Large window facing transmission wave front high precision detection system and method

Abstract

The present application relates to the technical field of industrial detection, and especially relates to a large-window transmission wavefront high-precision detection system and method, which is suitable for large-diameter window optical components with a diameter of no less than 1200 mm. A standard spherical mirror is self-detected and calibrated to obtain surface reference data, and a stable transmission detection light path is constructed in combination with a laser tracker to establish a spatial coordinate relationship. In the detection process, a thermal stability determination and low-order drift compensation strategy based on transmission wavefront timing is introduced, low-order aberration decomposition, change rate determination and drift modeling are performed, low-order phase drift caused by thermal gradient and environmental disturbance under large-diameter conditions is effectively inhibited, and the accuracy and consistency of the transmission wavefront detection result are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial testing technology, and in particular to a high-precision transmission wavefront testing system and method for large windows. Background Technology

[0002] In large optical systems, large-window plane mirrors are often used as protective windows, isolation windows, or light-transmitting interfaces. Their transmission performance directly affects the image quality and measurement accuracy of the system. Therefore, high-precision detection of their transmitted wavefront has become an important aspect of engineering applications. In existing technologies, transmitted wavefront detection typically relies on interferometers in conjunction with reference optical elements to complete the measurement, using wavefront results obtained from a single or limited number of orientations as the basis for detection. However, when the window diameter increases to the order of millimeters, existing detection methods struggle to guarantee the stability and consistency of measurement results in practical applications due to limitations in assembly and adjustment accuracy, the superposition of errors in reference elements, and fluctuations in environmental conditions.

[0003] Specifically, existing transmission wavefront detection schemes generally suffer from insufficient repeatability of detection results under large window conditions. Wavefront data acquired at different times or in different orientations show significant differences, making it difficult to form reliable judgment criteria. Simultaneously, the surface shape error of the reference optical element itself is often directly superimposed on the measurement results, and the lack of effective benchmark elimination methods results in insufficient directionality of the detection results towards the measured window itself. Furthermore, existing technologies often use fixed-time waiting or empirical judgment to handle the influence of environmental and thermal conditions, resulting in low-order drift components that vary over time in the detection results. This manifests as an unstable overall trend in the transmission wavefront, affecting the reliability of the final peak-valley and root-mean-square values.

[0004] Therefore, existing technologies cannot simultaneously achieve measurement accuracy, result stability, and engineering repeatability in high-precision transmission wavefront testing for large-aperture windows, and thus cannot meet the actual needs of high-precision optical systems for testing the transmission performance of large windows.

[0005] To address the above issues, this application presents a high-precision transmission wavefront detection system and method for large windows. Summary of the Invention

[0006] The technical problem this invention aims to solve is to address the shortcomings of existing technologies by providing a high-precision transmission wavefront detection system and method for large-aperture optical components with a diameter of not less than 1200 mm. It obtains surface profile reference data through self-calibration of a standard spherical mirror and establishes spatial coordinate relationships using a laser tracker, constructing a stable transmission test optical path. During the detection process, a thermal stability determination and low-order drift compensation strategy based on the transmission wavefront time sequence is introduced. Through low-order aberration decomposition, rate of change determination, and drift modeling, low-order phase drift caused by thermal gradients and environmental disturbances under large-aperture conditions is effectively suppressed, improving the accuracy and consistency of the transmission wavefront detection results.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A high-precision detection method for the transmission wavefront of a large window is provided, applied to a detection device. The detection device includes a plane mirror of the large window to be tested, a standard spherical mirror, an interferometer, and a laser tracker. The interferometer is used to acquire interference fringes and reconstruct the wavefront. The diameter of the plane mirror of the large window to be tested is not less than 1200 mm. The method includes:

[0009] The interferometer is used to perform self-calibration on the standard spherical mirror to obtain the surface reference data of the standard spherical mirror;

[0010] Based on the surface reference data, the standard spherical mirror, the interferometer, and the large window plane mirror to be tested are installed at a predetermined testing station. The spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large window plane mirror to be tested is established by the laser tracker to determine the measurement optical axis.

[0011] Based on the spatial coordinate relationship and the measurement optical axis, a transmission test optical path is constructed with the standard spherical mirror as the reference, so that the measurement beam returns to the interferometer after being transmitted through the large window plane mirror under test. The interferometer collects the transmission interference fringes and performs wavefront reconstruction to obtain the original transmission wavefront data and outputs the detection result.

[0012] The standard spherical mirror is self-calibrated using the interferometer, including:

[0013] The standard spherical mirror is installed at the self-test station, so that the outgoing measurement beam of the interferometer is reflected by the standard spherical mirror and returned to the interferometer, generating the self-test interference optical path of the standard spherical mirror.

[0014] Under the self-testing interference optical path, the standard spherical mirror is controlled to perform interference acquisition in at least two different orientations, the orientations including rotation about the optical axis of the standard spherical mirror and tilting about an axis perpendicular to the optical axis;

[0015] Wavefront reconstruction was performed on the interference fringes collected under various orientations to obtain the corresponding wavefront data;

[0016] The wavefront data is subjected to attitude registration and fusion operations to separate and eliminate common terms introduced by interferometer system errors, thereby obtaining the surface reference data of the standard spherical mirror.

[0017] The standard spherical mirror, the interferometer, and the large-window plane mirror to be tested are installed at a predetermined testing station, including:

[0018] The surface reference data is imported into the measurement software of the interferometer to generate a reference compensation model corresponding to the standard spherical mirror;

[0019] Based on the benchmark compensation model, the installation orientation of the standard spherical mirror in the predetermined testing station is determined, and the orientation mark is recorded after the standard spherical mirror is installed.

[0020] The interferometer is installed at the predetermined detection station and the output optical axis of the interferometer is initially calibrated so that the output optical axis of the interferometer is aligned with the reference line of the predetermined detection station.

[0021] The large window plane mirror to be tested is installed at the predetermined testing station, and the installation posture of the large window plane mirror to be tested is fixed.

[0022] The spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large-window plane mirror under test is established using the laser tracker, and the measurement optical axis is determined, including:

[0023] The laser tracker is placed at the predetermined measurement position of the predetermined detection station, and a detection station coordinate system is constructed, wherein the detection station coordinate system includes a reference line for defining the optical axis direction and a reference plane for defining the height and level.

[0024] At least one spatial measurement reference point is set on the standard spherical mirror, the interferometer, and the large window plane mirror to be measured, respectively. The spatial measurement reference point includes a target ball seat, a reflective target ball, and a preset target hole position.

[0025] The laser tracker measures the spatial measurement reference point to obtain the center position data of the standard spherical mirror, the spatial position data of the interferometer's output end, and the center position data of the large window plane mirror under test.

[0026] Spatial fitting is performed on the ball's center position data, spatial position data, and center position data to obtain the spatial coordinate relationship, which includes at least relative distance, relative height difference, and relative tilt angle.

[0027] Based on the spatial coordinate relationship, the measurement optical axis is calculated and determined to be the axis that passes through the center of the standard spherical mirror and points to the center of the large window plane mirror to be measured.

[0028] Based on the spatial coordinate relationship and the measurement optical axis, a transmission test optical path is constructed with the standard spherical mirror as a reference, including:

[0029] The measurement optical axis is input to the control system of the interferometer to generate optical path parameters for transmission testing;

[0030] According to the spatial coordinate relationship, the optical axis of the standard spherical mirror and the reflection detection optical axis of the large window plane mirror under test are adjusted to the first predetermined angle and the second predetermined angle, and the corresponding adjustment parameters are recorded, wherein the first predetermined angle is 30° and the second predetermined angle is 60°.

[0031] The interferometer acquires transmitted interference fringes and performs wavefront reconstruction to obtain raw transmitted wavefront data, outputting detection results including:

[0032] Under the detection postures corresponding to the first predetermined angle and the second predetermined angle, and in conjunction with the optical path parameters, the reflected interference fringes are collected by the Richter-Common method and the wavefront is reconstructed to obtain the original transmission wavefront data of the large window plane mirror under test.

[0033] Statistical analysis is performed on the raw transmitted wavefront data to obtain peak-valley values ​​and root mean square values, which are then output as detection results.

[0034] The method further includes placing the large-window plane mirror under test in a hot immersion environment for a stabilization time of not less than 2 hours, wherein a calibration strategy is executed after placement, the calibration strategy including:

[0035] While keeping the assembly and adjustment parameters of the transmission test optical path unchanged, at least N frames of transmission interference fringes are continuously acquired at predetermined time intervals and wavefront reconstruction is performed respectively to obtain the corresponding transmission time series data.

[0036] The transmission time series data is subjected to low-order aberration decomposition to obtain a low-order coefficient time series including defocus and astigmatism terms.

[0037] The rate of change of the low-order coefficients is calculated based on the time series of the low-order coefficients, and the rate of change is compared with a preset discrete threshold to determine whether the large window plane mirror under test is in a thermally stable state that can output detection results. If not, the large window plane mirror under test is placed back into the thermal immersion environment.

[0038] If so, drift modeling is performed on the time series of the low-order coefficients to obtain low-order drift components, wherein the low-order drift components are used to compensate for the drift of the original transmitted wavefront data.

[0039] The transmission time series data is subjected to low-order aberration decomposition, including:

[0040] Phase reconstruction is performed on each transmission wavefront data in the transmission time series data to obtain the corresponding effective aperture region;

[0041] The effective aperture regions corresponding to each frame of transmitted wavefront data are intersected to obtain a common effective aperture region, and the common effective aperture region is used as the unified decomposition aperture of each frame of transmitted wavefront data.

[0042] Within the unified decomposition caliber, establish a set of low-order aberration basis functions that includes at least defocus and astigmatism terms;

[0043] Based on the values ​​of the low-order aberration basis function set within the unified decomposition caliber, a basis function matrix is ​​constructed, and the basis function matrix is ​​pre-calculated once to generate a projection operator for low-order aberration decomposition.

