CN106596057A - Surface shape inspection method of large-aperture reflector assembly - Google Patents

Abstract

The invention relates to a method for inspecting the surface shape of a large-diameter reflector assembly, which belongs to the technical field of optical precision measurement. The mirror in the mirror assembly faces upward, and the surface shape data W _0(m×n) of the mirror in the mirror assembly is measured by an interferometer, and the mirror in the mirror assembly faces downward, and the mirror is measured by an interferometer The surface shape data W _180(i×j) of the mirror in the component; W _0(m×n) and step W _180(i×j) are subjected to surface shape data matrix normalization processing to obtain F _{0(a×b )} =W _a×b +G _0(a×b) +K _a×b and F _180(a×b) =W _a×b +G _180(a×b) +K _a×b , normalized The processed surface shape data F _0(a×b) and F _180(a×b) are image superimposed to eliminate the gravity deformation, and obtain the zero-gravity surface shape data of the mirror F=(F _0(a×b) +F _{180 (a×b)} )/2=W _a×b +K _a×b , if the root mean square value of the difference between F and W _a×b K _a×b is not greater than 0.003λ (λ=632.8nm), it is considered The assembly of the mirror assembly meets the requirements. If the root mean square value of the difference between F and W _a×b K _a×b is greater than 0.003λ, it is considered that the assembly of the mirror assembly does not meet the requirements, and the mirror assembly needs to be reassembled .

Description

A Surface Shape Inspection Method for Large-Aperture Mirror Components

technical field

The invention relates to a method for inspecting the surface shape of a large-diameter reflector assembly, which belongs to the technical field of optical precision measurement. The method is used to detect whether the assembly of the reflector assembly meets the requirements. The large-diameter refers to the diameter of the reflector is Ф800mm -Ф1200mm.

Background technique

The bipod discrete support structure of large-aperture optical parts has the advantages of simple assembly and simple mechanical boundary conditions, but the disadvantage is that the optical components deform greatly in the gravitational environment, and the wavefront shape detection results of the final optical components often include the original components of the optical components. Surface shape, gravity deformation, assembly stress deformation and other factors interfere with the fusion results, and due to the large gravity deformation error, the assembly stress deformation error will be submerged in the gravity error, resulting in the inability to accurately judge the assembly accuracy.

Since most optical remote sensors are installed and adjusted on the ground and used on-orbit, the test error caused by gravity will automatically disappear after the on-orbit. Therefore, when the optical remote sensor is installed and adjusted on the ground, the key is to distinguish the assembly stress error from the gravity error, and strictly control the assembly stress error to control its influence within the technical requirements.

Aiming at the problem of removing the gravity error of large-aperture reflectors, domestic scholars have proposed a method for unloading large-aperture space reflectors in the bare mirror stage. The gravity of the large-aperture space reflector is unloaded through a special support structure, and the deformation of the reflector introduced by the support force is strictly controlled. To meet the requirements of high-precision surface shape detection of mirrors, for details, please refer to the article "Research on Supporting Technology for High-precision Surface Shape Detection of Large-Aperture Space Mirrors" in "Applied Optics" in 2011, or "Space "Research on Lightweight High-Resolution Camera Offloading Technology" in the Proceedings of the Sixteenth Academic Conference of the Probing Professional Committee. It can be seen from the test methods that these methods have the disadvantages that the test accuracy is limited by the unloading structure, the gravity deformation cannot be completely eliminated, and it is not suitable for the zero-gravity surface shape test in the assembly stage of the mirror assembly.

Contents of the invention

The technical problem of the present invention is: to overcome the deficiencies of the prior art, to provide a surface shape inspection method for large-diameter reflector assemblies, and to provide a reasonable zero-gravity surface shape test for large-diameter discretely supported reflectors in the assembly and adjustment test stage. The engineering implementation method is simple, and the test method is simple and does not include simulation analysis and calculation errors, which improves the accuracy of zero-gravity surface shape testing.

