CN111366099B - Pre-analysis-based interference weighted sampling dephasing analysis method and measurement system under any cavity length - Google Patents

Abstract

The invention discloses a pre-analysis based interferometric weighted sampling dephasing analysis method and measurement system under any cavity length. The multi-surface transparent measured component is a transparent parallel plate. By pre-analyzing the weighted multi-step dephasing algorithm, Store the available parameters, the corresponding cavity length coefficients and the phase-shifting reference coefficients necessary for the algorithm in the pre-analysis matrix of the algorithm library; when using this algorithm to solve the phase, the corresponding parameter values in the algorithm library are directly called and automatically pass the distance measurement. The sensor measures the distance value from the measured part to the front surface of the reference mirror and the cavity length value, which breaks through the limitation of the fixed algorithm that can only be applied to the measurement under the fixed cavity length, and realizes the convenient design of the algorithm; the measuring system of the present invention It mainly includes a laser ranging sensor, a fixture combination and a matching device for a guide rail, and the fixture combination includes a fixture and a guide rail frame. The guide rail frame of the invention can assist the laser ranging sensor to realize the automatic measurement of the distance of the measured object, is easy to operate, and has high measurement accuracy.

Description

Interferometric weighted sampling solution phase analysis method and measurement system based on pre-analysis under arbitrary cavity length

technical field

The invention relates to an optical interferometric measurement method and device, in particular to an interference weighted sampling dephase analysis method and a measurement device, which are applied to the technical field of optical measurement technology.

Background technique

Optical interferometry is a measurement method based on the wave nature of light, which can achieve nanometer and sub-nanometer precision. one of the measurement methods. In interferometric measurement, the current mainstream measurement method is the PZT (Piezoelectric Transducer, PZT) phase-shifting technology based on piezoelectric ceramics, which has been used for a long time and the technology is relatively mature. However, because of the hardware errors and mechanical stress errors generated in the PZT phase shifting process, this technique cannot cope with higher-precision measurement requirements (such as nanoscale). Especially for the multi-surface interference measurement of transparent parallel flat crystals, the measurement requirements of this kind of sample are very high, and the existing measurement methods generally need to separate the multi-surface interference signals according to the different phase-shift frequencies (by setting the cavity length). coefficient, that is, the ratio of the interference cavity length to the product of thickness and refractive index), so if the error is large, the separation and measurement of transparent multi-surface samples cannot be completed at the same time. At the same time, when measuring multi-surface transparent components such as parallel transmission thin plates and semiconductor substrate coatings, a set of self-interference fringes will be formed on the front and rear surfaces, and the front and rear surfaces of the component interfere with the reference mirror respectively, resulting in two sets of fringes. The three groups of interference fringes eventually form superimposed cross-interference fringes. If the traditional phase-shifting algorithm is used to calculate multi-surface interference fringes, a large error will occur. At present, the commonly used techniques include point-by-point scanning measurement based on polarized light ellipsometer and white light interference methods. The commonly used method in industrial applications is to coat the back surface of the flat plate with a matte paint or vaseline that matches the refractive index of the material to measure each surface one by one to eliminate the influence of the reflection on the back surface on the measurement, but this method will affect the components. The surface is damaged and the topography of multiple surfaces cannot be measured simultaneously.

The measurement method based on wavelength-shifting interferometry is a typical high-precision, non-contact measurement technology developed in recent years. This measurement technology changes the output wavelength of the laser by dynamically changing the voltage value applied to the laser, so as to achieve the purpose of changing the phase value, that is, the phase-shifting operation. After the phase-shift operation, the interference light intensity distribution of the interference information of each surface of the tested object under different phase-shift values is collected by the CCD camera, and then the algorithm is used to calculate and separate the aliased interference information. Then, the independent component information of each part of the signal can be obtained. After the phase solution, the initial phase value of the measured surface can be obtained, and the surface shape change information of the measured surface can be obtained. This is the principle of wavelength-shifted interferometry.

The frequency of light intensity change of the interference fringe signal in the process of wavelength tuning and phase shifting is related to the optical path difference of the coherent light that interferes. Because the optical path differences of the multiple groups of interference fringe signals to be measured are not the same, even if the wavelength tuning amount is constant, the change frequencies corresponding to their respective interference fringes are not the same, and the interference frequency of each interference information is the optical path difference. function, in other words, the interference frequency is a function of the cavity length and the thickness and refractive index of the device under test. Using Fourier transform and signal analysis theory, multiple groups of interference fringes with different frequency changes can be separated by time-domain windowed Fourier transform. At the same time, the interference technology of wavelength tuning and phase-shifting algorithm can not only realize the detection of the three-dimensional contour of the front and rear surfaces of the object to be measured, but also measure the thickness change.

Because the optical path difference between different surfaces and the reference surface is different, the interference information can actually be distinguished according to the frequency, that is, through the sample with a given thickness and refractive index, using the weighted multi-part When the phase-shifting algorithm performs phase solution and measurement, the cavity length that meets the requirements of the algorithm is selected to satisfy the separation and sampling conditions, and then each interference information is separated. However, when the DUT is placed in certain positions of the interferometer, the interference frequencies determined by the optical path difference may overlap or be close to each other, and the extraction of interference information cannot be performed. Furthermore, many of the current phase-resolving algorithms adopt multi-step phase-shifting algorithms. For this kind of algorithm, the form of the sampling window in the frequency domain is strongly related to the setting law of the cavity length coefficient and the phase-shifting value. When each reference frequency is separated, the result is not necessarily able to achieve the purpose of dephasing. In other words, the frequency of each signal is different, and it is not necessarily possible to directly realize the separation of the signals, and it is also related to the setting of the cavity length coefficient and the phase shift value. Therefore, when using the multi-step phase-shifting algorithm for multi-surface interferometry, it is necessary to limit the settings of the cavity length coefficient and the phase-shift value. Often, an algorithm can only cope with one or a limited number of situations after it is designed. , when the cavity length coefficient changes, the separation of the phase solution and the signal cannot be performed, which is one of the main difficulties of the current multi-surface phase solution technology based on multi-step phase shifting.

At the same time, because the traditional algorithm design method does not pre-evaluate and analyze the algorithm, the specific scope of application of a multi-step phase solution algorithm is generally unknown, and only one more or a limited number of This kind of scheme is used for algorithm design and experiment formulation, which limits the application scope of this kind of algorithm to a large extent.

Not only that, the traditional design methods are mostly based on discrete parameter design for algorithm application, but in the actual process, due to the limitation of experimental conditions, generally cannot strictly meet the ideal conditions, so the algorithm needs to be adjusted according to known conditions during measurement. Fine-tuning, rather than fundamentally comprehensive analysis of algorithms and estimation and storage of valid data.