[0044] The transmission time series data is linearly projected using the projection operator to obtain the corresponding low-order coefficients, thereby generating a low-order coefficient time series. The low-order coefficients include at least the defocus coefficient and the astigmatism coefficient.

[0045] Drift modeling of the time series of the low-order coefficients includes:

[0046] The defocus coefficient and astigmatism coefficient in the low-order coefficient time series are respectively fitted according to the time sequence;

[0047] The low-order aberration compensation term corresponding to each time step is calculated based on the fitting results. The low-order aberration compensation terms are then summed and averaged to obtain the low-order drift component.

[0048] A high-precision transmission wavefront detection system for large windows, the system comprising:

[0049] The optical inspection module includes an interferometer, a standard spherical mirror, and an inspection station for carrying the large window plane mirror to be tested. It is used to construct a transmission inspection optical path based on the standard spherical mirror and to collect transmission interference fringes.

[0050] The spatial assembly and adjustment module includes a laser tracker, which is used to measure the spatial reference points of the standard spherical mirror, the interferometer and the large window plane mirror under test, establish spatial coordinate relationships and determine the measurement optical axis, and generate assembly and adjustment parameters for the transmission test optical path;

[0051] The data processing module is communicatively connected to the optical detection module and is used to perform wavefront reconstruction on the transmission interference fringes to obtain transmission time series data, perform low-order aberration decomposition on the transmission time series data to obtain low-order coefficient time series, perform drift modeling on the low-order coefficient time series to generate low-order drift components, and perform drift compensation on the transmission wavefront data.

[0052] The results output module calculates and outputs peak and valley values ​​and root mean square values ​​as detection results based on the drift-compensated transmitted wavefront data.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] This invention effectively reduces the superimposed influence of reference element surface shape errors on the transmission wavefront results by introducing a standard spherical mirror self-calibration and benchmark compensation mechanism, making the test results more focused on the transmission performance of the large window under test itself. Simultaneously, by combining a laser tracker to establish a unified spatial coordinate relationship, it ensures the geometric consistency of the measured optical axis under large aperture conditions, improving the repeatability of the test setup. During the test, by performing time-series acquisition of the transmission wavefront data, low-order aberration decomposition, and thermal stability determination based on the rate of change, it avoids the uncertainty caused by relying solely on a fixed waiting time. Furthermore, through low-order drift modeling and compensation, it eliminates the systematic phase drift caused by slow thermal evolution. Attached Figure Description

[0055] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0056] Figure 1 An exemplary application scenario diagram provided for an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of the detection device provided in an embodiment of the present invention;

[0058] Figure 3 This is a flowchart illustrating a high-precision transmission wavefront detection method for large windows provided in an embodiment of the present invention.

[0059] Reference numerals: 100, Detection device; 101, Large window plane mirror to be tested; 102, Detection station; 200, Interferometer; 201, Standard spherical mirror; 202, Laser tracker. Detailed Implementation

[0060] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0061] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0062] The high-precision transmission wavefront detection method for large windows described in this application is applicable to the evaluation of transmission wavefront quality for large-aperture planar window optical components, especially for the transmission performance acceptance, processing closure, and delivery retesting of large-window planar mirrors / optical windows with apertures reaching millimeters or larger, and further improved to the kilometer level. These large windows are typically used as protective windows, isolation windows, vacuum windows, or large-aperture light-transmitting interfaces in the optical path of a system. In engineering applications, the focus is not simply on the surface shape, but rather on whether the wavefront after the beam passes through meets the system's image quality or measurement accuracy requirements. Therefore, the transmission wavefront becomes a core acceptance indicator. The transmission wavefront can be understood as the phase distribution of the light wavefront on the exit surface after the beam passes through the large window under test. Specifically, after the light wave passes through the window's refraction and reflection, wavefront distortion may occur due to factors such as the window's surface shape, thickness variations, material inhomogeneity, or support / thermal effects. These distortions directly affect the beam quality and the system's imaging accuracy. The quality of the transmitted wavefront reflects the degree of change in the beam after passing through the window. Especially in applications with high image quality requirements, the accuracy of the transmitted wavefront directly affects the final performance of the system. It is important to emphasize that the transmitted wavefront of a large window is not determined by a single surface error, but by a comprehensive optical path difference response formed by the superposition of factors such as the surface shape of both sides, thickness consistency, material homogeneity, and clamping / gravity boundary conditions. When the aperture is enlarged to 1200mm and above, even a small perturbation of any of the above factors can manifest as a considerable low-frequency error in the wavefront, thus affecting the final determination of indicators such as PV and RMS.

[0063] In exemplary techniques, interferometry is typically used to construct the transmission test optical path. The wavefront is reconstructed by observing the interference fringes between the reference wave and the measurement wave, thus enabling a quantitative evaluation of the transmitted wavefront. In practice, the transmission test optical path often relies on high-precision reference elements, such as standard spherical mirrors, to provide a benchmark. Simultaneously, it is crucial to ensure that the spatial geometric relationship between the interferometer, the reference element, and the large window under test is controllable and reproducible. However, when the test object is upgraded to a large window, conventional small-aperture experience fails in several aspects: First, if the surface shape error of the standard spherical mirror itself is not effectively calibrated and included in the compensation, it will be directly miscalculated into the transmission wavefront of the tested part, resulting in a lack of traceability in the results; Second, with a large aperture, the optical path length is longer and the sensitivity is higher. Assembly deviations such as non-collinearity of the optical axis and slight relative attitude deviation will enter the wavefront results in the form of low-order terms such as defocus and astigmatism. Moreover, these low-order terms are highly similar to the actual processing errors and are difficult to distinguish through intuitive judgment; Third, the large window has a large heat capacity and a slow temperature field establishment. Even if the empirical condition of "immersion in heat for a certain time" is met, there may still be in-plane or thickness thermal gradients, causing the refractive index distribution to evolve slowly over time, resulting in low-frequency drift of the wavefront; Fourth, the gravitational deflection of the large window in the supported or suspended state cannot be ignored. Slight changes in boundary conditions can cause significant changes in the wavefront results, making it difficult to achieve consistency in retesting and cross-station comparison. The combined effect of these problems means that the common dilemma is not that the material cannot be measured, but that it can be measured but it is difficult to confirm that it represents the true transmission performance of the tested part, thus creating significant uncertainty in processing correction, acceptance judgment and dispute resolution.

[0064] Based on the aforementioned engineering constraints, the proposed solution does not simply provide a transmission test optical path, but rather forms a holistic methodological framework around how to obtain reproducible, compensable, and verifiable high-precision transmission wavefront results under large-window conditions. The framework first uses a standard spherical mirror as a reference benchmark, performing self-calibration on the standard spherical mirror using an interferometer to obtain surface reference data, establishing a measurement traceability chain from the source. On this basis, the standard spherical mirror, interferometer, and the large-window plane mirror under test are installed at a predetermined testing station, and a laser tracker is introduced to establish the spatial coordinate relationship between the three to determine the measurement optical axis. This transforms the optical path geometry from empirical light-based adjustments to coordinate-based adjustments, thereby improving the reproducibility of adjustments and the geometric stability under large-aperture conditions. Furthermore, the solution incorporates predetermined angle adjustments and the Reggio Emilia method's reflection interferometry acquisition logic when constructing the transmission test optical path. This allows the system to constrain and control low-order error-sensitive terms before transmission measurement, and also facilitates consistency verification of reflection and transmission measurement results when needed, reducing the risk of misjudgment caused by low-order aliasing.

[0065] refer to Figure 1 This is an exemplary application scenario diagram provided in the embodiments of this application.

[0066] like Figure 1As shown, the application scenario includes a testing device 100, a large-window plane mirror 101 to be tested, and a testing station 102. Both the testing device 100 and the large-window plane mirror 101 to be tested are arranged on the testing station 102. Figure 1 The diagram only schematically illustrates the relative positional relationship between the testing device 100 and the large window plane mirror 101 to be tested; the specific installation structure of the testing device 100 on the testing station 102 is not shown. Figure 1 It is fully shown in the middle.

[0067] The detection device 100 is used to emit a measurement beam and receive the return beam after passing through the large window plane mirror 101 to be measured, so as to obtain the transmission interference fringes and perform wavefront reconstruction.

[0068] The large window plane mirror 101 to be tested is installed on the testing station 102 and is located on the measuring optical axis of the testing device 100 in the testing state. After the measuring beam is emitted by the testing device 100, it is transmitted through the large window plane mirror 101 to be tested and returns to the testing device 100, thereby forming a transmission test optical path.

[0069] The testing station 102 is used to uniformly support and position the testing device 100 and the large-window plane mirror 101 to be tested, so as to ensure the stability of their spatial position and posture during the testing process. The specific structural form of the testing station 102 is not limited, as long as it can meet the requirements for assembly stability and repeatability during the large-aperture window transmission wavefront testing process.

[0070] refer to Figure 2 This is a schematic diagram of the detection device provided in the embodiments of this application.

[0071] like Figure 2 As shown, the detection device includes an interferometer 200, a standard spherical mirror 201, and a laser tracker 202.

[0072] The interferometer 200 is used to emit a measurement beam and receive the return beam after passing through the standard spherical mirror 201, so as to obtain interference fringes and perform wavefront reconstruction. The interferometer 200 serves as the measurement subject for transmission wavefront detection during the detection process, and its internal structure and specific type are not limited.

[0073] A standard spherical mirror 201 is set on the measurement optical axis of the interferometer 200 and is used as an optical reference element for transmission detection. The measurement beam is emitted by the interferometer 200 and propagates along the measurement optical axis, forming a reference reflection or reference at the standard spherical mirror 201, thereby providing a stable optical reference for subsequent transmission wavefront detection.

[0074] The laser tracker 202 is positioned to the side of the measuring optical axis and is used to measure the spatial position of the interferometer 200 and the standard spherical mirror 201 in the inspection station, so as to establish and calibrate the spatial coordinate relationship between the interferometer 200 and the standard spherical mirror 201, thereby providing a spatial reference for the assembly and adjustment of the inspection device and the determination of the measuring optical axis.