Technical solution of the present invention is:

A method for inspecting the surface shape of a large-diameter reflector assembly, characterized in that: the reflector assembly includes a reflector, six legs and a support plate, the reflector to be tested and the six legs are fixedly connected together by gluing, And the six outriggers are fixedly connected with the support plate;

The device used in this test method includes a test tower, a folding mirror, a compensator, and an interferometer; the spherical wavefront beam or the plane wavefront beam emitted from the interferometer passes through the compensated aspheric wavefront beam of the compensator and then reaches the folding mirror, and the beam After passing through the refraction of the folding mirror, it reaches the mirror, and the beam returns to the interferometer in the same way after passing through the mirror to form interference fringes;

The test tower includes a top support platform and a side plate, and the top support platform has a through hole that allows the reflector and six support legs to pass through and cannot allow the support plate to pass through; one of the side plates has a through hole for the passage of the beam emitted by the interferometer;

When the mirror surface of the mirror faces upward, the top support platform is used to support the folding mirror, compensator and interferometer, and the cavity surrounded by the top support platform, side plates and the ground is used to hold the mirror assembly; interferometer, compensator , The folded mirror is placed on the top support platform of the test tower from right to left, and the mirror assembly is placed inside the test tower;

When the mirror surface of the reflector is facing down, the reflector and six support legs in the reflector assembly pass through the through holes on the top support platform and are limited on the top support platform by the support plate; To the right, it is placed in the cavity surrounded by the top support platform, side plates and the ground. The interferometer is placed outside the test tower, and the beam emitted by the interferometer can be transmitted to the compensator through the through hole on the side plate;

The steps of the method include:

(1) The mirror surface of the mirror assembly is facing upward, and the surface shape data W _0(m×n) of the mirror in the mirror assembly is measured by an interferometer, and W _0(m×n) = W _m×n + G _0(m×n) +K _m×n ; among them, W _m×n is the original surface shape data of the reflector part, W _m×n is the known value, and G _0(m×n) is the gravity surface shape data, G _0(m×n) is an unknown value, K _m×n is an assembly error surface shape data, and K _m×n is an unknown value;

(2) The mirror in the mirror assembly faces downward, and the surface shape data W _180(i×j) of the mirror in the mirror assembly is measured by an interferometer, and W _180(i×j) =W _i×j + G _180(i×j) +K _i×j ; Among them, W _i×j is the original surface shape data of the reflector parts, W _i×j is the known value, and G _180(i×j) is the gravity surface shape data, G _180(i×j) is an unknown value, K _i×j is an assembly error surface shape data, and K _i×j is an unknown value;

(3) The surface shape data W _{0 (m×n)} obtained in step (1) and the surface shape data W _{180 (i×j)} obtained in step (2) are subjected to surface shape data matrix normalization processing to obtain normalized Obtain the surface shape data F _0(a×b) =W _a×b +G _0(a×b) +K _a×b of the reflector mirror in the reflector assembly when the surface of the reflector assembly is turned upward The surface shape data F _180(a×b) of the mirror facing downward =W _a×b +G _180(a×b) +K _a×b , where W _a×b is the original surface shape of the mirror part data, W _a×b is a known value; G _0(a×b) is the gravitational surface data after normalization processing when the mirror surface in the mirror assembly is facing up; G _180(a×b) is When the reflector in the reflector assembly faces downward, the gravity surface shape data after normalization processing;

(4) Superimpose the normalized surface shape data F _0(a×b) and F _180(a×b ) obtained in step (3) to eliminate gravity deformation and obtain the zero-gravity surface shape data of the mirror F=(F _0(a×b) +F _180(a×b) )/2=W _a×b +K _a×b , if the difference between F and W _a×b is the root mean square of K _a×b If the value is not greater than 0.003λ (λ=632.8nm), it is considered that the assembly of the mirror assembly meets the requirements. If the root mean square value of the difference between F and W _a×b K _a×b is greater than 0.003λ, it is considered that the assembly of the mirror assembly is If the requirements are not met, the reflector assembly needs to be reassembled.