In terms of hardware, the traditional method mostly uses a scale or a scaled slide rail to measure the distance of the DUT. This measurement method has low accuracy, and when the outer edge of the fixture cannot be on the same plane as the DUT, it must be included in the measurement. large measurement errors.

SUMMARY OF THE INVENTION

In order to solve the problems of the prior art, the purpose of the present invention is to overcome the deficiencies of the prior art, and to provide a pre-analysis based method and measurement system for interferometric weighted sampling for phase resolution under arbitrary cavity length. Especially for the problem that an algorithm in the weighted multi-step sampling algorithm can only be applied to a narrow range of discrete cavity length conditions, the present invention proposes a pre-analysis based interferometric weighted sampling phase solution technology with arbitrary cavity lengths To deal with the interferometric measurement of multi-surface transparent DUTs. The multi-surface transparent test piece targeted by the present invention is a transparent parallel flat plate. The principle is as follows: when the weighted multi-step sampling algorithm is used for dephasing calculation, not only the aliased interference frequencies need to be separated, but also the interference frequencies are independent of each other and there should be a certain distance between the frequency components so that the sampling function can be performed. At the same time, the weighted sampling window function needs to be able to accurately sample the signal of the target frequency in the frequency domain and not include information other than the target signal into the sampling. When the above three conditions are met, multi-step weighting can be used. The sampling algorithm extracts the target information and completes the phase solution operation. However, in the actual application process, after a weighted sampling algorithm is generally designed, the algorithm cannot be changed, and it can only process one or several discrete values with a fixed range of cavity length coefficients, and it is impossible to perform the designed algorithm. Dynamically adjust to cope with different cavity length conditions. However, for the actual measurement process, it is impossible to guarantee that the cavity length can well meet the conditions required by the algorithm. The reason is that changing the cavity length means changing the distance between the measured object and the front surface of the reference mirror, and this distance should not be too large or too large. Small: If it is too small, the experimental conditions cannot be satisfied, the reference mirror and the DUT are fixed on the fixture, and the fixture itself occupies a certain space; if it is too large, the size of the anti-vibration test bench where the interferometer is located may not meet the preset cavity length conditions and introducing errors. And if a designed algorithm only satisfies a few discrete cavity length values, the cavity length needs to be adjusted during the measurement process, that is, the distance from the test piece to the reference mirror needs to be adjusted through a scale or a scale guide. Errors will inevitably be introduced in the adjustment process, so that the current actual cavity length value cannot strictly meet the ideal cavity length conditions required by the algorithm. The measurement error is very large or even impossible to measure. The method proposed by the present invention is designed for different cavity length coefficients based on the pre-designed algorithm applicable coefficient library, and the distance measurement sensor is used to measure the distance of the measured piece with high precision, so the above problems can be well avoided .

In order to achieve the above-mentioned purpose of invention and creation, the present invention adopts the following technical solutions:

A pre-analysis based interferometric weighted sampling solution phase analysis method with arbitrary cavity length, comprising the following processes and steps:

a. According to the characteristics of the weighted multi-step sampling algorithm, double iteration is performed on the phase-shifting reference coefficient and cavity length coefficient within a certain range;

b. Take the mean square error of the maximum error for the results of the algorithm based on Zernike polynomial interference simulation and multi-step weighted discrete sampling. When the separation error of each information is less than 1.2um, the algorithm is determined to be available, otherwise it is determined to be unavailable, and then the This information is stored in a pre-analysis matrix;

c. Use the laser ranging sensor to measure the distance of the DUT, set the thickness and reflectivity information of the DUT, call the pre-analysis matrix, select and design the specific parameters in the algorithm, and select the available parameters from the algorithm library. value for weighted multi-step sampling calculation;

d. After the algorithm parameters are selected, the interferogram acquisition of the interferometer is carried out. The number of acquisition frames and the phase-shift value are related to the phase-shift reference coefficient. During the measurement, set the thickness value and refractive index of the input test piece to be measured. Measurement of each surface information of the test piece.

As a preferred technical solution of the present invention, in the process of formulating the iteration and value determination method, for each cavity length coefficient, the phase-shifting reference coefficient value corresponding to the right scale within 1 step is made to satisfy the current phase-shifting coefficient at the same time. For the condition that the reference coefficient and its right side are adjacent to the phase-shift reference coefficient, take the scale value on the right side of the scale range where the actual cavity length coefficient is located as the estimated value, and then select the phase-shift reference coefficient based on the estimated value; After the positioning and selection of the cavity length coefficient and the sampling function, the calculation of the sampling function is performed to complete the algorithm design preparation process.

As a preferred technical solution of the present invention, the aliasing interferometry of the multi-surface transparent DUT is performed. By performing double iteration on different cavity length coefficients and phase-shifting reference coefficients, the phase-shifting reference coefficients are firstly iterated, and in the iterative process The iteration of the cavity length coefficient is included in the calculation process, so as to obtain the unwrapped phase results of each surface solution under different cavity length coefficients and different phase-shifting reference coefficients, and then solve the surface shape results of each surface, mainly considering the measured The front surface, the back surface and the thickness variation of the part are obtained.

A measurement system based on pre-analysis interference weighted sampling under arbitrary cavity length, adopts the pre-analysis-based interference weighted sampling and dephase analysis method under arbitrary cavity length of the present invention, the measurement system mainly includes a laser ranging sensor, a fixture combination and The matching device of the guide rail, the clamp combination includes a clamp and a guide rail frame; the two ends of the guide rail frame are provided with adjustable reference arms, and the inner side of the adjustable reference arm has a rack and a toothed knob for cooperative transmission; the bottom end of the knob is a screw rod, which is connected with the inside of the guide rail frame. The threaded hole at the bottom end is threaded, and the knob and the adjustable reference arm are located at each end of the rail frame; the adjustable reference arm is coated with a high reflectivity coating according to the laser wavelength to match the work of the laser ranging sensor; the measured mirror is Clamp combination clamping; after the clamp clamps the mirror under test, it is assembled with the guide rail frame through the clamp assembly threaded hole; the guide rail frame is composed of two upper and lower hollow cavities, and the lower part is slotted, so that the guide rail frame is clamped on the guide rail and performs coordinated movement ;The upper part of the guide rail frame opens the guide rail frame assembly threaded hole to assemble the upper and lower structure of the guide rail frame through bolts; the adjustable reference arm is clamped in the grooves on both sides of the guide rail frame through the reference arm guide grooves at both ends and moves; the adjustable reference arm can be controlled by rotating the knob. Adjust the reference arm to move back and forth, so as to realize the correction with the front surface of the measured mirror and the distance measurement of the laser ranging sensor; the laser ranging sensors are arranged on the front and rear sides of the bottom of the interferometer, and the outer surface of the reference mirror is used as the Benchmark, measure the average distance of the calibrated adjustable reference arm as the absolute distance between the measured mirror and the reference mirror, so as to obtain the cavity length value; the laser ranging sensor is connected to the upper computer signal, and the computer reads the measured distance. The mean value is used as the data information for calculating the cavity length coefficient.