[0075] In this structural configuration, the interferometer 200, the standard spherical mirror 201, and the laser tracker 202 together constitute a detection device for implementing the high-precision detection method for large-window transmission wavefronts of this application. Through the coordinated cooperation of optical measurement and spatial measurement, it achieves control over the assembly accuracy and measurement stability during the detection of large-window transmission wavefronts.

[0076] Next, with reference to the accompanying drawings, the method of this application regarding the high-precision detection method for transmission wavefronts with large windows will be further elaborated. Figure 3 The method shown is applied to a detection device, which includes a standard spherical mirror, an interferometer, and a laser tracker. The interferometer is used to acquire interference fringes and reconstruct the wavefront. The diameter of the large-window plane mirror to be tested is not less than 1200 mm. The method includes:

[0077] S1: Perform self-calibration on the standard spherical mirror using the interferometer to obtain the surface reference data of the standard spherical mirror;

[0078] In this embodiment, the standard spherical mirror serves as a reference in the transmission test optical path, and its surface shape error directly contributes to the interferometric measurement results. By constructing a self-testing interferometric optical path for the standard spherical mirror without introducing the large-window plane mirror under test, and using an interferometer to collect and reconstruct its reflected interference fringes, the actual surface shape distribution of the standard spherical mirror under the current testing environment can be obtained.

[0079] Those skilled in the art will understand that the surface profile reference data is used to characterize the inherent error components introduced by the standard spherical mirror to the wavefront under measurement conditions, and thus serve as a basis for compensation in the subsequent transmission wavefront reconstruction process.

[0080] S2: Based on the surface reference data, install the standard spherical mirror, the interferometer, and the large window plane mirror to be tested at the predetermined testing station;

[0081] In this embodiment, the predetermined testing station provides a unified bearing and installation environment for the standard spherical mirror, interferometer, and the large-window plane mirror under test, ensuring that each optical component is in a stable and repeatable spatial state during the testing process. Based on the obtained surface reference data, the working orientation of the standard spherical mirror can be determined during the installation phase, ensuring that it maintains an attitude relationship consistent with the reference data during subsequent transmission testing, thereby guaranteeing the effectiveness of the compensation strategy. This installation process is not limited to a specific mechanical structure; those skilled in the art can select different installation methods according to actual testing needs, as long as the installation and adjustment parameters remain stable during the testing process and no additional uncontrollable deformation is introduced.

[0082] S3: Establish the spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large window plane mirror to be measured using the laser tracker, and determine the measurement optical axis;

[0083] In this embodiment, a laser tracker is used to acquire the spatial position information of each optical element in the testing station, thereby establishing a unified spatial coordinate relationship. By measuring the position of the center of the standard spherical mirror, the position of the interferometer's exit end, and the center position of the large-window plane mirror under test, the relative position and orientation relationship between the three can be determined, and the measured optical axis can be calculated accordingly. Those skilled in the art will understand that the measured optical axis is used to characterize the geometric reference direction of beam propagation during transmission testing, and its determination helps to reduce deviations caused by manual light adjustment or experience-based setup under large-aperture conditions.

[0084] S4: Based on the spatial coordinate relationship and the measurement optical axis, construct a transmission test optical path with the standard spherical mirror as the reference, so that the measurement beam returns to the interferometer after being transmitted through the large window plane mirror to be tested. The interferometer collects the transmission interference fringes and performs wavefront reconstruction to obtain the original transmission wavefront data and outputs the detection result.

[0085] In this embodiment, the transmission test optical path uses a standard spherical mirror as a reference. By constraining the measurement optical axis, the measurement beam returns to the interferometer after passing through the large-window plane mirror under test, forming stable interference conditions. The transmission interference fringes acquired by the interferometer contain optical path difference information introduced by factors such as the surface shape, thickness variation, and material refractive index distribution of the two sides of the large-window plane mirror under test. The original transmission wavefront data can be obtained through wavefront reconstruction. Furthermore, by combining the aforementioned surface shape reference data and spatial assembly information, the original transmission wavefront data can be further analyzed and compensated to output test results with engineering significance. Those skilled in the art will understand that the test results may include wavefront evaluation indicators such as peak-to-valley values ​​and root mean square values, and their specific output format can be set according to actual test requirements. This application does not impose further limitations.

[0086] Before detailing the specific technical aspects of the steps, this application's embodiments need to reiterate:

[0087] When performing transmission wavefront testing on window-type optical components, the measurement results are often not solely determined by the inherent optical errors of the component under test (DUT), but are also constrained by the thermal evolution and assembly boundary conditions during the testing process. Unlike common small-diameter components, when the aperture is increased to at least 1200 mm, the heat capacity of the DUT increases significantly, and the timescale for heat exchange and temperature field homogenization is lengthened. In engineering testing, even with constant temperature chambers, isolation enclosures, or similar thermal immersion measures, the surface and thickness of the DUT may still maintain a slowly changing temperature gradient over a considerable period. This temperature gradient alters the refractive index distribution of the material and introduces additional phase disturbances through air refractive index fluctuations and stationary thermal convection, resulting in time-varying low-frequency components in the transmission wavefront data. Since these low-frequency components often exhibit mathematically similar characteristics to low-order aberrations such as defocus and astigmatism, they can easily be mistaken for inherent transmission errors of the DUT without differentiation, thus affecting the reliability of the final judgment and the consistency of retests.

[0088] It's easy to see that to address the issue of temperature drift affecting interferometric detection, a thermal stability determination step is typically implemented. This could involve waiting a predetermined time after immersion in heat, or setting a threshold for the fluctuation amplitude of several measurements within a certain time window to determine whether a stable state suitable for outputting detection results has been reached. While such measures are effective in testing small- and medium-diameter devices, for large-diameter windows, relying solely on fixed waiting times or coarse-grained fluctuation judgments based on overall indicators often fails to capture the objective reality of the slow, localized evolution of the temperature field. Firstly, immersion time and temperature field uniformity are not strictly equivalent; the temperature gradient may continue to decay even after the waiting time is met. Secondly, short-term stability of the overall PV or RMS does not necessarily mean that lower-order terms have converged. PV represents peak-to-valley values, and RMS represents root mean square values. Especially under large-diameter conditions, the slow drift of lower-order terms may be masked by high-frequency noise, leading to misjudgments of stability. Therefore, if the existing thermal stability determination logic is still used under the large window condition, there may be engineering dilemmas such as premature output leading to results with thermal drift or excessive waiting leading to decreased detection efficiency and still not being able to explain the source of drift.

[0089] Understandably, even after thermal stability determination is introduced into the detection process, the usual approach is still to perform full-aperture, full-volume low-order aberration decomposition on each frame of wavefront data to extract the determination criteria and drift model input. This processing chain introduces new adverse effects in large-aperture scenarios: First, the amount of transmitted wavefront data increases exponentially with the increase of aperture and sampling density, and performing full basis function fitting on each frame, such as fitting multi-order Zernike terms, will bring a significant computational burden; Second, the key to thermal stability determination and drift compensation often focuses on a few low-order terms such as defocus and astigmatism, but full decomposition will inevitably generate a large number of high-order coefficients that are irrelevant to the determination, which not only increases computational redundancy but also makes the fitting process more sensitive to mask boundaries, edge noise, and local distortion, which may amplify coefficient jitter; Third, when a complete set of coefficients is output for each frame, the drift modeling stage requires additional screening, fitting, and back-substitution processes, which lengthens the data processing chain, increases parameter dependence, and reduces engineering reproducibility.

[0090] Those skilled in the art will understand that, in the above situation, the processing objective and the computational behavior are not equivalent: the information dimension required for the judgment is low, while the processing introduces high-dimensional fitting and unnecessary output, thereby causing unnecessary complexity and error propagation channels.

[0091] Next, we will further elaborate on the technical content of the self-test calibration method in this application.

[0092] It should be noted that the standard spherical mirror in this application is not only used as a passive reflection element in the transmission test optical path, but also introduced into the overall method as an optical reference carrier for transmission wavefront detection. In the scenario of large-window transmission wavefront detection, the detection result is essentially the superposition of the optical path difference introduced by the large-window plane mirror under test and various systematic errors in the reference optical path. If the surface shape error of the reference element itself is not clearly identified and distinguished, it will inevitably be mixed into the transmission wavefront result, thereby weakening the authenticity and traceability of the detection result. Therefore, the standard spherical mirror is specifically used in this application to establish the reference wavefront model in the transmission detection process. Its role is not only to participate in the optical path construction, but also to provide a quantifiable and compensable reference for subsequent transmission wavefront reconstruction.

[0093] Based on this, the surface profile reference data obtained from the self-test calibration is used to characterize the true surface profile of the standard spherical mirror under actual testing conditions and current environmental conditions. Unlike relying solely on the factory-specified accuracy or historical test results, this application directly constructs a self-test interference optical path for the standard spherical mirror using an interferometer without introducing the large-window plane mirror under test. It then collects and reconstructs the reflected interference fringes to obtain the surface profile distribution of the standard spherical mirror in the current testing configuration. Those skilled in the art will understand that this surface profile reference data essentially reflects the instantaneous reference error of the standard spherical mirror in the testing system, which includes the mirror's own manufacturing errors, low-order deviations introduced by the assembly and adjustment posture, and the combined influence of the testing environment on the reference optical path.

[0094] Understandably, the core purpose of the self-calibration process is to distinguish and quantify the surface shape error of the standard spherical mirror and the interferometer system error without relying on external high-order references, thereby providing a reliable reference for subsequent transmitted wavefront testing. In practice, the standard spherical mirror is first installed in a dedicated self-calibration station. This station ensures that the standard spherical mirror possesses sufficient mechanical stability and repeatable attitude adjustment capability during the self-calibration process. The outgoing measurement beam from the interferometer propagates along its own measurement optical axis, is reflected by the standard spherical mirror, and returns to the interferometer, interfering with the reference beam to form a self-calibration interferometric optical path for the standard spherical mirror. Under this optical path configuration, the phase information contained in the interference fringes simultaneously superimposes the surface shape error of the standard spherical mirror and the interferometer's own optical path error. This embodiment achieves effective separation of the two through subsequent attitude changes and data fusion.

