A CNC machining method for small-scale mirror errors

Abstract

The present invention discloses a CNC machining method for small-scale errors of a mirror, the method comprising performing wavelet decomposition based on a residual surface shape to obtain a reconstruction coefficient; performing inverse wavelet transform on the reconstruction coefficient to reconstruct the surface shape, extract the positive value part, and calculate the residual; repeating the aforementioned steps on the residual, and stopping the iteration if the maximum value of the extracted positive value part is less than a threshold value δ to obtain a small-scale error surface shape; performing binarization, opening operation, and closing operation on the small-scale error surface shape to obtain an area where the small-scale error is located, and generating a final machining path according to the distribution of the area; selecting a suitable polishing tool, and controlling a CNC machine tool or a mechanical arm to use the polishing tool to perform local polishing on the small-scale error of the mirror according to the generated path. The mirror polishing method based on small-scale error area extraction proposed by the present invention performs targeted corrections on small-scale errors occurring in mirror machining without polishing the entire mirror, thereby optimizing the polishing effect and improving machining efficiency.

Description

Technical Field

[0001] The invention belongs to the field of optical mirror processing, and in particular relates to a numerical control processing method for small-scale mirror errors. Background Art

[0002] High-precision, large-aperture optical mirrors play an important role in several cutting-edge fields. They are one of the core components with the longest processing cycle and the most technical content in related equipment. Current applications mainly include: (1) Astronomy. At the end of the last century, two 10-meter-diameter Keck Telescopes (Keck Telescope, USA) and four 8.2-meter Very Large Telescopes (VLT) were built in the world. With the help of advanced optical telescopes, a number of major achievements at the level of Nobel Prize in Physics have been made internationally; (2) Defense and space fields. In recent years, the demand for two types of difficult mirrors has increased sharply. One is high-steepness aspheric surfaces. The huge asphericity puts high demands on the grinding and polishing efficiency; the other is long strip reflectors. Compared with traditional circular mirrors, their rotational asymmetry brings great difficulties to automated processing. In addition, both often have high surface quality requirements, which further increases the difficulty of processing.

[0003] In the 1970s, computer-controlled optical surface shaping technology (CCOS) was proposed and gradually developed. This technology uses a small polishing tool controlled by a computer to grind or polish optical components based on quantitative surface shape data, so that the surface shape error gradually converges and eventually meets the accuracy requirements. Due to its determinism and stability, this technology has been widely used.

[0004] When CCOS technology is applied for full-aperture processing, since the size of the polishing tool is much smaller than the size of the workpiece, and the convolution effect of the processing path, while the surface error converges, periodic small-scale errors will gradually appear. This error will cause light scattering, thereby affecting the imaging contrast of the optical system.

[0005] Small-scale errors usually appear as periodic rings or regional fragments. If a smaller tool is used to directly perform full-caliber corrections, the processing efficiency will be reduced. Therefore, it is necessary to make targeted corrections to local small-scale errors to optimize the grinding and polishing effect and improve processing efficiency. Summary of the invention

[0006] In view of the fact that small-scale errors in the form of rings or fragments often occur during mirror processing, the present invention aims to propose a CNC processing method for small-scale errors of mirrors. This method is used to extract small-scale errors and determine local processing paths to achieve targeted corrections, optimize the polishing effect, and improve processing efficiency.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A CNC machining method for small-scale errors of a mirror surface, wherein the small-scale errors of the mirror surface are first extracted and the area where the small-scale errors are located is individually corrected, and the mirror surface after the small-scale errors are corrected is then subjected to full-aperture grinding and polishing; the small-scale error extraction and correction method comprises:

[0009] Step 1: Read the residual surface Z1(x, y), perform n-layer wavelet decomposition on it, obtain the coefficient matrix of each layer, and select the m coefficient matrices with the smallest size as the reconstruction coefficients;

[0010] Step 2: Perform inverse wavelet transform on the reconstruction coefficients obtained in step 1, reconstruct the surface shape, extract the positive part h1(x, y), and calculate the residual Z2(x, y). The formula is as follows:

[0011] Z2(x,y)=Z1(x,y)-h1(x,y);

[0012] Step 3: Repeat steps 1 and 2 for the residual Z2(x, y) obtained in step 2, and determine whether the maximum value of the extracted positive part is less than the threshold δ. If it is less than the threshold δ, stop the iteration and calculate the small-scale error surface shape H(x, y). The formula is as follows:

[0013]

[0014] Where k is the number of iterations;

[0015] Step 4: Binarize, open and close the small-scale error surface H(x, y) obtained in step 3 to obtain the area where the small-scale error is located;

[0016] Step 5: Generate intra-region processing paths and plan inter-region processing paths for the regions extracted in step 4, and integrate them to obtain the final processing path;

[0017] Step 6: Refer to the area size extracted in step 4, select a suitable grinding and polishing tool, set the processing path obtained in step 5 on the computer, and control the CNC machine tool or robotic arm to use the grinding and polishing tool to perform local grinding and polishing on the small-scale errors on the mirror surface.