As a preferred technical solution of the present invention, the measured part is clamped by the clamps combined with the clamps, so that the measured transparent parallel plate can be perpendicular to the worktable, so that the measured parallel plate is parallel to the outer surface of the reference mirror of the interferometer, and by adjusting the clamp position, so that the front and rear surfaces of the tested object and the outer surface of the reference mirror interfere respectively.

As a preferred technical solution of the present invention, the bottom end of the guide rail frame cooperates with the guide rail on the worktable to move, and the vertical height center of the clamp is aligned with the center of the outer surface of the interferometer.

It is preferable to use a wavelength phase-shifting interferometer when acquiring the interferogram, and the number of acquisition frames and the phase-shift value are respectively related to the phase-shift reference coefficient.

Preferably, the interferometer is connected to the computer using its own data cable, and the computer directly reads the acquired interferogram collected by the CCD camera in the interferometer.

The high-precision laser ranging sensor is preferred, and the German JENOPTIK high-precision laser ranging sensor, model JGS-70, is preferred. The sensor is arranged on a worktable parallel to the outer surface of the reference mirror of the interferometer, and one is installed at each end of the interferometer. , The signal measured by the high-precision laser ranging sensor is transmitted to the computer through the data transmission line for reading, and the average of the two measured distance values is taken as the actual distance value. Calibrate the sensor and take the outer surface of the reference mirror as the benchmark, and the measured distance of the DUT at this time is the absolute distance.

As a preferred technical solution of the present invention, the fixture is divided into two parts, one part is the part that clamps the tested piece, here a vacuum self-centering mirror frame can be selected, and the fixture is selected according to the diameter of the tested piece, and the When the diameter of the test piece is between 30-200mm, it is preferable to select the fixture model: FPSTA-4SCML-8V, and the fixture needs to be installed on the rail frame. The other part is the guide rail frame matched with the fixture. The bottom of the fixture is drilled with threaded holes and bolted to the guide rail frame. The bottom end of the guide rail frame is matched with the guide rail and can be embedded on the guide rail. The two ends of the guide rail frame are provided with threaded holes for matching with the clamps. The two ends of the guide rail frame are provided with adjustable reference arms, and the inner side of the adjustable reference arm has a rack that can cooperate with the knob. The knob and the adjustable reference arm are located at each end of the rail frame. The reference arm is coated with a high reflectivity coating, preferably 1030-i0 coating material according to the laser wavelength, in order to cooperate with the work of the laser ranging sensor.

Preferably, the multi-surface dephasing algorithm includes 6 algorithms, which correspond to different phase-shifting reference coefficients respectively, and can deal with different distribution situations of cavity length coefficients.

Preferably, the guide rail is a high-precision measurement guide rail, and the GCM-720205M model guide rail is preferably used.

CCD camera is preferred, DFK37AUX273 model and its supporting data cable are preferred.

A wavelength tunable laser is preferred, and a wavelength tunable laser of the TLB-6804 series from Newfocus is preferred.

Principle of the present invention:

The design process of the algorithm proposed by the present invention is as follows: first, the main follow function is set based on experience, and through the accurate design of the multi-surface interference phase shift measurement algorithm, the Zernike fitting method is used to completely simulate the multi-surface interference acquisition process and the algorithm In the processing process, the unwrapped phase results of each surface solved under different cavity length coefficients and different phase-shifting reference coefficients are obtained through double iteration with different cavity length coefficients and phase-shifting reference coefficients, and then the surface shape of each surface is solved. As a result, three pieces of information, the front surface, the back surface, and the thickness variation of the DUT, are mainly considered here. De-tilt the surface profile results, and then solve the residuals with their respective mean values. The residual error here is the peak value of the distance that each data deviates from the mean value, that is, the maximum value of the mean difference. Store residual data and make decisions. The judgment condition is selected as follows: in the double iteration, under the condition of the current cavity length coefficient and the phase shift reference coefficient, the residual results of the three interference information of the current surface, the back surface and the thickness change are simultaneously less than 1.2nm, then it is determined that the Conditional algorithms apply. The double iterative results of the cavity length coefficient and the phase-shift reference coefficient are solved and placed in the pre-analysis matrix. When using this algorithm, a high-precision laser ranging sensor is used to measure the distance from the measured object to the reference mirror, and the value is read into the phase-solving algorithm designed by the present invention, and then only the values of thickness and refractive index need to be input, and then the Automatic calculation of cavity length coefficient and phase shift reference coefficient. In the solution process of the pre-analysis matrix, because the calculation process under different cavity length coefficients is inevitably discrete data, the data processing method designed by the present invention is: searching for the actual cavity length coefficient, there are always two discrete data. The actual cavity length coefficient is between these two values, so the larger value next to it is selected as the estimated value, and then the preparatory process of the algorithm is completed by searching the phase-shifting reference coefficient for the estimated value.

The next step is to bring the current adaptive phase-shifting reference coefficients and actual cavity length values searched from the pre-analysis matrix into the dephasing algorithm, and design the phase-shifting value and the number of interferogram acquisition frames and weighting through different phase-shifting reference coefficients. Sampling function, and then calculate the arctangent of the data to obtain the wrapping phase. After de-wrapping and de-tilting, the surface shape of the corresponding measured surface can be obtained, and the multi-surface interferometric measurement process can be completed.

When constructing the pre-analysis matrix, the iterative step size of the phase-shifting reference coefficient is 1, and the iterative step size of the cavity length coefficient is 0.1. The parameters of the double iteration are discrete, so it must not satisfy all the cavity length coefficient cases. The determination method is: for each cavity length coefficient, the phase-shifting reference coefficient value corresponding to the right scale within one step must satisfy the conditions of the current phase-shifting reference coefficient and its right-hand adjacent phase-shifting reference coefficient at the same time, Therefore, the processing method adopted here is to take the scale value on the right side of the scale range where the actual cavity length coefficient is located as the estimated value, and then select the phase-shift reference coefficient through the estimated value. After the positioning and selection of the phase-shifting reference coefficient and the cavity length coefficient, the sampling function can be calculated. At this time, the algorithm design process is completed. When using, only the thickness and reflectivity of the test piece can be input to automatically specify the algorithm. design.