The advantage of the present invention compared with prior art is:

(1) It provides a simpler engineering implementation method for the zero-gravity surface shape test of the large-diameter discretely supported mirror assembly;

(2) It does not require the support of mechanical simulation data, and will not introduce additional simulation calculation errors.

(3) The test results do not include the gravity deformation error, which provides an objective technical basis for accurately locating the assembly stress error.

Description of drawings

Fig. 1 is a block flow diagram of the inventive method;

Fig. 2 is the structural representation of the device when the mirror surface is upward;

Fig. 3 is the structural representation of the device when the reflector mirror faces downward;

Figure 4 is a simulation data map.

detailed description

A method for inspecting the surface shape of a large-diameter reflector assembly is mainly aimed at testing the zero-gravity surface shape after the large-diameter reflector and the bipod structure are assembled into discrete support components by glue injection and positioning, that is, the reflector to be tested and the six The four legs are fixedly connected together by gluing, and the six legs are fixedly connected with the support plate, and the reflector, the six legs and the support plate form the reflector assembly 5; the device used in this test method includes a test tower 4 , folding mirror 3, compensator 2 and interferometer 1; the spherical wavefront beam or planar wavefront beam sent from interferometer 1 passes through the aspheric wavefront beam after compensator 2 and reaches folding mirror 3, and the beam passes through folding mirror 3 reaches the reflector after refraction, and the light beam returns to the interferometer 1 to form interference fringes after passing through the reflector;

Described test tower 4 comprises top support platform and side plate, and the through hole that can make reflector and six support legs pass and can not make support plate pass is arranged on the top support platform; Wherein a side plate has through hole, uses The passage of the light beam emitted by the interferometer 1;

When the mirror surface of the reflector is upward, the top support platform is used to support the folding mirror 3, the compensator 2 and the interferometer 1, and the cavity surrounded by the top support platform, side plates and the ground is used to hold the reflector assembly 5; The instrument 1, the compensator 2, and the folding mirror 3 are placed on the top support platform of the test tower 4 from right to left, and the reflector assembly 5 is placed inside the test tower 4;

When the mirror surface of the reflector is facing down, the reflector in the reflector assembly 5 and the six support legs pass through the through holes on the top support platform and are limited on the top support platform by the support plate; folding mirror 3 and compensator 2 is placed in the cavity surrounded by the top support platform, side plates and ground from left to right, the interferometer 1 is placed outside the test tower 4, and the light beam emitted by the interferometer 1 can be transmitted to the test tower through the through hole on the side plate On compensator 2;

The steps of the method include:

(1) The mirror surface in the mirror assembly 5 faces upward, and the surface shape data W _0(m×n) of the mirror in the mirror assembly 5 is measured with the interferometer 1, and W _0(m×n) =W _{m ×n} +G _0(m×n) +K _m×n ; Among them, W _m×n is the original surface shape data of the reflector parts, W _m×n is the known value, G _0(m×n) is the gravity surface data, G _0(m×n) is an unknown value, K _m×n is the assembly error surface data, and K _m×n is an unknown value;

(2) The mirror in the mirror assembly 5 faces downward, and the surface shape data W _180(i×j) of the mirror in the mirror assembly 5 is measured by the interferometer 1, and W _180(i×j) =W _{i ×j} +G _180(i×j) +K _i×j ; Among them, W _i×j is the original surface shape data when the mirror part is in state, W _i×j is a known value, G _180(i×j) is the gravity surface data, G _180(i×j) is an unknown value, K _i×j is the assembly error surface data, and K _i×j is an unknown value;