[0095] After establishing the self-testing interferometric optical path, the system error and mirror error are decoupled by controlling the standard spherical mirror to perform interferometric acquisition under at least two different orientations. Specifically, rotation around the optical axis of the standard spherical mirror changes the projection direction of the spherical mirror shape error in the interferometer coordinate system, while tilting around an axis perpendicular to the optical axis introduces different incident and reflection geometric relationships, allowing the standard spherical mirror shape error to exhibit distinguishable spatial distribution characteristics under different orientations. In contrast, the interferometer's own system errors, such as reference surface errors and optical path distortion, remain unchanged or exhibit consistent changes during the aforementioned orientation changes. Those skilled in the art will understand that the multiple sets of interferometric fringe data obtained in this way mathematically constitute a set of observation samples where "system errors remain unchanged and mirror errors undergo predictable changes," providing a data foundation for subsequent error separation.

[0096] After completing multi-pose interferometric acquisition, wavefront reconstruction was performed on the interference fringes obtained in each pose to obtain the corresponding wavefront data. Subsequently, these wavefront data underwent attitude registration processing, mapping them uniformly to the same coordinate system of the spherical mirror itself, thus eliminating the geometric transformations introduced by different poses. Based on this, by fusing multiple sets of registered wavefront data, a common term that remains unchanged in each pose can be extracted and identified as a component introduced by the interferometer system error; simultaneously, components exhibiting consistent rotation or transformation patterns with attitude changes are attributed to the surface shape error of the standard spherical mirror itself. Through the above data processing, the final surface shape reference data can accurately reflect the actual surface shape distribution of the standard spherical mirror under the testing state. This surface shape reference data can then be used as a compensation reference in the transmission wavefront detection process, enabling subsequent measurement results to have higher accuracy and repeatability under large-aperture conditions.

[0097] It should be noted that the wavefront reconstruction described in this application can be implemented using phase shift interferometry, Fourier transform phase extraction, or wavefront reconstruction based on phase expansion, etc. The specific algorithm can be selected according to the method of acquiring the interference fringes and the type of interferometer, as long as it can recover the corresponding phase distribution from the acquired interference fringes and form wavefront data. This application does not limit this. Meanwhile, the attitude registration and fusion operations can be processed using coordinate registration methods based on rigid body transformation models, rotation and translation solution methods based on least squares fitting, or multi-dataset registration methods based on correlation analysis, etc. After registration, common error terms can be extracted and the surface shape error of the standard spherical mirror can be separated through weighted averaging, minimum variance fusion, or other equivalent data fusion methods. This application will not elaborate further on these methods.

[0098] In one example, the standard spherical mirror, the interferometer, and the large-window plane mirror to be tested are installed at a predetermined testing station, including:

[0099] S2.1: Import the surface reference data into the measurement software of the interferometer to generate a reference compensation model corresponding to the standard spherical mirror;

[0100] Specifically, the surface profile reference data of a standard spherical mirror is spatial phase distribution data established with the solid surface of the spherical mirror as a reference object. This data will participate in imaging as part of the reference optical path in the subsequent transmission test optical path, and its corresponding phase distribution will be deterministically superimposed on the wavefront result reconstructed by the interferometer. If this superposition relationship is not explicitly modeled and processed, the transmission wavefront result output by the interferometer will simultaneously contain the surface profile error of the standard spherical mirror and the optical path difference information introduced by the large-window plane mirror under test, thereby reducing the directivity of the test result to the tested object. Therefore, before introducing the standard spherical mirror into transmission testing, the surface profile reference data obtained from self-calibration needs to be converted into a compensation model that can be directly called by the interferometer measurement software, so that it can participate in the calculation in the form of known compensation terms during the wavefront reconstruction or data output stage.

[0101] In this embodiment, the surface reference data is stored in the form of a two-dimensional grid. Each grid point corresponds to a sampling position on the working surface of the standard spherical mirror, and the phase deviation value of that position relative to the ideal sphere is recorded. The reference data includes, during generation, conventional information such as the sampling aperture range, sampling interval, reference zero point position, and phase positive and negative directions. When the reference data is imported into the interferometer measurement software, the software first parses the metadata and creates a unique reference data entry corresponding to the standard spherical mirror. This entry identifies the applicable object, data source, and callable status of the reference data. Subsequently, the measurement software performs matching processing on the reference data according to the current interferometer measurement parameters, which include at least the actual measurement aperture size, the center position of the measurement area, and the wavefront sampling density. During the matching process, if the sampling point array of the reference data is inconsistent with the wavefront point array obtained by the interferometer in terms of spatial resolution or point distribution, the reference data is mapped to the point structure consistent with the real-time wavefront data through interpolation. If the effective aperture of the reference data is larger than the real-time measurement aperture, the excess part is clipped. If it is smaller than the real-time measurement aperture, it only participates in compensation within its coverage area.

[0102] Furthermore, to ensure that the reference compensation model is consistent with the actual installation state of the standard spherical mirror at the testing station, the attitude information recorded during the self-calibration stage must be incorporated when generating the compensation model. Specifically, the reference data is established in the self-calibration coordinate system during self-calibration, which is not necessarily consistent with the measurement coordinate system used by the interferometer in subsequent transmission testing. Therefore, when generating the compensation model, the measurement software performs coordinate transformation processing on the reference data based on the recorded spherical mirror installation orientation information, so that the spatial distribution direction in the reference data is consistent with the direction in the interferometer measurement coordinate system. The coordinate transformation includes at least the overall translation adjustment of the reference data to align with the center of the measurement aperture, the rotation adjustment around the measurement optical axis to match the spherical mirror orientation, and, if necessary, mirror mapping of the coordinate directions to eliminate the difference in imaging direction between self-calibration and transmission testing. After completing the coordinate transformation, the measurement software solidifies the processed reference data into a reference compensation model, and in the subsequent transmission wavefront reconstruction process, subtracts the phase distribution corresponding to this compensation model from the real-time reconstructed wavefront data point by point, thereby eliminating the influence of the standard spherical mirror surface shape error on the test results.

[0103] This application provides an exemplary data structure for a benchmark compensation model, which includes:

[0104] A reference phase distribution data unit is used to characterize the working surface of a standard spherical mirror. The reference phase distribution data unit is stored in the form of a two-dimensional data array to record the phase deviation value of the standard spherical mirror relative to the ideal sphere at each sampling position. Each data point corresponds to a spatial sampling position on the working surface of the standard spherical mirror. The two-dimensional data array contains at least sampling aperture range information and sampling point spatial distribution information.

[0105] A coordinate mapping parameter unit is used to describe the spatial attributes of the reference phase distribution data unit. The coordinate mapping parameter unit is used to define the correspondence between the reference phase distribution data and the interferometer measurement coordinate system. Its content includes at least the center position offset of the reference data, the rotation angle information around the measurement optical axis, and the coordinate direction mapping relationship when necessary. It is used to correctly map the reference phase distribution data to the corresponding spatial position in the transmitted wavefront reconstruction result during the compensation process.

[0106] The aperture and resolution matching parameter unit is used to limit the applicable range of the reference compensation model. The aperture and resolution matching parameter unit is used to record the effective aperture size, aperture center position and sampling resolution information corresponding to the reference compensation model, so that when the interferometer measurement aperture or sampling density changes, the reference phase distribution data can be clipped, interpolated or resampled, thereby ensuring the consistency of the compensation model and the real-time measured wavefront data in terms of spatial scale.

[0107] S2.2: Based on the benchmark compensation model, determine the installation orientation of the standard spherical mirror in the predetermined testing station, and record the orientation mark after the standard spherical mirror is installed;

[0108] Specifically, the reference phase distribution data unit and coordinate mapping parameter unit included in the reference compensation model jointly define the spatial orientation relationship of the standard spherical mirror surface error in the measurement coordinate system. This orientation relationship is solidified into part of the compensation model after self-calibration. Since the reference phase distribution data is not perfectly symmetrically distributed in all directions on the spherical mirror surface, its local phase fluctuations have a clear directional characteristic in the two-dimensional data array. Therefore, only when the actual installation orientation of the standard spherical mirror in the predetermined testing station is consistent with the orientation definition used in the compensation model can the reference phase distribution data correctly correspond to the spatial position in the actual optical path during the transmitted wavefront compensation process. If the installation orientation is rotated or mirrored, even if the compensation model is invoked, the compensation term and the actual error term cannot be superimposed and canceled due to spatial mapping mismatch.

[0109] In this embodiment, the process of determining the installation orientation of the standard spherical mirror is based on the coordinate mapping parameter unit in the reference compensation model. First, the zero-azimuth definition information of the reference phase distribution data relative to the measurement coordinate system is read from the coordinate mapping parameter unit. This zero-azimuth definition information includes at least a preset reference direction and a corresponding rotation angle reference in the reference data. Subsequently, a station reference direction consistent with the measurement coordinate system is established at a predetermined testing station. This station reference direction can be defined by a mechanical reference line, a station reference edge, or a direction line formed by laser projection. During the installation of the standard spherical mirror, the rotation angle of the spherical mirror around its optical axis is adjusted to align the preset orientation mark on the spherical mirror with the station reference direction, thereby ensuring that the actual orientation of the spherical mirror is consistent with the coordinate mapping relationship in the reference compensation model. This adjustment process can be completed manually by rotating the mirror and reading the angle scale, or it can be assisted by a limiting structure or positioning pins, as long as the orientation consistency requirement is met.

[0110] Furthermore, after the standard spherical mirror is installed and locked, the orientation markings are recorded to establish a long-term correspondence between the spherical mirror entity and the reference compensation model. The orientation marking record includes at least a description of the correspondence between the spherical mirror orientation markings and the reference direction of the testing station, as well as the matching reference compensation model identification information. Through this recording method, when the standard spherical mirror needs to be reinstalled or reset during subsequent testing, it can be directly restored to an installation orientation consistent with the compensation model based on the orientation markings, without needing to re-perform self-calibration or regenerate the compensation model.