Compared with the prior art, the present invention has the following obvious outstanding substantive features and significant advantages:

1. The present invention analyzes the weighted multi-step phase-solving algorithm in advance, and stores the available parameters, the corresponding cavity length coefficients and the necessary phase-shifting reference coefficients of the algorithm in the pre-analysis matrix of the algorithm library; utilize this algorithm to solve the phase problem. The corresponding parameter value in the algorithm library is directly called and the distance value between the measured part and the front surface of the reference mirror is automatically measured by the ranging sensor, that is, the cavity length value is measured, which breaks through the fixed algorithm and can only be applied to the fixed cavity. The long-term measurement limitation problem realizes the convenient design of the algorithm;

2. Based on the characteristics of the weighted multi-step sampling algorithm, the present invention also realizes the weighted multi-step interferometric measurement technology under any cavity length, breaks through the technical bottleneck, and broadens the application scope of the weighted sampling wavelength phase-shifting and de-phase algorithm;

3. The present invention constructs an available database of algorithm parameters by pre-analyzing the algorithm characteristics. When actually measuring, only need to input the thickness value and the reflectivity value to automatically search the parameter library to automatically design the algorithm and perform non-contact multi-surface. Simultaneous measurement of information, and the algorithm calculation speed is fast, the calculation cost is low, and the measurement effect is good, which can solve the problem of algorithm adaptability under any cavity length; Automatic measurement makes the realization of this technology more convenient; the present invention carries out pre-analysis and processing from the perspective of algorithm adaptability, and comprehensively and meticulously analyzes the characteristics of the phase-solving algorithm, so as to reverse the design of the algorithm and the measurement scheme based on this. Formulated; the innovative technology proposed by the present invention is beneficial to the application of interferometric measurement technology and the development of multi-surface measurement technology, and develops a design idea of measurement technology for those skilled in the field;

4. The present invention reverses the algorithm design, the formulation of the experimental acquisition scheme and the automatic design of the algorithm after obtaining the phase shift value suitable for the current working condition through pre-analysis; it can well complete the phase-solving bottleneck problem under any cavity length. , and because the algorithm is automatically designed, the technical application cost and design cost are reduced, and the application range of the wavelength phase-shifting interference technology is broadened.

Description of drawings

FIG. 1 is a schematic structural diagram of a reference arm of a measurement system according to a preferred embodiment of the present invention.

FIG. 2 is a schematic diagram of the knob structure of the measurement system according to the preferred embodiment of the present invention.

FIG. 3 is a front view of the structure of the guide rail frame of the measurement system according to the preferred embodiment of the present invention.

FIG. 4 is a top view of the structure of the guide rail frame of the measurement system according to the preferred embodiment of the present invention.

FIG. 5 is a left side view of the structure of the guide rail frame of the measuring system according to the preferred embodiment of the present invention.

FIG. 6 is a schematic diagram of the overall system structure of the measurement system according to the preferred embodiment of the present invention.

FIG. 7 is a flowchart of an algorithm adopted by the measurement system according to the preferred embodiment of the present invention.

FIG. 8 is an aliasing interferogram collected by an interference weighted sampling dephase analysis method according to a preferred embodiment of the present invention.

FIG. 9 is a diagram of a phase resolution result obtained by an interference weighted sampling phase resolution analysis method according to a preferred embodiment of the present invention.

FIG. 10 is a residual diagram of a result obtained by an interference weighted sampling dephase analysis method according to a preferred embodiment of the present invention.

FIG. 11 is a schematic diagram of a pre-analysis parameter library adopted by the interference-weighted sampling phase-resolving analysis method according to the preferred embodiment of the present invention.

Detailed ways

The above scheme will be further described below in conjunction with specific embodiments, and preferred embodiments of the present invention are described in detail as follows:

In the present embodiment, referring to FIG. 7 , a pre-analysis method for phase analysis based on interference weighted sampling under arbitrary cavity length includes the following processes and steps:

a. According to the characteristics of the weighted multi-step sampling algorithm, double iteration is performed on the phase-shifting reference coefficient and cavity length coefficient within a certain range;

b. Take the mean square error of the maximum error for the results of the algorithm based on Zernike polynomial interference simulation and multi-step weighted discrete sampling. When the separation error of each information is less than 1.2um, the algorithm is determined to be available, otherwise it is determined to be unavailable, and then the This information is stored in a pre-analysis matrix;

c. Use the laser ranging sensor to measure the distance of the DUT, set the thickness and reflectivity information of the DUT, call the pre-analysis matrix, select and design the specific parameters in the algorithm, and select the available parameters from the algorithm library. In the actual use of this embodiment, because the hardware design part uses the laser ranging sensor to realize the automatic measurement of the distance of the measured object, only the thickness and reflectivity of the measured piece need to be input during operation. , then after calling the pre-analysis matrix, the specific parameters in the algorithm can be automatically selected and designed, and the available values will be selected from the 6 algorithms to be used for weighted multi-step sampling calculation;

d. After the algorithm parameters are selected, the experimental plan is formulated and the interferometer interferogram acquisition is carried out. The number of acquisition frames and the phase shift value are related to the phase shift reference coefficient. When measuring, set the thickness value of the input test piece to be the same as the Refractive index, measure the information of each surface of the tested object. This embodiment is based on the pre-analyzed interferometric weighted sampling dephasing analysis method under any cavity length, at the level of automatic design of the algorithm.

In this embodiment, referring to FIG. 7 , in the process of formulating the iteration and value determination method, for each cavity length coefficient, the phase-shifting reference coefficient value corresponding to the right scale within 1 step is made to satisfy the current For the condition that the phase-shift reference coefficient and its right side are adjacent to the phase-shift reference coefficient, take the scale value on the right side of the scale range where the actual cavity length coefficient is located as the estimated value, and then select the phase-shift reference coefficient based on the estimated value; After the positioning and selection of the reference coefficient and the cavity length coefficient, the calculation of the sampling function is carried out to complete the preparation process of the algorithm design.

In this embodiment, referring to FIG. 7 , the aliasing interferometry of the multi-surface transparent DUT is performed. By performing double iteration on different cavity length coefficients and phase-shifting reference coefficients, the phase-shifting reference coefficients are firstly iterated, and then the iteration In the process, the iteration of the cavity length coefficient is included in the calculation process, so as to obtain the unwrapped phase results of each surface solution under different cavity length coefficients and different phase shifting reference coefficients, and then solve the surface shape results of each surface, and mainly consider The three information of the front surface, back surface and thickness change of the tested part.

In this embodiment, in terms of hardware, the laser ranging sensor 20, the matching device of the clamp combination and the guide rail are used to realize the high-precision realization of the automation algorithm.