(3) The surface shape data W _{0 (m×n)} obtained in step (1) and the surface shape data W _{180 (i×j)} obtained in step (2) are subjected to surface shape data matrix normalization processing to obtain normalized Obtain the surface shape data F _0(a×b) =W _a×b +G _0(a×b) +K _a×b and the mirror assembly 5 when the reflector mirror in the reflector assembly 5 is facing upwards after de-processing F _180(a×b) = W _a×b +G _180(a×b) +K _a×b of the surface shape data F 180(a×b) of the mirror facing downward, W _a×b is the original state of the mirror part Surface data, W _{a × b} is a known value; G _{0 (a × b)} is when the reflector mirror in the reflector assembly 5 faces upwards, the gravity surface data after normalization processing; G _{180 (a × b)} is the gravitational surface shape data after normalization processing when the reflector in the reflector assembly 5 faces downward;

(4) Superimpose the normalized surface shape data F _0(a×b) and F _180(a×b ) obtained in step (3) to eliminate gravity deformation and obtain the zero-gravity surface shape data of the mirror F=(F _0(a×b) +F _180(a×b) )/2=W _a×b +K _a×b , if the difference between F and W _a×b is the root mean square of K _a×b If the value is not greater than 0.003λ (λ=632.8nm), it is considered that the assembly of the mirror assembly (5) meets the requirements. If the root mean square value of the difference between F and W _a×b K _a×b is greater than 0.003λ, it is considered that the mirror assembly (5) meets the requirements. The assembly of component 5 does not meet the requirements, and the mirror component 5 needs to be reassembled.

A test method capable of removing the gravity error of a large-diameter discretely supported mirror assembly. The large-diameter discretely supported mirror assembly is used to deform the mirror into an elastic deformation under the action of a gravitational field, and the deformation is vertically upward and vertically downward on the mirror surface. The characteristics of the same value and opposite phase during the test, the surface shape test image is obtained by vertically upward and vertically downward of the mirror surface, and then the graphic data matrix is normalized for the surface shape data with different matrix dimensions and center positions, and then The averaged images are superimposed, and finally the surface shape under the gravity of the large-aperture discrete support mirror components is obtained. As shown in the finite element simulation results in Figure 4, Figure 4(a) is the simulated surface deformation diagram obtained from the mirror surface vertically upward, and Figure 4(b) is the simulated surface deformation diagram obtained from the mirror surface vertically downward , without considering other external forces and the shape of the naked mirror, the simulation surface deformation diagrams in Figure 4(a) and Figure 4(b) have the same value but opposite phases, and the superimposed figure 4(c) zero-gravity simulation surface shape deformation diagram. This method is correct from both theoretical and simulation point of view.

The process flow of the method of the present invention is shown in Figure 1, and the test principle diagrams are shown in Figures 2 and 3. The invention maintains no internal change in the state of the discrete support component of the large-diameter reflector, and the deformation of the reflector is within the range of elastic deformation, and has recoverability and repeatability. 180°flip test mirror assembly, the wavefront difference of the optical parts caused by gravity is theoretically equal, but the phase is opposite, through uniform superimposition of images, the influence of gravity can be effectively removed, and the zero-gravity surface shape of the mirror can be obtained. The specific method flow is as follows:

(1) The mirror surface in the mirror assembly 5 faces upward, and the surface shape data W _0(m×n) of the mirror in the mirror assembly 5 is measured with the interferometer 1, and W _0(m×n) =W _{m ×n} +G _0(m×n) +K _m×n ; Among them, W _m×n is the original surface shape data when the mirror part is in state, W _m×n is the known value, G _0(m×n) is the gravity surface data, G _0(m×n) is an unknown value, K _m×n is the assembly error surface data, and K _m×n is an unknown value;

(2) The mirror in the mirror assembly 5 faces downward, and the surface shape data W _180(i×j) of the mirror in the mirror assembly 5 is measured by the interferometer 1, and W _180(i×j) =W _{i ×j} +G _180(i×j) +K _i×j ; Among them, W _i×j is the original surface shape data when the mirror part is in state, W _i×j is a known value, G _180(i×j) is the gravity surface data, G _180(i×j) is an unknown value, K _i×j is the assembly error surface data, and K _i×j is an unknown value;