[0111] Those skilled in the art will understand that the specific form of the orientation mark can be a scribing line, a marker point, a positioning hole, or a combination thereof, as long as it can uniquely characterize the installation orientation of the spherical mirror around the optical axis. This application does not impose further limitations here.

[0112] S2.3: Install the interferometer at the predetermined detection station and perform initial calibration on the outgoing optical axis of the interferometer so that the outgoing optical axis of the interferometer is aligned with the reference line of the predetermined detection station;

[0113] Specifically, under large-window transmission wavefront testing conditions, the interferometer's output optical axis not only determines the propagation direction of the measurement beam but also directly participates in establishing subsequent spatial coordinate relationships and the geometric configuration of the transmission test optical path. When there is a deviation between the interferometer's output optical axis and the geometric reference line of the testing station, this deviation will be superimposed on the transmission wavefront results in the form of deterministic low-order components, exhibiting stable but difficult-to-distinguish systematic error characteristics in measurements at different orientations or times. Since this type of error is mathematically similar to the defocusing and astigmatism transmission errors of the large-window plane mirror under test, failing to constrain the interferometer's optical axis during the installation phase will increase the complexity of subsequent low-order determination and drift modeling. Therefore, during the interferometer installation process, it is necessary to use the reference line of the testing station as a geometric reference to initially correct the interferometer's output optical axis, making it a stable directional reference throughout the entire testing process.

[0114] In this embodiment, the interferometer is mounted on an adjustable mounting structure at a predetermined testing station. This adjustable mounting structure has at least height adjustment and angle fine-tuning capabilities for adjusting the spatial attitude of the interferometer. A baseline is predefined at the testing station. This baseline can be formed through the mechanical reference structure of the station body, the direction of the guide rail, or a spatial reference line established by an alignment device, and is used to characterize the desired measurement optical axis direction at the testing station. After the interferometer is installed, a rough alignment is first performed so that the emitted beam from the interferometer propagates approximately along the baseline direction and covers the predetermined working area of ​​the standard spherical mirror. Subsequently, by fine-tuning the mounting structure, the pitch and yaw angles of the interferometer are gradually adjusted so that the propagation direction of the emitted beam in space tends to be consistent with the baseline. During this adjustment process, the interferometer's own echo signal strength, fringe contrast, or spot position stability can be used as auxiliary criteria to determine whether the emitted optical axis is in a reasonable alignment state.

[0115] S2.4: Install the large window plane mirror to be tested at the predetermined testing station and fix the installation posture of the large window plane mirror to be tested.

[0116] In yet another example, the spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large-window plane mirror under test is established using the laser tracker, and the measurement optical axis is determined, including:

[0117] S3.1: Arrange the laser tracker at the predetermined measurement position of the predetermined detection station and construct the detection station coordinate system, wherein the detection station coordinate system includes a reference line for defining the optical axis direction and a reference plane for defining the height and level.

[0118] Specifically, the assembly and adjustment geometry of large-aperture transmission testing spans long distances and large apertures. Simply relying on visual alignment or interference fringe status for on-the-spot measurement and adjustment can easily lead to uncontrollable differences between different batches and different operators. Therefore, before determining the optical axis, a coordinate system for the testing station is established using a laser tracker. This provides a unified reference for all subsequent position and attitude data, which helps to transform the assembly and adjustment process into a recordable and reproducible spatial measurement process.

[0119] In this embodiment, the laser tracker is positioned at a measurement location on the side or end of a predetermined testing station. This location ensures visual coverage of the standard spherical mirror, the interferometer's output end, and the spatial measurement reference point of the large-window plane mirror under test, while minimizing the intersection of the measurement optical path with personnel passage areas. After completing self-testing and station orientation, the laser tracker selects three non-collinear reference points on the testing station as the basis for establishing the coordinate system. These reference points can be preset target holes, fixed target ball seats, or repeatedly accessible reference surface corners on the station. By measuring the three reference points, a station reference plane is fitted and defined as the horizontal reference of the station coordinate system. Subsequently, two more points are selected on this plane to define the station reference line. This reference line can be set along the direction of the station guide rail, the direction of the mechanical reference edge, or the direction of a preset alignment line, and is defined as the axial reference direction of the station coordinate system. After the coordinate system is established, the laser tracker outputs the origin, axial direction, and plane normal direction of the station coordinate system and stores them in association with the current measurement task.

[0120] S3.2: At least one spatial measurement reference point is set on the standard spherical mirror, the interferometer, and the large window plane mirror to be measured, respectively. The spatial measurement reference point includes a target ball seat, a reflective target ball, and a preset target hole position.

[0121] Specifically, establishing spatial coordinate relationships relies on feature points that can be stably tracked by a laser tracker. However, large-aperture optical components typically have high reflectivity or low texture, making it easy to encounter problems such as unstable reflection paths or uncertain measurement points when directly using the optical surface as the measurement object. Therefore, it is necessary to set spatial measurement reference points on each object, converting key geometric features of the measured object, such as the center of a sphere, the center of the exit end, and the center of the window, into repeatable geometric points. The arrangement of reference points not only affects measurement accuracy but also the solvability of attitude and center position during subsequent fitting. Therefore, the number and location of these reference points must meet the minimum observable condition.

[0122] In this embodiment, a spatial measurement reference point for determining the center position of the sphere is set on the standard spherical mirror. For example, a target ball or a reflecting target ball can be installed on the outer periphery of the spherical mirror, and the center position is determined by combining known structural parameters of the spherical mirror, such as the geometric offset of the target ball relative to the center of the sphere. When the spherical mirror body has a preset target hole position, a reflecting target ball can also be directly installed on the hole position as a measurement point. A spatial measurement reference point for the exit end is set on the interferometer. For example, a target ball can be set near the exit port of the interferometer, so that the reference point has a fixed geometric offset relationship with the exit optical axis, or the calibration hole position on the interferometer housing can be used as a reference point and its fixed parameters relative to the exit optical axis can be bound in the software. A spatial measurement reference point for determining the center position and attitude of the large window plane mirror under test is set on the plane mirror under test. For example, at least three reflecting targets can be arranged at equal angles on the outer periphery of the window, or a set of measurement points can be formed using preset holes on the support fixture, so that the window center and normal direction can be obtained subsequently through point set fitting.

[0123] S3.3: The laser tracker is used to measure the spatial measurement reference point to obtain the center position data of the standard spherical mirror, the spatial position data of the interferometer output end, and the center position data of the large window plane mirror to be measured;

[0124] S3.4: Perform spatial fitting on the center position data, spatial position data, and center position data to obtain the spatial coordinate relationship, which includes at least the relative distance, relative height difference, and relative tilt angle;

[0125] Specifically, the spatial coordinate relationship between the three includes not only distance but also relative height difference and relative tilt angle information. These quantities together determine whether the transmission test optical path can form a stable echo and effective aperture coverage within a large aperture range. If only "distance" is used as the basis for assembly, situations may arise where the beam can reach the target but the echo cannot return completely, or the effective aperture may deviate, causing lower-order terms to be amplified. Therefore, it is necessary to form a set of parameters containing position and attitude constraints through spatial fitting, so that the subsequent optical path construction and angle assembly have a directly applicable geometric basis.

[0126] In this embodiment, spatial fitting uses the workstation coordinate system as a unified reference, projecting the sphere center position, the exit end position, and the window center position onto the same coordinate system. Relative distances can be directly calculated from the coordinate differences between the three points, forming distance parameters such as "distance from the interferometer exit end to the sphere center" and "distance from the sphere center to the window center." The relative height difference is obtained from the coordinate differences of each point in the normal direction of the workstation reference plane, used to determine whether the optical axis meets the workstation layout and optical path obstruction conditions in the height direction. The relative tilt angle is derived from the angle between the connecting line direction and the workstation reference line direction, used to characterize whether the line connecting the interferometer, the sphere center, and the window center is consistent with the workstation axis. For ease of subsequent use, the above parameters are stored in the form of a spatial coordinate relationship parameter set. This parameter set includes not only numerical values ​​but also the corresponding point pair definitions and coordinate system version numbers to avoid mixing different coordinate systems.

[0127] S3.5: Based on the spatial coordinate relationship, calculate and determine the measuring optical axis as the axis that passes through the center of the standard spherical mirror and points to the center of the large window plane mirror to be measured;

[0128] Specifically, when the transmission test optical path is referenced to a standard spherical mirror, the center of the sphere provides a geometrically stable reference point, allowing the propagation direction of the measurement beam to be defined "point-to-point" without relying on fringe experience. Simultaneously, the center of the window, as the light transmission center of the test piece, determines the coverage position of the effective aperture on the test piece. Defining the measurement optical axis as an axis passing through the center of the sphere and pointing towards the center of the window facilitates control of the beam incident position and echo path under large aperture conditions, reducing aberrations introduced and effective aperture shift caused by off-axis incident light. This results in subsequent transmission wavefront reconstruction that more closely approximates the true response of the test piece under the expected light transmission conditions.

[0129] In this embodiment, the optical axis is determined using the coordinates of the sphere's center and the window's center as inputs, forming a spatial direction line from the sphere's center to the window's center. This direction line is used as the direction of the optical axis. Simultaneously, the sphere's center is used as a reference point on the optical axis to ensure that the optical axis definition is consistent with the standard spherical mirror reference. After the optical axis is determined, its consistency is compared with the interferometer's output optical axis. If there is a difference in direction, the adjustable mounting structure of the interferometer or the fine-tuning mechanism of the spherical mirror / window is adjusted to gradually bring the interferometer's output beam direction closer to the measured optical axis direction. After adjustment, the laser tracker can be used again to re-measure the positions of the output reference point, the sphere's center, and the window's center to verify that the collinearity of the three in the workstation coordinate system meets the preset requirements. The final measured optical axis parameters are then associated and recorded with the corresponding installation and adjustment parameters.

