree is increased by 1.7%, the feasibility of the method is proved, and 4, in order to meet the requirement of the performance test of the adaptive optical system, In this paper, the image of the corrected white-light-fiber light source is used to calculate the flow ratio (sr) and the sr is used as the evaluation index of the performance test, and a method for performing the performance test on the self-adaptive optical system in a laboratory is designed and implemented. the method adopts an optical transfer function integration method to calculate the sr, so that the problem that the measurement target is not consistent with the ideal target energy in the sr calculation formula is avoided; in order to simulate the actual factors which influence the correction effect of the adaptive optical system, According to the method, different sh-wfs image signal-to-noise ratios can be simulated by changing the brightness of the light source, and the length and the greenwood frequency of different fried air are simulated by changing the position and the rotation speed of the turbulence simulator respectively. Finally, the performance test method designed in this paper is used in the performance test of the 97-unit self-adaptive optical experimental system. the performance test results show that for medium sh-wfs image signal-to-noise ratio, the self-adaptive optical experimental system can better carry out the closed-loop correction under the condition that the length of the fried atmosphere is more than 5cm and the greenwood frequency is lower than 60hz. The stability of the closed-loop control of the region reconstruction method is inferior to that of the mode reconstruction method, and the stability evaluation and improvement are studied by using the lyapunov stability theory. The error propagation factor based on the lyapunov stability theory can cover the effect of the integral gain and the response matrix on the stability. Therefore, the error propagation factor is used as the evaluation criterion of closed-loop control stability. and the corresponding relation between the sh-wfs sub-aperture and the deformation mirror actuator is determined by using the soutwell corresponding method to avoid the waffle mode which can be caused by the fried corresponding method and improve the stability. At the same time, the singular value filtering method of the response matrix is adopted to improve the stability, and the influence of the singular value filtering on the stability of the response matrix by the error propagation factor is proposed. By calculating the error propagation factor, the effect of different integral gain and the different number of singular values of the response matrix on the stability of closed-loop control is analyzed. The results of the analysis show that the singular value filtering of the response matrix can improve the stability of closed-loop control, and the analysis results show that the stability can be maintained by appropriately filtering the singular values of some of the response matrices when the integral gain is high. Finally, the system performance after 13 minimum singular values of the response matrix is measured by the experiment, and the measurement results show that the pre-wave correction capability of the adaptive optical system can still be better, and the small part singular value of the filter response matrix is not affected by the system performance.
【学位授予单位】：中国科学院研究生院(长春光学精密机械与物理研究所)
【学位级别】：博士
【学位授予年份】：2016
【分类号】：O439

【相似文献】