Referring to FIGS. 1 to 7 , a measurement system based on pre-analysis interference weighted sampling with arbitrary cavity lengths, using the pre-analysis-based interference weighted sampling and phase-resolving analysis method with arbitrary cavity lengths in this embodiment, the measurement system mainly includes The laser ranging sensor 20, the matching device of the clamp assembly 18 and the guide rail 19, the clamp assembly 18 includes a clamp and a guide rail frame 7;

Both ends of the guide rail frame 7 are provided with adjustable reference arms 1. The adjustable reference arm 1 has a rack 3 on the inside and a knob 4 with teeth 5 for cooperative transmission; 8. Screw connection, knob 4 and adjustable reference arm 1 at each end of the guide rail frame 7; the adjustable reference arm 1 is coated with high reflectivity coating 2 according to the laser wavelength to cooperate with the work of the laser ranging sensor 20;

The mirror under test 17 is clamped by the clamp combination 18; after the clamp clamps the reference mirror 16, it is combined with the guide rail frame 7 through the clamp assembly threaded hole 11; The frame 7 is clamped on the guide rail 19 and performs coordinated movement; the upper part of the guide rail frame 7 opens the guide frame assembly threaded hole 10 to assemble the upper and lower structures of the guide rail frame through the bolts 9; in the grooves on both sides of the rail frame 7 and move;

The adjustable reference arm 1 is controlled to move back and forth by rotating the knob 4, thereby realizing the correction with the front surface of the measured mirror 17 and the distance measurement of the laser ranging sensor 20; the laser ranging sensor 20 is arranged on each of the front and rear sides of the bottom of the interferometer. One, taking the outer surface of the reference mirror 16 as the benchmark, and measuring the average distance of the calibrated adjustable reference arm 1 as the absolute distance value between the measured mirror 17 and the reference mirror 16, so as to obtain the cavity length value; laser ranging sensor 20 is connected with the upper computer signal, and the computer reads the average value of the measured distance value as the data information for calculating the cavity length coefficient.

In this embodiment, referring to FIGS. 3 to 6 , the object to be tested is clamped by the fixtures of the fixture assembly 18, so that the transparent parallel plate to be measured can be perpendicular to the worktable, so that the parallel plate to be measured is connected to the reference mirror 16 of the interferometer. The outer surfaces are parallel, and by adjusting the position of the fixture, the front and rear surfaces of the tested object and the outer surface of the reference mirror 16 are respectively interfered.

In this embodiment, referring to FIGS. 3-6 , the bottom end of the guide rail frame 7 cooperates with the guide rail 19 on the worktable to move, and the vertical height center of the fixture is aligned with the center of the outer surface of the interferometer.

With reference to the accompanying drawings and detailed descriptions of specific parts in the drawings:

The wavelength phase-shifting interferometer can separate the independent components of the aliased interference signal from the acquired interference image by setting different phase-shifting values and acquisition frames according to the weighted multi-step sampling algorithm. The method in this embodiment is aimed at transparent The parallel plate can realize non-contact simultaneous measurement of each surface of the DUT. The phase of the wavelength-tunable phase-shifting initial interference signal is expressed as:

θ ₀ is the initial phase of the component surface, h(x, y) is the geometric change of the component surface topography, that is, the distribution of the optical path difference. It can be seen from the above formula that the shape of the sample surface can be obtained by measuring the initial phase value. When the wavelength changes, the phase value θ(x,y) and its Taylor series expansion can be expressed as:

Among them, λ ₀ is the starting wavelength of the laser, k is the number of phase shift steps, that is, the number of captured images, and Δλ is the tuning amount of a single wavelength change corresponding to one phase shift. Three main signals are mainly considered here, namely the front surface signal, the back surface signal and the thickness variation signal. T is the average thickness of the element to be measured, and the signal frequency v _p,q (x, y) of each group of interference fringes formed by the reflection of the measuring beam can be expressed as the following general formula:

为折射率随波长的变化率。所述各表面干涉光强信号是时间域内的周期信号，因此通过欧拉公式和傅里叶变换可以变换为频率域，表示为：where p and q are integers reflecting the double beam paths, n is the refractive index of the element to be measured,

is the rate of change of refractive index with wavelength. Each surface interference light intensity signal is a periodic signal in the time domain, so it can be transformed into the frequency domain through Euler's formula and Fourier transform, which is expressed as:

Among them, W(v) is the Fourier transform form of the window function w(t) and needs to satisfy the condition that the value is 0 at one frequency and two frequency. Then the solution of the phase distribution to be measured can be expressed as:

In the actual measurement process, the phase-shifting interferometry technique can be regarded as a discrete sampling of the whole based on a few smoothly arranged variables. In the case of discrete sampling, the relationship between the window function w(t) and the phase shift value σ _k is:

Among them, k represents the number of phase shifts, and δ[] represents the Dirac delta function. The light intensity interference signal can be rewritten in the form of Fourier transform, and the phase to be measured can be expressed as:

称为移相算法的特征系数，常数const是当v＝0时传输窗口的复相位。以上是加权多步采样测量技术的理论基础。in,

Called the characteristic coefficient of the phase-shifting algorithm, the constant const is the complex phase of the transmission window when v=0. The above is the theoretical basis of the weighted multi-step sampling measurement technique.

When using this technique for actual measurement, it is necessary to solve the pre-analysis matrix first. After solving the pre-analysis matrix, it can be stored for direct use in subsequent measurements, without the need to solve the matrix for each measurement. The solution process of the algorithm pre-analysis matrix is very simple, as shown in the algorithm flow chart in Figure 7:

The first step is to set the main follow function based on experience. Here, based on different cavity length coefficients and phase shift reference coefficients, 6 groups of sampling weight functions are used. It can completely simulate the multi-surface interference acquisition process and the algorithm processing process in a combined way. Through the double iteration of different cavity length coefficients and phase-shifting reference coefficients, the phase-shifting reference coefficients are firstly iterated. As well as the phase results after de-wrapping of each surface under different phase-shifting reference coefficients, and then solve the surface shape results of each surface. Here, the three information of the front surface, rear surface and thickness change of the tested part are mainly considered. After de-slope processing is performed on each surface shape result, the residual is obtained. The residual here is the peak distance of each data deviating from the mean value, which is the largest average difference. Residual data are stored in a pre-analysis matrix and decisions and magnitudes are made. The judgment condition is selected as follows: in the double iteration, under the condition of the current cavity length coefficient and the phase shift reference coefficient, the residual results of the three interference information of the current surface, the back surface and the thickness change are simultaneously less than 1.2nm, then it is determined that the Conditional algorithms apply. After the double iterative results of the cavity length coefficient and the phase-shifting reference coefficient are solved, they are placed in the pre-analysis matrix to cover the old parameters, and the pre-analysis parameter library shown in Fig. 11 can be obtained.