[0130] For example, the following numerical example is provided to illustrate how to complete centering, attitude determination, and locking recording based on spatial measurement data when installing and fixing the large-window plane mirror to be tested at a predetermined testing station. This example is only used to explain the calculation link, field organization, and dimensional relationships. The selected parameters and values ​​are illustrative and do not represent actual calibration results or engineering recommendations.

[0131] In this example, the baseline of the inspection station is defined as the positive X-axis of the station coordinate system, and the measurement optical axis is agreed to coincide with this X-axis; the measurement result of the center of the standard spherical mirror is used to provide the point through which the optical axis passes. The laser tracker measured the coordinates of the center of the standard spherical mirror as (X=2500.000mm, Y=0.000mm, Z=1200.000mm), and set up three repeatedly accessible spatial measurement reference points A, B, and C on the large window plane mirror to be measured (for example, by attaching a target ball base or installing a reflective target ball using preset holes). The coordinate measurement results are as follows: point A (X=1600.200mm, Y=+602.500mm, Z=1201.350mm), point B (X=1600.150mm, Y=-602.600mm, Z=1199.250mm), and point C (X=1600.100mm, Y=+0.050mm, Z=1799.900mm). Based on this, by fitting the plane containing points A, B, and C, the current "working surface spatial attitude" of the plane mirror can be obtained. Simultaneously, taking the geometric center of points A, B, and C as the approximate center position of the plane mirror under test, the coordinates of the center point are (X≈1600.150mm, Y≈-0.017mm, Z≈1400.167mm). It can be seen that the center of the plane mirror has a lateral offset of approximately 0.017mm from the theoretical plane of the optical axis in the Y direction, which can be considered a relatively small centering error in engineering. However, the difference in Z values ​​between the three points clearly shows that the plane mirror is tilted relative to the work position coordinate system: the Z height difference between points A and B at their positive and negative symmetrical positions in the Y direction is approximately 2.100mm, between 1201.350mm and 1199.250mm, indicating a significant asymmetrical tilt of the plane mirror in the pitch direction around the Z axis. At the same time, the Z height difference of point C relative to A and B also suggests a tilt component in the Y direction. If this tilt is not corrected, it will manifest as uneven fringe density and effective aperture shift in subsequent interferometric measurements. Furthermore, low-order terms such as astigmatism / defocus are easily amplified by assembly and adjustment errors, affecting the input stability of subsequent thermal stability determination and drift modeling.

[0132] Furthermore, the testing station provides a three-point support adjustment mechanism for the large-window plane mirror under test, located at support positions corresponding to A, B, and C. Each support point has the capability for fine-tuning with a screw or shim, and the adjustment amount is recorded in millimeters as the dimension of height change. Based on the measured height difference of the three points, the support point corresponding to point A is first adjusted down by 1.050 mm and the support point corresponding to point B is adjusted up by 1.050 mm to eliminate the 2.100 mm height difference between A and B, making the height of the plane mirror more consistent in the positive and negative Y-directions. Then, using point C as a reference, the support point at point C is finely adjusted in the opposite direction to make the average height of point C consistent with that of points A and B. For example, after completing the first round of adjustments, measurements were taken again to obtain the adjusted spatial measurement reference points A' (X=1600.205mm, Y=+602.498mm, Z=1200.320mm), B' (X=1600.152mm, Y=-602.595mm, Z=1200.290mm), and C' (X=1600.098mm, Y=+0.048mm, Z=1200.310mm). The maximum difference in Z values ​​among the three points was approximately 0.030mm, indicating that the working surface of the plane mirror was basically parallel to the reference plane of the workstation, and the tilt was compressed to a controllable range. At this point, the center point position was checked again, and the Y-direction offset remained within the order of 0.1mm. If further alignment was required, the entire plane mirror could be finely adjusted within 0.10mm in the Y direction using a translation mechanism to make the center point Y closer to 0. After this translation adjustment was completed, measurements were taken again to confirm the consistency between the center point and the optical axis plane. The "first determine the orientation, then center" sequence is used to avoid repeated iterations caused by centering before the tilt has converged, thereby reducing the ineffective movement and stress changes of large-diameter components at the work station.

[0133] Furthermore, after completing the attitude adjustment and alignment, the installation attitude of the large window plane mirror under test is fixed and recorded. In this example, fixing includes: tightening the three-point support fine-tuning screws and recording the final readings, such as support point A -1.050mm, support point B +1.050mm, and support point C -0.980mm; recording the coordinates of the center point of the plane mirror and the coordinates of the three-point remeasurement; recording the orientation of the plane mirror relative to the reference line of the testing station; and if the plane mirror has adjustable rotation around the optical axis, recording the alignment status of its orientation mark. To ensure that the attitude does not drift during subsequent timing data acquisition, a short-term retest can be performed after locking: Without changing the workstation status, the heights of points A', B', and C' are repeatedly measured at 10-minute intervals. If the difference in the Z-values ​​of the three points exceeds the preset tolerance (exemplarily 0.020 mm), it indicates that the support point has rebounded or the contact surface has slipped, requiring relocking or adjustment of the contact conditions. The tolerance value can be determined based on the diameter of the part under test, the support span, and the target wavefront accuracy requirements; this example is only for illustration. Those skilled in the art will understand that three-point support, four-point support, or suspension support can all achieve the above-mentioned "space measurement-attitude adjustment-retest locking" chain, as long as the large-window plane mirror under test maintains a reproducible spatial attitude during the testing period. This application does not impose further limitations here.

[0134] Next, we will further elaborate on the technical aspects of the detection results in this application.

[0135] In one example, the interferometer acquires transmitted interference fringes and performs wavefront reconstruction to obtain raw transmitted wavefront data, and outputs the detection results, including:

[0136] Under the detection postures corresponding to the first predetermined angle and the second predetermined angle, and in conjunction with the optical path parameters, the reflected interference fringes are collected by the Richter-Common method and the wavefront is reconstructed to obtain the original transmission wavefront data of the large window plane mirror under test.

[0137] Statistical analysis is performed on the raw transmitted wavefront data to obtain peak-valley values ​​and root mean square values, which are then output as detection results.

[0138] It should be noted that the acquisition of transmission interference fringes and wavefront reconstruction is not simply about obtaining interference results under a single attitude. Instead, it combines the geometric characteristics of large-window transmission detection in engineering applications, using different detection attitudes to constrain and verify the stability of the transmission wavefront results. Specifically, when the aperture of the large-window plane mirror under test is large, the transmission wavefront obtained under a single incident attitude is easily affected by residual assembly errors, slight optical path off-axis, and low-order thermal drift components, thus statistically masking the true transmission characteristics of the measured object. By repeatedly acquiring transmission interference fringes at different predetermined angles, the measurement beam can experience different propagation paths and incident conditions inside the window, allowing the attitude-related systematic components to exhibit identifiable variation patterns in the multi-attitude results. Meanwhile, the inherent transmission error of the large-window plane mirror under test remains consistent under different attitudes, thus providing a basis for comparison in subsequent data analysis.

[0139] In this embodiment, under the detection postures corresponding to the first and second predetermined angles, the interferometer controls the output direction of the measurement beam according to the established optical path parameters, so that the beam returns after being transmitted through the large window plane mirror under test, forming a reflected interference optical path. The interference fringes are acquired using the Rickey-Comman method, which, while ensuring the contrast and stability of the interference fringes, acquires multiple interference images through phase modulation or equivalent methods, and reconstructs the original transmitted wavefront data under the corresponding posture. During the wavefront reconstruction process, the reference compensation model and spatial coordinate relationships introduced in the previous steps are invoked to ensure that the reconstructed original transmitted wavefront data maintains consistency in aperture, coordinate direction, and reference reference. Those skilled in the art will understand that the Rickey-Comman method, as a mature interferometric measurement method, can vary in its specific implementation depending on the type of interferometer and the phase modulation method, as long as it can stably acquire reflected interference fringes and recover the phase distribution.

[0140] It should be noted that in this application, the first predetermined included angle is 30° and the second predetermined included angle is 60°.

[0141] After obtaining the raw transmission wavefront data under various testing postures, statistical analysis is performed on the data to form the final testing results. Specifically, firstly, the raw transmission wavefront data is masked within a unified effective aperture range to eliminate the influence of edge occlusion, fringe distortion, or insufficient signal-to-noise ratio regions on the statistical results. Subsequently, the peak-to-valley values ​​and root-mean-square values ​​of the transmission wavefront are calculated within this effective aperture range to quantify the maximum phase fluctuations and overall energy distribution introduced by the large-window plane mirror under transmission conditions. By comparing the peak-to-valley values ​​and root-mean-square values ​​obtained under different testing postures, the consistency of the testing results can be further judged. When the statistical results under multiple postures are within a reasonable convergence range, it indicates that the transmission wavefront results are not sensitive to posture changes and have high reliability. Finally, the peak-to-valley values ​​and root-mean-square values ​​that meet the consistency requirements are output as the testing results to evaluate whether the large-window plane mirror under test meets the corresponding optical performance requirements. Those skilled in the art can adjust the number of testing postures, the statistical aperture range, and the consistency criteria under different engineering conditions based on the above embodiments, and this application does not further limit this.

[0142] Next, we will further elaborate on the technical content of the correction strategy in this application.

[0143] In one example, the method further includes placing the large window plane mirror under test in a hot immersion environment, and the hot immersion stabilization time is not less than 2 hours.

[0144] Understandably, hot immersion typically refers to placing the optical element under test in a controlled environment, allowing it to fully exchange heat with the surrounding air and supporting structure, thus gradually bringing the overall temperature of the element closer to and stabilizing the ambient temperature. In existing technologies, hot immersion is usually achieved through a constant-temperature laboratory, a sealed enclosure, or an isolated cavity. The engineering assumption is that after the element has been placed in this environment for a predetermined time, its internal temperature distribution will be essentially the same as its surface temperature distribution, thereby reducing the impact of refractive index fluctuations and geometric deformations caused by temperature changes on optical measurements to an acceptable range. Therefore, in the testing of small- and medium-diameter optical elements, hot immersion is often considered a common pretreatment step, usually requiring only a fixed waiting time to meet the measurement requirements.