(3) The surface shape data W _{0 (m×n)} obtained in step (1) and the surface shape data W _{180 (i×j)} obtained in step (2) are subjected to surface shape data matrix normalization processing to obtain normalized Obtain the surface shape data F _0(a×b) =W _a×b +G _0(a×b) +K _a×b and the mirror assembly 5 when the reflector mirror in the reflector assembly 5 is facing upwards after de-processing F _180(a×b) = W _a×b +G _180(a×b) +K _a×b of the surface shape data F 180(a×b) of the mirror facing downward, W _a×b is the original state of the mirror part Surface data, W _{a × b} is a known value; G _{0 (a × b)} is when the reflector mirror in the reflector assembly 5 faces upwards, the gravity surface data after normalization processing; G _{180 (a × b)} is the gravitational surface shape data after normalization processing when the reflector in the reflector assembly 5 faces downward;

(4) Superimpose the normalized surface shape data F _0(a×b) and F _180(a×b ) obtained in step (3) to eliminate gravity deformation and obtain the zero-gravity surface shape data of the mirror F=(F _0(a×b) +F _180(a×b) )/2=W _a×b +K _a×b , if the difference between F and W _a×b is the root mean square of K _a×b If the value is not greater than 0.003λ (λ=632.8nm), it is considered that the assembly of the mirror assembly 5 meets the requirements. If the root mean square value of the difference between F and W _a×b K _a×b is greater than 0.003λ, it is considered that the assembly of the mirror assembly 5 is The assembly does not meet the requirements, and the mirror assembly 5 needs to be reassembled.

A test method capable of removing gravity errors of a large-diameter discretely supported mirror assembly, which is characterized in that: the large-diameter discretely supported mirror assembly is used to deform the mirror under the action of a gravitational field into elastic deformation, and the deformation is vertically upward and downward on the mirror surface. The characteristics of the same value and opposite phase in the vertical downward test, the surface shape test image is obtained by vertically upward and vertical downward of the mirror surface, and then the graphic data matrix regression is performed on the surface shape data of different matrix dimensions and center positions Then, the average image is superimposed, and finally the surface shape under the gravity of the large-aperture discrete support mirror components is obtained. The method of the invention has simple testing principles, strong engineering implementation, does not include simulation analysis and calculation errors, and has high testing accuracy; adopts objective testing principles and data processing methods, and meets the reliability requirements of space applications; it can be applied to large-diameter mirrors Zero-gravity surface shape test during component assembly and adjustment stage.

The present invention will be further described below in conjunction with drawings and embodiments.

Example 1

A method for inspecting the surface shape of a large-diameter reflector assembly is mainly aimed at testing the zero-gravity surface shape after the large-diameter reflector and the bipod structure are assembled into discrete support components by glue injection and positioning, that is, the reflector to be tested and the six The four legs are fixedly connected together by gluing, and the six legs are fixedly connected with the support plate, and the reflector, the six legs and the support plate form the reflector assembly 5; the device used in this test method includes a test tower 4 , folding mirror 3, compensator 2 and interferometer 1; the spherical wavefront beam or planar wavefront beam sent from interferometer 1 passes through the aspheric wavefront beam after compensator 2 and reaches folding mirror 3, and the beam passes through folding mirror 3 reaches the reflector after refraction, and the light beam returns to the interferometer 1 to form interference fringes after passing through the reflector;

Described test tower 4 comprises top support platform and side plate, and the through hole that can make reflector and six support legs pass and can not make support plate pass is arranged on the top support platform; Wherein a side plate has through hole, uses The passage of the light beam emitted by the interferometer 1;