When using this algorithm, a high-precision laser ranging sensor is used to measure the distance from the measured object to the reference mirror and read this value into the phase solution algorithm designed in this paper, and then only need to input the values of thickness and refractive index to automatically Calculate the cavity length coefficient and the phase shift reference coefficient. In the process of solving the pre-analysis matrix, because the calculation process under different cavity length coefficients is inevitably discrete data, the processing method selected by the method in this embodiment is: searching for the actual cavity length coefficients, there are always two The discrete value makes the actual cavity length coefficient between the two, so the larger value next to it is selected as the estimated value, and then the phase-shifting reference coefficient is searched for the estimated value to complete the preparation process of the algorithm. In notation, N represents the phase-shifting reference coefficient, and M represents the cavity length coefficient, which is the ratio of the cavity length from the front surface to the reference mirror to the optical path length of the thickness change.

The total acquisition frame number H under different N and the sampling basis functions V ₁ -V ₆ are attached later.

k=1...H, j is an imaginary unit, so the sampling function can be expressed as:

where V _i ' is the normalized (0-1) result of the sampled basis function V _i , where i=1...6. The next step is to bring the current adaptive phase-shifting reference coefficients and actual cavity length values searched from the pre-analysis matrix into the dephasing algorithm, and design the phase-shifting value and the number of interferogram acquisition frames and weighting through different phase-shifting reference coefficients. Sampling function, and then calculate the arctangent of the data to obtain the wrapping phase. After taking wrapping and de-tilting, the surface shape of the corresponding measured surface can be obtained by simple calculation, and the multi-surface interferometric measurement process can be completed. During the search, because the iteration parameters are discrete when constructing the pre-analysis matrix, the iteration step size of N is 1, and the iteration step size of M is 0.1, so it must not be able to satisfy all the cavity length coefficient cases. According to the schematic diagram of the pre-analysis parameter library in Fig. 11, for a certain cavity length coefficient, the N value corresponding to the right scale within 1 step must satisfy the conditions of N and N+1, so the processing adopted here The method is to take the right scale value of the scale range where the actual cavity length coefficient is located as the estimated value, and then select N based on the estimated value. After the positioning and selection of the phase-shifting reference coefficient N and the cavity length coefficient M, the calculation of the sampling function can be performed, and the algorithm design process is completed at this time.

The total number of acquisition frames, the phase shift value of each frame and the corresponding total number of acquisition frames H set based on experience can be determined through the N value selected above, and the phase shift value is 2π/N.

It can be seen from FIG. 6 that the thin line frame in the figure represents the interior of the interferometer. When the interferometer is working, a tunable beam is emitted by the wavelength-tunable laser 12, and is divided into two beams by the beam splitter prism 13, and one beam passes through the second lens. 21, the diaphragm 22 and the third lens 23 enter the CCD camera 24 of the observation system. After a part of the first lens 14 and the collimating lens 15 are collimated, the light beam passes through the reference mirror 16 and is transmitted to the measured mirror 17, where it is reflected and then The reverse optical path enters the observation system, and the interference occurs between the light beams. At this time, the observation system can collect the interference light intensity cross fringes with multi-surface interference information aliasing, as shown in Figure 8. The interferogram collected by the CCD camera 24 is connected with the computer through its own data line for data transmission and reading and writing, so that the algorithm in the computer can directly act on the interferogram.

The mirror under test 17 is clamped by a clamp assembly 18 , wherein the clamp assembly 18 includes a clamp and a guide rail frame 7 . As can be seen from FIG. 4 , after the fixture clamps the mirror 16 under test, the fixture is assembled with the guide rail frame 7 through the threaded hole 11 of the fixture. The structure of the guide rail frame 7 can be seen from Figure 3, Figure 4, and Figure 5. The guide rail frame 7 is composed of two upper and lower hollow cavities with a certain thickness. The upper part of the guide rail frame 7 opens the guide rail frame mounting screw hole 10 through the bolt 9 to assemble the upper and lower structure of the guide rail frame. It can be seen from FIG. 5 that the reference arm 1 can be clamped in the grooves on both sides of the guide rail frame 7 through the reference arm guide grooves 25 at both ends and move. A knob 4 is placed on each of the left and right sides of the guide rail frame 7, and the lower part thereof is assembled through its own screw 6 and the threaded hole 8 on the frame body.

And it can be seen from Figure 2 that there are teeth 5 on the knob 4. It can be seen from Figure 1 that the knob 4 can cooperate with the rack 3 placed on the reference arm 1 at both ends of the frame body, and the surface on the reference arm 1 can be matched. Coating material 2 is applied to the part of the interferometer, and this coating can improve the reflectivity of the material, so as to cooperate with the laser ranging. Therefore, by rotating the knob 4 , the reference arm 1 can be moved back and forth, so as to realize the correction with the front surface of the measured mirror 17 and the distance measurement of the laser ranging sensor 20 . The laser ranging sensor 20 can be pasted on the front and rear sides of the bottom of the interferometer, and both take the outer surface of the reference mirror 16 as a benchmark, and measure the average distance of the reference arm 1 after calibration as the distance between the measured mirror 17 and the reference mirror 16. The absolute distance value, the cavity length value can be obtained, which is brought into the algorithm design process. At the same time, the laser ranging sensor 20 is connected to the computer through its own data line, and the computer reads the average value of the measured distance values as necessary data for calculating the cavity length coefficient.

After completing the above process, the algorithm pre-analysis and data acquisition are completed. The next step is to directly input the thickness and refractive index of the measured part in the calculation during the measurement, and read the current actual cavity length value, which is Measured by the ranging sensor, the acquired interferogram can be calculated by the weighted sampling function that matches the current cavity length value, and the phase solution result can be obtained. After simple calculation, the topography distribution of each surface of the measured element can be obtained. . As shown in Fig. 9 and Fig. 10, the residual unit of Fig. 10 is um. It can be seen from the results that the technology proposed by the present invention can well realize the multi-surface interference phase solution and measurement process of the transparent parallel plate, and the residual is in the sub-sub At the nanometer level, the measurement accuracy is very high.