[0145] However, for large-aperture window optical components, the assumption of stability over a fixed time does not always hold. Due to the significantly increased aperture and thickness of large-window plane mirrors, their heat capacity and thermal inertia increase accordingly, significantly lengthening the establishment and homogenization process of the internal temperature field. Even when the apparent ambient temperature has been reached, a slowly decaying temperature gradient may still exist in the in-plane or thickness direction. Furthermore, the support method of the large window in the testing station, its contact area with the station structure, and local airflow conditions all further affect the heat exchange path, leading to differences in the thermal equilibrium state of different regions. Optically, this temperature gradient manifests as a non-uniform change in refractive index distribution, introducing a time-varying low-frequency phase component into the transmitted wavefront, often with a shape highly similar to low-order aberrations such as defocus and astigmatism. Compared to small-window components, even after the same immersion time, the transmitted wavefront of a large-window plane mirror may still be in a slowly evolving state. Direct testing can easily misjudge the lingering thermal effects as inherent transmission errors of the tested component, thus affecting the accuracy and repeatability of the test results.

[0146] In one example, the correction strategy includes:

[0147] A1: While keeping the installation and adjustment parameters of the transmission test optical path unchanged, at least N frames of transmission interference fringes are continuously acquired at predetermined time intervals and wavefront reconstruction is performed respectively to obtain the corresponding transmission time series data.

[0148] Specifically, even after the hot immersion process is complete and the surface temperature stabilizes, a slowly decaying temperature gradient may still exist inside and on the surface of the large-diameter window, causing the low-frequency components in the transmitted wavefront to drift continuously over time. Acquiring only a single frame of data makes it impossible to distinguish this drift component from the inherent transmission error of the test piece, and repeated measurements at different time points will produce inconsistent results. Therefore, acquiring multiple frames of time-series data while keeping the transmission test optical path adjustment parameters constant ensures that "constant geometric conditions" are a prerequisite, thus allowing the time-series fluctuations to primarily reflect the changing trend of thermally related disturbances rather than changes in the adjustment parameters.

[0149] In this embodiment, after the transmission test optical path is assembled and adjusted, the locking status of the interferometer, standard spherical mirror, and the large-window plane mirror under test is confirmed, and the locking position is kept unchanged during the acquisition to avoid low-order term jumps caused by slight loosening. At least N frames of transmission interference fringes are continuously acquired at predetermined time intervals. The predetermined time interval is used to balance the time scale of thermal drift and acquisition efficiency. For example, the predetermined time interval can be 30s or 60s; N is used to cover at least one observable slow drift window. For example, N can be 20 frames, 30 frames, or 60 frames. Wavefront reconstruction is performed immediately after each frame of fringes is acquired, and the transmission wavefront data corresponding to that frame is output and transmitted time-series data is formed in chronological order. In addition to the two-dimensional wavefront array, each frame of data also records the acquisition timestamp, attitude number, such as the first or second predetermined angle, effective aperture mask, and fringe quality indicators such as contrast or effective pixel ratio, for subsequent rejection of abnormal frames.

[0150] A2: Perform low-order aberration decomposition on the transmission time series data to obtain a low-order coefficient time series including defocus terms and astigmatism terms;

[0151] Specifically, thermal gradients and air refractive index fluctuations are more likely to manifest as low-frequency deformations in transmitted wavefronts, with their main energy concentrated in low-order terms such as defocus and astigmatism; while local processing textures or high-frequency noise in the window contribute little to thermal stability assessment. Decomposing each frame of the transmitted wavefront into low-order terms and forming a coefficient time series can transform complex two-dimensional wavefront changes into a comparable sequence of a few scalars, allowing subsequent judgments and modeling to avoid relying on high-order fitting of the full aperture, thereby reducing the complexity of the computational chain and minimizing parameter coupling.

[0152] In one example, low-order aberration decomposition is performed on the transmission time series data, including:

[0153] Phase reconstruction is performed on each transmission wavefront data in the transmission time series data to obtain the corresponding effective aperture region;

[0154] The effective aperture regions corresponding to each frame of transmitted wavefront data are intersected to obtain a common effective aperture region, and the common effective aperture region is used as the unified decomposition aperture of each frame of transmitted wavefront data.

[0155] Within the unified decomposition caliber, establish a set of low-order aberration basis functions that includes at least defocus and astigmatism terms;

[0156] Based on the values ​​of the low-order aberration basis function set within the unified decomposition caliber, a basis function matrix is ​​constructed, and the basis function matrix is ​​pre-calculated once to generate a projection operator for low-order aberration decomposition.

[0157] The transmission time series data is linearly projected using the projection operator to obtain the corresponding low-order coefficients, thereby generating a low-order coefficient time series. The low-order coefficients include at least the defocus coefficient and the astigmatism coefficient.

[0158] Specifically, the processing logic for low-order aberration decomposition of transmission time-series data is not aimed at obtaining a complete aberration spectrum, but rather at constructing the information dimensions required for thermal stability determination and low-order drift modeling. The changes in the transmission wavefront over time originate from both the refractive index changes caused by the slow evolution of the internal temperature field of the measured large-window plane mirror, and the low-frequency phase drift caused by air disturbances and thermal conduction from the supporting structure in the measurement optical path. These disturbances are holistic and continuous in spatial scale, manifesting as overall wavefront undulations or gradual changes along a certain direction, with their energy distribution concentrated in low-order modes such as defocus and astigmatism. In contrast, high-frequency processing textures, local defects, or fringe noise have weak correlation in the time dimension and contribute limitedly to thermal state determination. Based on this characteristic, by performing low-order decomposition of the wavefront data and forming a coefficient time series, the complex two-dimensional phase changes can be compressed into a small number of scalar sequences that can be compared over time while maintaining clear physical meaning, providing stable input for subsequent determination and modeling.

[0159] In this embodiment, low-order aberration decomposition is first established on the basis of a unified effective aperture. Since transmitted wavefront data acquired at different time points may differ in their effective aperture range due to fringe quality, occlusion boundaries, or minor beam offsets, directly decomposing each frame separately would make the low-order coefficients incomparable in spatial coverage, thus introducing spurious changes unrelated to the thermal process. Therefore, an intersection operation is performed on the effective aperture regions corresponding to the transmitted wavefronts of each frame to obtain a common effective aperture region, which is then used as the unified decomposition aperture, ensuring that the low-order coefficients of all frames originate from phase information within the same spatial range. Subsequently, a set of low-order aberration basis functions, including defocus and astigmatism terms, is constructed within the unified decomposition aperture. This set of basis functions spatially describes only the overall deformation characteristics and does not include high-frequency oscillation components. Based on the values ​​of this basis function set within the unified aperture, a basis function matrix is ​​constructed, and this matrix is ​​pre-calculated once before the start of temporal processing to obtain the projection operator used for low-order aberration decomposition. By applying the pre-computed projection operator to each frame of transmitted wavefront data, the corresponding low-order coefficients can be obtained directly without repeating the complete basis function fitting process on each frame.

[0160] Furthermore, the above processing method ensures high consistency and repeatability of low-order coefficient extraction in the temporal dimension. On the one hand, the projection operator is generated under a unified aperture and fixed basis function set, so that the coefficients at different time points only reflect the changes in the wavefront itself, without being affected by fitting weights, initial value selection, or numerical convergence paths. On the other hand, the one-time pre-computation method avoids repeated matrix construction and solving under large aperture and high sampling density conditions, thereby reducing the complexity of the computational chain while ensuring numerical stability. The defocus coefficient and astigmatism coefficient time series obtained in this way can fully characterize the main changing trends related to the thermal state in the transmitted wavefront with a lower data dimension, providing input data with clear physical meaning and controlled noise for subsequent rate of change determination and drift modeling. Based on the above embodiments, those skilled in the art can construct corresponding basis function sets and projection operators under different apertures and sampling conditions to achieve low-order aberration decomposition of transmitted wavefront time series data. This application does not further limit the specific basis function form or matrix size.

[0161] A3: Calculate the rate of change of the low-order coefficients based on the time series of the low-order coefficients, and compare the rate of change with a preset discrete threshold to determine whether the large window plane mirror under test is in a thermally stable state that can output detection results. If not, place the large window plane mirror under test back into the thermal immersion environment.

[0162] Specifically, the engineering significance of hot immersion lies in reducing temperature differences, but under large-diameter conditions, there may still be a slowly decaying temperature gradient; this gradient manifests as a continuous drift of low-order terms on the wavefront, so using the rate of change of low-order coefficients as a criterion is more effective in reflecting whether output conditions are met than simply relying on waiting time.

[0163] In this embodiment, the changes between adjacent sampling points are calculated for the defocus coefficient time series and the astigmatism coefficient time series, and converted into a change rate sequence by combining the sampling interval. To avoid misjudgment caused by a single outlier, the change rate sequence is not directly judged by extreme values, but a sliding window statistical method is used for robust judgment. For example, the most recent K frames, such as K=10 frames, can be taken as the judgment window, and the median or mean of the defocus change rate and the astigmatism change rate within the window are calculated and compared with a preset discrete threshold. The discrete threshold can be established by combining historical stable working condition data: under the same detection device and the same work position, a known stable window or no-load condition is selected, and no less than 30 frames of data are collected to obtain the statistical distribution of the defocus and astigmatism change rates. Then, the 95th percentile of this distribution is taken as the threshold benchmark, and a safety margin is added in engineering to form the final threshold. For example, the defocus rate threshold can be set to 0.002λ / min, and the astigmatism rate threshold can be set to 0.0015λ / min, where λ is the operating wavelength of the interferometer. This value is used to illustrate the threshold dimension and the method of value selection. If wavelength normalization is not used, nanometer-dimensional thresholds can also be used. For example, the defocus rate threshold can be set to 1.2 nm / min, and the astigmatism rate threshold can be set to 0.9 nm / min. If both types of rates of change within the judgment window are lower than their corresponding thresholds, the output condition is considered met; if either rate of change is higher than the threshold, the state is considered to still be thermally unstable.