As shown in Figure 2, when the mirror surface of the reflector is facing upward, the top support platform is used to support the folding mirror 3, the compensator 2 and the interferometer 1, and the cavity surrounded by the top support platform, side plates and the ground is used to hold The mirror assembly 5; the interferometer 1, the compensator 2, and the folding mirror 3 are placed on the top support platform of the test tower 4 from right to left, and the mirror assembly 5 is placed inside the test tower 4;

As shown in Figure 3, when the mirror surface of the reflector is facing down, the reflector in the reflector assembly 5 and the six support legs pass through the through holes on the top support platform and are limited on the top support platform by the support plate; Folding mirror 3 and compensator 2 are placed in the cavity surrounded by the top support platform, side plates and ground from left to right, interferometer 1 is placed outside the test tower 4, and the beam emitted by interferometer 1 can pass through the side plates The through hole on the top is transmitted to the compensator 2;

As shown in Figure 1, the steps of the method include:

(1) The reflector mirror in the reflector assembly 5 faces upwards, and the surface shape data W _{0 (m × n)} =0.118λ (λ=632.8nm) of the reflector in the interferometer 1 measurement reflector assembly 5=0.118λ (λ=632.8nm, this data is root mean square statistical value (RMS, all the following specific data are RMS), and W _0(m×n) =W _m×n +G _0(m×n) +K _m×n ; where, W _m×n is The original surface shape data of the mirror parts in the state, W _m×n = 0.023λ, G _0(m×n) is the gravity surface data, G _0(m×n) is the unknown value, K _m×n is the assembly error Surface shape data, K _m×n is an unknown value;

(2) The reflector mirror in the reflector assembly 5 faces downward, and the surface shape data W _180(i×j) =0.120λ of the reflector in the reflector assembly 5 is measured with the interferometer 1, and W _180(i×j) =W _i×j +G _180(i×j) +K _i×j ; where, W _i×j is the original surface shape data when the reflector part is in state, W _i×j =0.023λ, G _{180(i× j)} is the gravity surface data, G _180(i×j) is an unknown value, K _i×j is the assembly error surface data, and K _i×j is an unknown value;

(3) The surface shape data W _{0 (m×n)} obtained in step (1) and the surface shape data W _{180 (i×j)} obtained in step (2) are subjected to surface shape data matrix normalization processing to obtain normalized Obtain the surface shape data F _0(a×b) =W _a×b +G _0(a×b) +K _a×b =W _{0(m ×n)} = 0.118λ and the surface shape data F _180(a×b) =W _a×b +G _180(a×b) +K _a×b =W when the mirror in the mirror assembly 5 faces downward _180(i×j) = 0.120λ, where W _a×b is the original surface shape data of the reflector part state, W _a×b =W _m×n =W _i×j =0.023λ; G _{0(a× b)} is the _{gravitational} surface shape data after the normalization process when the reflector mirror in the reflector assembly 5 faces upwards; Gravity surface data after processing;

(4) Superimpose the normalized surface shape data F _0(a×b) and F _180(a×b ) obtained in step (3) to eliminate gravity deformation and obtain the zero-gravity surface shape data of the mirror F=(F _0(a×b) +F _180(a×b) )/2=W _a×b +K _a×b ＝0.025λ, and finally K _a×b ＝0.002λ. The root mean square value of the difference of W _{a × b} K _{a × b} < 0.003λ (λ = 632.8nm), the assembly stress error of the mirror assembly 5 is well controlled and meets the requirements, and there is no need to reassemble the mirror assembly 5 .

