Research On The Interferometric Detection Technology For The Surface Shape Of The Primary Mirror Of The Segmented Telescope

With the intensive research on frontier scientific problems such as astronomical observation and cosmic exploration,the human requirements for the performance of astronomical telescopes have gradually increased,including the telescope’s resolution and light gathering capability,which requires an increasing aperture of the telescope’s primary mirror.The development of traditional monolithic telescope’s primary mirror is limited by processing,cost,transportation and other factors,that make it difficult to break through the 8m magnitude,therefore segmented telescope came into being.This has two problems,firstly,the optical performance of the segmented telescope is affected by the accuracy of the individual submirrors,which requires high-precision surface shape measurement technology to ensure the surface shape accuracy of individual sub-mirrors.Secondly,after the accuracy of the submirrors is achieved,if there are co-phase errors in the segmented primary mirror,it will lead to a serious degradation of the imaging quality of the telescope system.Therefore,it is also necessary to ensure that the segmented primary mirror has a continuous and consistent phase.Based on the above problems,this paper carries out research on the interference detection technology of the primary mirror surface shape of the segmented telescope,and the main work of the article is as follows.1.For the influence of the reference surface shape error on the detection accuracy,the absolute shape of the measured surface is obtained by the shift-rotation method.For the problem that the Zernike polynomials do not have orthogonal completeness on non-circular domain,orthogonal polynomials on non-circular domain are obtained by recursive iterations of the Gram-Schmidt orthogonalization method,thereby characterizing and analyzing the surface shape.For the problem that it is difficult to achieve a balance between measurement accuracy,efficiency and cost,an absolute measurement method for irregular aperture surfaces based on deep learning is proposed.Using deep learning technology to achieve fast pixel-level accuracy prediction of the shift and rotation of the measured surface.Combined with the shift-rotation method,the absolute detection of the measured surface shape with fast,high accuracy and low cost is realized.Simulation and experimental results prove the correctness of the method,which solves to some extent the problems of long measurement cycles,low efficiency and expensive precision translation and rotation stage of traditional methods.2.For the problem of co-phase detection of segmented mirrors,the co-phase error measurement methods based on vortex beam interference are proposed,one based on vortex beam and spherical beam interference,and the other based on vortex beam and plane beam interference.Studied and derived the mathematical relationship between the co-phase error(piston and tip/tilt)and phase characteristics,and verified the correctness of the relationship by simulations in MATLAB and OpticStudio.A dual-wavelength algorithm is applied to expand the measurement range of the piston error.Experiments of segmented mirrors co-phase error measurement were conducted,which are quantitative measurement experiments based on the spatial light modulator and qualitative measurement experiments based on designed and processed segmented mirrors.The experimental results verify the validity of both methods.The methods enrich the technique of co-phase detection of segmented primary mirrors and provide corresponding theoretical reference and support.