[0164] A4: If so, perform drift modeling on the time series of the low-order coefficients to obtain low-order drift components, wherein the low-order drift components are used to compensate for the drift of the original transmitted wavefront data.

[0165] In one example, drift modeling of the time series of the lower-order coefficients includes:

[0166] The defocus coefficient and astigmatism coefficient in the low-order coefficient time series are respectively fitted according to the time sequence;

[0167] The low-order aberration compensation term corresponding to each time step is calculated based on the fitting results. The low-order aberration compensation terms are then summed and averaged to obtain the low-order drift component.

[0168] It should be noted that the use of low-order drift components for drift compensation of the original transmitted wavefront data can be understood as follows: the common trend reflecting continuous changes over time in the time series of low-order coefficients is explicitly removed from the original transmitted wavefront data, so that the final output transmitted wavefront result no longer contains time-dependent phase components introduced by slow thermal evolution or environmental stabilization processes. This compensation is not a correction for random fluctuations at a single moment, but rather a processing of low-order variation components that persist across multiple frames of data and can be mathematically described by a time function. When performing drift compensation on the original transmitted wavefront data, the low-order drift components are subtracted point-by-point from the corresponding transmitted wavefront data, so that the compensated wavefront data no longer exhibits systematic low-order drift in the time dimension, but only retains the phase distribution related to the inherent transmission performance of the measured large-window plane mirror.

[0169] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A high-precision detection method for transmission wavefronts with large windows, applied to a detection device, characterized in that, The detection device includes a standard spherical mirror, an interferometer, and a laser tracker, wherein the interferometer is used to acquire interference fringes and reconstruct the wavefront, and the diameter of the large-window plane mirror to be tested is not less than 1200 mm. The method includes: The interferometer is used to perform self-calibration on the standard spherical mirror to obtain the surface reference data of the standard spherical mirror; Based on the surface reference data, the standard spherical mirror, the interferometer, and the large window plane mirror to be tested are installed at a predetermined testing station. The spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large window plane mirror to be tested is established by the laser tracker to determine the measurement optical axis. Based on the spatial coordinate relationship and the measurement optical axis, a transmission test optical path is constructed with the standard spherical mirror as the reference, so that the measurement beam returns to the interferometer after being transmitted through the large window plane mirror under test. The interferometer collects the transmission interference fringes and performs wavefront reconstruction to obtain the original transmission wavefront data and outputs the detection result.

2. The high-precision transmission wavefront detection method for large windows according to claim 1, characterized in that, The standard spherical mirror is self-calibrated using the interferometer, including: The standard spherical mirror is installed at the self-test station, so that the outgoing measurement beam of the interferometer is reflected by the standard spherical mirror and returned to the interferometer, generating the self-test interference optical path of the standard spherical mirror. Under the self-testing interference optical path, the standard spherical mirror is controlled to perform interference acquisition in at least two different orientations, the orientations including rotation about the optical axis of the standard spherical mirror and tilting about an axis perpendicular to the optical axis; Wavefront reconstruction was performed on the interference fringes collected under various orientations to obtain the corresponding wavefront data; The wavefront data is subjected to attitude registration and fusion operations to separate and eliminate common terms introduced by interferometer system errors, thereby obtaining the surface reference data of the standard spherical mirror.

3. The high-precision transmission wavefront detection method for large windows according to claim 1, characterized in that, The standard spherical mirror, the interferometer, and the large-window plane mirror to be tested are installed at a predetermined testing station, including: The surface reference data is imported into the measurement software of the interferometer to generate a reference compensation model corresponding to the standard spherical mirror; Based on the benchmark compensation model, the installation orientation of the standard spherical mirror in the predetermined testing station is determined, and the orientation mark is recorded after the standard spherical mirror is installed. The interferometer is installed at the predetermined detection station and the output optical axis of the interferometer is initially calibrated so that the output optical axis of the interferometer is aligned with the reference line of the predetermined detection station. The large window plane mirror to be tested is installed at the predetermined testing station, and the installation posture of the large window plane mirror to be tested is fixed.

4. The high-precision transmission wavefront detection method for large windows according to claim 3, characterized in that, The spatial coordinate relationship between the standard spherical mirror, the interferometer, and the large-window plane mirror under test is established using the laser tracker, and the measurement optical axis is determined, including: The laser tracker is placed at the predetermined measurement position of the predetermined detection station, and a detection station coordinate system is constructed, wherein the detection station coordinate system includes a reference line for defining the optical axis direction and a reference plane for defining the height and level. At least one spatial measurement reference point is set on the standard spherical mirror, the interferometer, and the large window plane mirror to be measured, respectively. The spatial measurement reference point includes a target ball seat, a reflective target ball, and a preset target hole position. The laser tracker measures the spatial measurement reference point to obtain the center position data of the standard spherical mirror, the spatial position data of the interferometer's output end, and the center position data of the large window plane mirror under test. Spatial fitting is performed on the ball's center position data, spatial position data, and center position data to obtain the spatial coordinate relationship, which includes at least relative distance, relative height difference, and relative tilt angle. Based on the spatial coordinate relationship, the measurement optical axis is calculated and determined to be the axis that passes through the center of the standard spherical mirror and points to the center of the large window plane mirror to be measured.

5. The high-precision transmission wavefront detection method for large windows according to claim 4, characterized in that, Based on the spatial coordinate relationship and the measurement optical axis, a transmission test optical path is constructed with the standard spherical mirror as a reference, including: The measurement optical axis is input to the control system of the interferometer to generate optical path parameters for transmission testing; According to the spatial coordinate relationship, the optical axis of the standard spherical mirror and the reflection detection optical axis of the large window plane mirror under test are adjusted to the first predetermined angle and the second predetermined angle, and the corresponding adjustment parameters are recorded, wherein the first predetermined angle is 30° and the second predetermined angle is 60°.

6. The high-precision transmission wavefront detection method for large windows according to claim 5, characterized in that, The interferometer acquires transmitted interference fringes and performs wavefront reconstruction to obtain raw transmitted wavefront data, outputting detection results including: Under the detection postures corresponding to the first predetermined angle and the second predetermined angle, and in conjunction with the optical path parameters, the reflected interference fringes are collected by the Richter-Common method and the wavefront is reconstructed to obtain the original transmission wavefront data of the large window plane mirror under test. Statistical analysis is performed on the raw transmitted wavefront data to obtain peak-valley values ​​and root mean square values, which are then output as detection results.

7. The high-precision transmission wavefront detection method for large windows according to claim 1, characterized in that, The method further includes placing the large-window plane mirror under test in a hot immersion environment for a stabilization time of not less than 2 hours, wherein a calibration strategy is executed after placement, the calibration strategy including: While keeping the assembly and adjustment parameters of the transmission test optical path unchanged, at least N frames of transmission interference fringes are continuously acquired at predetermined time intervals and wavefront reconstruction is performed respectively to obtain the corresponding transmission time series data. The transmission time series data is subjected to low-order aberration decomposition to obtain a low-order coefficient time series including defocus and astigmatism terms. The rate of change of the low-order coefficients is calculated based on the time series of the low-order coefficients, and the rate of change is compared with a preset discrete threshold to determine whether the large window plane mirror under test is in a thermally stable state that can output detection results. If not, the large window plane mirror under test is placed back into the thermal immersion environment. If so, drift modeling is performed on the time series of the low-order coefficients to obtain low-order drift components, wherein the low-order drift components are used to compensate for the drift of the original transmitted wavefront data.

8. The high-precision transmission wavefront detection method for large windows according to claim 7, characterized in that, The transmission time series data is subjected to low-order aberration decomposition, including: Phase reconstruction is performed on each transmission wavefront data in the transmission time series data to obtain the corresponding effective aperture region; The effective aperture regions corresponding to each frame of transmitted wavefront data are intersected to obtain a common effective aperture region, and the common effective aperture region is used as the unified decomposition aperture of each frame of transmitted wavefront data. Within the unified decomposition caliber, establish a set of low-order aberration basis functions that includes at least defocus and astigmatism terms; Based on the values ​​of the low-order aberration basis function set within the unified decomposition caliber, a basis function matrix is ​​constructed, and the basis function matrix is ​​pre-calculated once to generate a projection operator for low-order aberration decomposition. The transmission time series data is linearly projected using the projection operator to obtain the corresponding low-order coefficients, thereby generating a low-order coefficient time series. The low-order coefficients include at least the defocus coefficient and the astigmatism coefficient.

9. The high-precision transmission wavefront detection method for large windows according to claim 7, characterized in that, Drift modeling of the time series of the low-order coefficients includes: The defocus coefficient and astigmatism coefficient in the low-order coefficient time series are respectively fitted according to the time sequence; The low-order aberration compensation term corresponding to each time step is calculated based on the fitting results. The low-order aberration compensation terms are then summed and averaged to obtain the low-order drift component.

10. A high-precision transmission wavefront detection system for large windows, used to implement the high-precision transmission wavefront detection method for large windows as described in any one of claims 1-9, characterized in that, The system includes: The optical inspection module includes an interferometer, a standard spherical mirror, and an inspection station for carrying the large window plane mirror to be tested. It is used to construct a transmission inspection optical path based on the standard spherical mirror and to collect transmission interference fringes. The spatial assembly and adjustment module includes a laser tracker, which is used to measure the spatial reference points of the standard spherical mirror, the interferometer and the large window plane mirror under test, establish spatial coordinate relationships and determine the measurement optical axis, and generate assembly and adjustment parameters for the transmission test optical path; The data processing module is communicatively connected to the optical detection module and is used to perform wavefront reconstruction on the transmission interference fringes to obtain transmission time series data, perform low-order aberration decomposition on the transmission time series data to obtain low-order coefficient time series, perform drift modeling on the low-order coefficient time series to generate low-order drift components, and perform drift compensation on the transmission wavefront data. The results output module calculates and outputs peak and valley values ​​and root mean square values ​​as detection results based on the drift-compensated transmitted wavefront data.

Images