Claims (1)

1. the face shape method of inspection of a kind of large caliber reflecting mirror component, it is characterised in that：Mirror assembly (5) including reflecting mirror, Six supporting legs and gripper shoe, reflecting mirror to be tested is fixed together with six supporting legs by gluing, and six supporting legs with Gripper shoe is fixed together；

The device that the method for testing is used includes test tower (4), fold mirror (3), compensator (2) and interferometer (1)；From interferometer (1) the spheric wave front light beam for sending or plane wave front light beam are reached after non-spherical wavefront light beam after the compensation of compensator (2) In fold mirror (3), light beam is reached after the refraction of fold mirror (3) on reflecting mirror, light beam by reflecting mirror after backtracking to interference Instrument forms interference fringe on (1)；

Including top braces platform and side plate, having on top braces platform can make reflecting mirror and six to described test tower (4) Individual supporting leg passes through and can not make the through hole that gripper shoe passes through；Through hole is carried on one of side plate, is sent for interferometer (1) Light beam pass through；

When reflecting mirror minute surface upward when, top braces platform be used for support fold mirror (3), compensator (2) and interferometer (1), top The cavity that portion's support platform, side plate and ground surround is used to hold mirror assembly (5)；Interferometer (1), compensator (2), fold mirror (3) it is placed on from right to left on the top braces platform of test tower (4), mirror assembly (5) is placed on the interior of test tower (4) Portion；

When reflecting mirror minute surface down when, reflecting mirror in mirror assembly (5) and six supporting legs pass through top braces platform On through hole and carried out by gripper shoe spacing on top braces platform；Fold mirror (3) and compensator (2) are placed on from left to right In the cavity surrounded by top braces platform, side plate and ground, interferometer (1) is placed on test tower (4) outside, and interferometer (1) light beam for sending can be transferred on compensator (2) by the through hole on side plate；

The step of the method, includes：

(1) mirror in mirror assembly (5) faces up, and with interferometer (1) reflecting mirror in mirror assembly (5) is measured Face graphic data W_0(m×n), and W_0(m×n)=W_m×n+G_0(m×n)+K_m×n；Wherein, W_m×nFor reflecting mirror part status when original face figurate number According to W_m×nFor given value, G_0(m×n)For gravity face graphic data, G_0(m×n)For unknown-value, K_m×nFor rigging error face graphic data, K_m×n For unknown-value；

(2) mirror in mirror assembly (5) faces down, and with interferometer (1) reflecting mirror in mirror assembly (5) is measured Face graphic data W_180(i×j), and W_180(i×j)=W_i×j+G_180(i×j)+K_i×j；Wherein, W_i×jFor reflecting mirror part status when original face Graphic data, W_i×jFor given value, G_180(i×j)For gravity face graphic data, G_180(i×j)For unknown-value, K_i×jFor rigging error face figurate number According to K_i×jFor unknown-value；

(3) face graphic data W for obtaining step (1)_0(m×n)Face graphic data W obtained with step (2)_180(i×j)The face of carrying out graphic data Matrix normalization process, the face figurate number when mirror for obtaining obtaining after normalized in mirror assembly (5) faces up According to F_0(a×b)=W_a×b+G_0(a×b)+K_a×bFace graphic data F when facing down with the mirror in mirror assembly (5)_180(a×b)= W_a×b+G_180(a×b)+K_a×b, W_a×bFor reflecting mirror part status when original face graphic data, W_a×bFor given value；G_0(a×b)For reflection Gravity face graphic data when mirror in mirror assembly (5) faces up, after normalized；G_180(a×b)For mirror assembly (5) the gravity face graphic data when mirror in faces down, after normalized；

(4) face graphic data F after the normalized for obtaining step (3)_0(a×b)And F_180(a×b)Image overlay is carried out, weight is eliminated Force deformation, obtains surface shape of reflector in zero gravity state data F=(F_0(a×b)+F_180(a×b))/2=W_a×b+K_a×bIf, F and W_a×bDifference Value K_a×bRoot-mean-square value be not more than 0.003 λ (λ=632.8nm), it is believed that the assembling of mirror assembly (5) meets the requirements, if F With W_a×bDifference K_a×bRoot-mean-square value be more than 0.003 λ, it is believed that the assembling of mirror assembly (5) is undesirable, need again Mirror assembly (5) is assembled.