Attachment: The specific composition of V ₁ -V ₆ corresponding to different N:

N=9, number of acquisition frames H=57, V ₁ :

V ₁ ＝[0.2222,1.5556,6.2222,18.6667,46.6667,102.6667,205.3333,381.3333,667.3333,1241.3333,1899.3333,2837.3333,4125.3333,5833.3333,8025.3333,10752,14042,17892,22129.3333,26931.3333,32041.3333,37294.6667,42494.6667, 47422.6667,51851.3333,55561.3333,58361.3333,60111.3333,60749.3333,60111.3333,58361.3333,55561.3333,51851.3333,47422.6667,42494.6667,37294.6667,32041.3333,26931.3333,22129.3333,17892,14042,10752,8025.3333,5833.3333,4125.3333,2837.3333,1899.3333,1241.3333, 667.3333, 381.3333, 205.3333, 102.6667, 46.6667, 18.6667, 6.2222, 1.5556];

N=10, number of acquisition frames H=64, V ₂ :

V ₂ = [0.2, 1.4, 5.6, 16.8, 42, 92.4, 184.8, 343.2, 600.6, 1001, 1894.2, 2759.4, 3967.6, 5602.8, 7752, 10500, 13923, 18081, 23009, 28707, 34839, 419 57372,65373,73227,80661,87395,93159,97713,100870,102522,102522,100870,97713,93159,87395,80661,73227,65373,57372,49476,41905,34839,28707,23009,18081,13923, 10500,7752,5602.8,3967.6,2759.4,1894.2,1001,600.6,343.2,184.8,92.4,42,16.8,5.6,1.4,0.2];

N=11, the number of acquisition frames H=71, V ₃ :

V ₃ ＝[1.2727,5.0909,15.2727,38.1818,84,168,312,546,910,1456,2836.9091,3954.3636,5485.4545,7528.3636,10186.9091,13566,17766,22876,28966,36078,44216,52750.7273,62767.0909,73534.3636,84859.0909,96500.7273,108178, 119578,130368,140210,148778,155778,160970.7273,164197.0909,165406.3636,164197.0909,160970.7273,155778,148778,140210,130368,119578,108178,96500.7273,84859.0909,73534.3636,62767.0909,52750.7273,44216,36078,28966,22876, 17766,13566,10186.9091,7528.3636,5485.4545,3954.3636,2836.9091,1456,910,546,312,168,84,38.1818,15.2727,5.0909,1.2727,0.1818];

N=12, the number of acquisition frames H=78, V ₄ :

V ₄ ＝[0.1667,1.1667,4.6667,14,35,77,154,286,500.5,834.1667,1334.6667,2062.6667,4170.8333,5591.8333,7505.3333,10024,13268.5,17363.5,22432.6667,28592.6667,35946.1667,44573.8333,54525.3333,65809.3333,77308,91084, 105910,121580.6667,137837.1667,154372.1667,170837.3333,186853.3333,202022.3333,215943,228228,238524,246534.1667,252043.1667,254944.6667,254944.6667,252043.1667,246534.1667,238524,228228,215943,202022.3333,186853.3333,170837.3333,154372.1667,137837.1667,121580.6667, 105910,91084,77308,65809.3333,54525.3333,44573.8333,35946.1667,28592.6667,22432.6667,17363.5,13268.5,10024,7505.3333,5591.8333,4170.8333,2062.6667,1334.6667,834.1667,500.5,286,154,77,35,14,4.6667,1.1667, 0.1667];

N=13, the number of acquisition frames H=85, V ₅ :

V ₅ ＝[0.1538,1.0769,4.3077,12.9231,32.3077,71.0769,142.1538,264,462,770,1232,1904,2856,6021.0769,7803.5385,10166.1538,13236.4615,17152.1538,22057.5385,28099.0769,35420,44154,54418,66304,79870, 95130,110198.3077,128676.1538,148560.6154,169635.8462,191626.6154,214202.1538,236982.3077,259546,281442,302202,321356,338450,353066,364844.3077,373508.1538,378890.6154,380963.8462,378890.6154,373508.1538,364844.3077,353066,338450,321356,302202, 281442,259546,236982.3077,214202.1538,191626.6154,169635.8462,148560.6154,128676.1538,110198.3077,95130,79870,66304,54418,44154,35420,28099.0769,22057.5385,17152.1538,13236.4615,10166.1538,7803.5385,6021.0769,2856,1904,1232, 770,462,264];

N=14, the number of acquisition frames H=92, V ₆ :

V ₆ = [0.1429,1,4,12,30,66,132,245.1429,429,715,1144,1768,2652,3876,8539.1429,10748,13634,17340,22021,27841,34,969,43570.1495829,58 113895,134095,153372.1429,177637,203727,231417,260417,290375,320882,351479.1429,381667,410917,438685,464427,487617,507767,524449.1429,537320,546147,550837,550837,546147,537320,524449.1429,507767, 487617,464427,438685,410917,381667,351479.1429,320882,290375,260417,231417,203727,177637,153372.1429,134095,113895,95810,79820,65858,53820,43574.1429,34969,27841,22021,17340,13634, 10748, 8539.1429].

In this embodiment, an interferometric weighted sampling dephasing analysis method and measurement system based on pre-analysis with arbitrary cavity lengths to deal with the aliasing interferometry problem of multi-surface transparent DUTs, including algorithm design and hardware implementation. The automation design level is mainly divided into the following processes. Firstly, according to the characteristics of the weighted multi-step sampling algorithm, double iteration is performed on the phase-shifting reference coefficient and cavity length coefficient within a certain range; then the mean square error of the maximum error is obtained for the algorithm results based on Zernike polynomial interference simulation and multi-step weighted discrete sampling. To judge, when the separation error of each information is less than 1.2um, the algorithm is judged to be available, otherwise it is judged to be unavailable, and then the information is stored in the pre-analysis matrix; in actual use, because the hardware design part uses laser ranging sensors Realize the automatic measurement of the distance of the DUT, so you only need to input the thickness and reflectivity information of the DUT during operation. After calling the pre-analysis matrix, the specific parameters in the algorithm can be automatically selected and designed. In the algorithm, the available values are selected for weighted multi-step sampling calculation; then after the algorithm parameters are selected, the experimental scheme is formulated and the interferometer interferogram acquisition is carried out, in which the number of acquisition frames and the phase shift value are related to the phase shift reference coefficient. When measuring, only need to input the thickness value and refractive index of the measured part, and then the measurement of each surface information can be performed automatically. In terms of hardware: use the laser ranging sensor, the combination of the fixture and the guide rail to realize the high-precision realization of the automation algorithm. The fixture combination includes two parts: the fixture and the guide rail frame. Algorithm design and experimental acquisition scheme formulation and algorithm automatic design are carried out in reverse after the phase-shift value that is suitable for the current working condition obtained by pre-analysis. The technology can well solve the phase bottleneck problem under any cavity length, and because the algorithm is automatically designed, the technical application cost and design cost are reduced, and the application scope of the wavelength-shifting interferometry technology is broadened. The method and test system of the present embodiment solve the problem of phase resolution in interferometric measurement, especially the problem of measurement limitation that an algorithm in the weighted multi-step sampling algorithm is intelligently applied to a narrow range of discrete cavity length conditions. The method and test system of the present embodiment automatically design a phase-resolving technical scheme for algorithm parameters by pre-analyzing the phase-resolving algorithm, including the algorithm library design method, the algorithm and the automatic design technology of the experimental scheme; it can well complete any cavity length. And because the algorithm is automatically designed, the technical application cost and design cost are reduced, and the application scope of the wavelength phase-shifting interference technology is broadened. The method is simple, easy to implement, low in cost, and suitable for promotion.

