| 其他摘要 | The disturbance of the earth's atmosphere to the astronomical and space observations is the main problem that the ground-based telescope can not avoid. Atmosphere turbulence causes the reduction of the angular resolution of the telescope. When telescope aperture greater than the seeing disk, the angular resolution of the telescope is mainly limited by atmospheric turbulence. There are many high-resolution observation techniques have been developed to eliminate or reduce the effects of atmospheric turbulence disturbance. Adaptive optics technology can effectively decrease the interference of atmospheric turbulence. The traditional adaptive optics system’s correction FOV (field of view) is limited by the isoplanatic angle. MCAO (multi-conjugate adaptive optics) technology can perform with wide correction FOV. MCAO's design and optimization of performance need to know the characteristics of atmospheric turbulence profiles. The next-generation large aperture telescope must be equipped with a wide-field adaptive optics system in order to fully realize its performance of large aperture. Therefore, it is of great significance to measure the atmospheric turbulence profiles of the site. At present, there are many methods for measuring turbulence profiles. Among them, the methods (such as PML(profiler of moon limb) and S-DIMM+(solar differential Image motion monitor plus)) based on variance or covariance of image motion of extended object have the advantages of high altitude resolution and high measurement accuracy. These methods are indirect measurement methods. When measure turbulence profiles, it can only get limited number of data to calculate variance or covariance of image motion data. It is necessary to convert the integral equation into a discrete linear equation. Then, solve the equation to get the turbulence profiles by the optimization method. The measurement error has a great influence on the inversion accuracy of the turbulence profiles. In this paper, some methods have been proposed for improving the profiles measurement accuracy and anti-noise performance. Firstly, the inversion method of turbulent profiles was studied. In this paper, the turbulence profiles inversion method based on the simulated annealing algorithm is adopted. In order to improve the accuracy of inversion, a inversion method with constraint based on simulated annealing algorithm was proposed. Smooth constraint were added according to the natural characteristics of turbulence profiles. The smooth constraint significantly improved the inversion accuracy of turbulence profiles.Secondly, a wide FOV turbulence profiles measurement method is proposed. In the process of turbulence profiles inversion, it is found that increasing the amount of data and increasing the number of linear equations can improve the inversion accuracy of turbulence profiles and the anti-noise performance. Increasing the FOV is a more effective way to increase linear equations. However, the layer sample coefficient matrix has a large discretization error, especially in the large FOV, which limits the increase of the observation FOV. By using the layer integral coefficient matrix, the discretization error can be effectively reduced, and the observation FOV can be increased, thereby improving the measurement accuracy of the turbulence profiles measurement method and the anti-noise performance of the measurement error, noise, and so on.Then, The ill-conditioned characteristic of turbulence profiles measurement method of differential image motion covariance has been analyzed. Increasing the subapertures baseline can reduce the seriousness of ill-conditioned problem, but it also increases the influence of measurement error. Therefore, increasing the baseline cannot improve the measurement accuracy of the turbulence profiles. Increasing the number of subapertures is another effective method to increase the measurement data and linear equations number, which can improve measurement accuracy. Finally, the noise analysis is performed for the single aperture limb motion turbulence profiles measurement method. It is found that the noise has a great influence on this method. To reduce the influence of noise, a noise separation method is proposed. The noise is separated by adding noise separation terms. The result of simulation experiments shows that the noise separation method improves measurement accuracy and reduces the influence of noise. |
修改评论