The embodiments of the present invention have been described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned embodiments, and various changes can also be made according to the purpose of the invention and creation of the present invention. Changes, modifications, substitutions, combinations or simplifications should be equivalent substitution methods, as long as they meet the purpose of the present invention, as long as they do not deviate from the pre-analysis-based interference-weighted sampling dephasing analysis method and measurement for any cavity length in the present invention The technical principle and inventive concept of the system all belong to the protection scope of the present invention.

Claims (6)

1. an arbitrary cavity length based on pre-analysis, an interference weighted sampling dephase analysis method, is characterized in that, comprises the following process and steps: a. According to the characteristics of the weighted multi-step sampling algorithm, double iteration is performed on the phase-shifting reference coefficient and cavity length coefficient within a certain range; b. The algorithm results based on Zernike polynomial interference simulation and weighted multi-step sampling algorithm are judged by the mean square error of the maximum error. When the separation error of each information is less than 1.2um, the algorithm is judged to be available, otherwise it is judged to be unavailable, and then the This information is stored in a pre-analysis matrix; c. Use the laser ranging sensor to measure the distance of the DUT, set the thickness and reflectivity information of the DUT, call the pre-analysis matrix, select and design the specific parameters in the algorithm, and select the available parameters from the algorithm library. value for weighted multi-step sampling calculation; d. After the algorithm parameters are selected, the interferogram acquisition of the interferometer is carried out. The number of acquisition frames and the phase-shift value are related to the phase-shift reference coefficient. During the measurement, set the thickness value and refractive index of the input test piece to be measured. Measurement of each surface information of the test piece. 2. The interferometric weighted sampling solution phase analysis method based on pre-analysis according to claim 1, characterized in that: in the process of formulating iteration and value determination method, for each cavity length coefficient, make 1 The phase-shifting reference coefficient value corresponding to the right scale within the step distance satisfies the condition of the current phase-shifting reference coefficient and its adjacent phase-shifting reference coefficient on the right side, and the right scale value of the scale range where the actual cavity length coefficient is located is taken as the estimated value , and then select the phase-shift reference coefficient through the estimated value; after the positioning and selection of the phase-shift reference coefficient and the cavity length coefficient, the calculation of the sampling function is performed to complete the algorithm design preparation process. 3. The interferometric weighted sampling solution phase analysis method based on pre-analysis according to claim 1, characterized in that: carrying out the aliasing interferometry of the multi-surface transparent DUT, by comparing different cavity length coefficients and shifts. The phase reference coefficients are double-iterated. First, the phase-shift reference coefficients are iterated. In the iterative process, the iteration of the cavity length coefficients is included in the calculation process, so as to obtain the solution of each surface under different cavity length coefficients and different phase-shift reference coefficients. To wrap the later phase results, and then solve the surface shape results, and mainly consider the three information of the front surface, the back surface and the thickness change of the tested part. 4. A measurement system based on the interference weighted sampling under the arbitrary cavity length based on pre-analysis, adopts the interference weighted sampling solution phase analysis method based on the arbitrary cavity length based on the pre-analysis of claim 1, it is characterized in that: the described measurement system mainly A matching device comprising a laser ranging sensor (20), a clamp assembly (18) and a guide rail (19), the clamp assembly (18) comprising a clamp and a guide rail frame (7); Adjustable reference arms (1) are provided at both ends of the guide rail frame (7), and the adjustable reference arm (1) has a rack (3) on the inner side to cooperate with a knob (4) with teeth (5) for transmission; the bottom of the knob (4) The end is a screw (6), which is threadedly connected to the threaded hole (8) at the inner bottom end of the guide rail frame (7). The knob (4) and the adjustable reference arm (1) are located at each end of the guide rail frame (7); adjustable The reference arm (1) is coated with a high reflectivity coating (2) according to the wavelength of the laser, so as to cooperate with the work of the laser ranging sensor (20); The mirror under test (17) is clamped by the fixture assembly (18); after the reference mirror (16) is clamped by the fixture, it is combined with the guide rail frame (7) through the clamp assembly screw hole (11); It is composed of a hollow cavity, and the bottom is slotted so that the guide rail frame (7) can be clamped on the guide rail (19) and cooperated with the movement; the upper part of the guide rail frame (7) is opened with threaded holes (10) for assembly of the guide rail frame through bolts (9). Assembly of the upper and lower structures of the guide rail frame; the adjustable reference arm (1) is clamped in the grooves on both sides of the guide rail frame (7) through the reference arm guide grooves (25) at both ends and moves; The adjustable reference arm (1) is controlled by rotating the knob (4) to move back and forth, so as to realize the correction with the front surface of the measured mirror (17) and the distance measurement of the laser ranging sensor (20); the laser ranging sensor (20) One is arranged on the front and back sides of the bottom of the interferometer, and the outer surface of the reference mirror (16) is used as the benchmark to measure the average distance of the calibrated adjustable reference arm (1) as the measured mirror (17) and the reference mirror. The absolute distance value of (16) is obtained to obtain the cavity length value; the laser ranging sensor (20) is connected to the upper computer signal, and the computer reads the average value of the measured distance value as the data information for calculating the cavity length coefficient. 5. The measurement system based on the interference weighted sampling under the arbitrary cavity length of the pre-analysis according to claim 4, is characterized in that: the measured part is clamped by the clamp of the clamp combination (18), so that the measured transparent parallel plate can be vertically On the workbench, make the parallel plate under test parallel to the outer surface of the reference mirror (16) of the interferometer, and adjust the position of the fixture so that the front and rear surfaces of the test piece and the outer surface of the reference mirror (16) interfere respectively. . 6. The measurement system of interference weighted sampling based on pre-analyzed arbitrary cavity length according to claim 4, characterized in that: the bottom end of the guide rail frame (7) cooperates with the guide rail (19) on the worktable to move, The vertical height center of the fixture is aligned with the center of the outer surface of the interferometer.

Images