[0018] Furthermore, in step 1, the number of wavelet decomposition layers is not a fixed value, and is preliminarily set in combination with the size of the residual matrix, and then adjusted according to the obtained reconstructed surface shape; the number m of coefficient matrices selected as reconstruction coefficients is not a fixed value, and is set and modified in combination with the reconstructed surface shape.

[0019] Furthermore, in step 3, the threshold δ is not a fixed value, but is set according to the change of the maximum value of the positive part extracted in each round of iteration.

[0020] Furthermore, in step 4, the threshold of the binarization process and the size of the structural element of the opening and closing operations are set according to the small-scale error area, and then adjusted according to the extraction result.

[0021] Furthermore, in step 6, after the targeted correction of small-scale errors is completed, the mirror surface can continue to be ground and polished at the full aperture.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] Compared with the current mainstream CNC small tool technology for full-aperture grinding and polishing correction of mirror surfaces, the present invention proposes a mirror polishing method based on small-scale error area extraction, which can make targeted corrections to small-scale errors occurring in mirror processing without the need to polish the entire mirror surface, thereby optimizing the polishing effect and improving processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the method of the present invention.

[0025] Figure 2 is the residual surface diagram.

[0026] Figure 3 It is the extracted small-scale error surface map.

[0027] Figure 4 It is a binary image of small-scale error area.

[0028] Figure 5 It is a path planning map. DETAILED DESCRIPTION

[0029] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0030] like Figure 1 As shown, the present invention is a CNC machining method for small-scale errors of a mirror surface. During machining, the small-scale errors of the mirror surface are first extracted and the area where the small-scale errors are located is individually corrected. After the small-scale errors are corrected, the mirror surface is further subjected to full-aperture grinding and polishing. The method for extracting and correcting small-scale errors includes:

[0031] Step 1: Read the residual surface Z1(x, y), perform n-layer wavelet decomposition on it, obtain the coefficient matrix of each layer, and select the m coefficient matrices with the smallest size as the reconstruction coefficients;

[0032] Among them, the number of wavelet decomposition layers is not a fixed value. It is preliminarily set in combination with the size of the residual matrix, and then adjusted according to the reconstructed surface shape, generally set to 4 to 5 layers. The number m of coefficient matrices selected as reconstruction coefficients is not a fixed value. It is set and modified in combination with the reconstructed surface shape, and generally 2 are selected.

[0033] Step 2: Perform inverse wavelet transform on the reconstruction coefficients obtained in step 1, reconstruct the surface shape, extract the positive part h1(x, y), and calculate the residual Z2(x, y). The formula is as follows:

[0034] Z2(x,y)=Z1(x,y)-h1(x,y);

[0035] Step 3: Repeat steps 1 and 2 for the residual Z2(x, y) obtained in step 2, and determine whether the maximum value of the extracted positive part is less than the threshold δ. If it is less than the threshold δ, stop the iteration and calculate the small-scale error surface shape H(x, y). The formula is as follows:

[0036]

[0037] Where k is the number of iterations.

[0038] The threshold δ is not a fixed value, but is set according to the change of the maximum value of the positive part extracted in each round of iteration.

[0039] Step 4: Binarize, open and close the small-scale error surface H(x, y) obtained in step 3 to obtain the area where the small-scale error is located;

[0040] Among them, the threshold of binarization processing and the size of the structural element of the opening and closing operation are set according to the small-scale error area and then adjusted according to the extraction result.

[0041] Step 5: Generate intra-region processing paths and plan inter-region processing paths for the regions extracted in step 4, and integrate them to obtain the final processing path;

[0042] Step 6: Refer to the area size extracted in step 4, select a suitable grinding and polishing tool, set the processing path obtained in step 5 on the computer, and control the CNC machine tool or robotic arm to use the grinding and polishing tool to perform local grinding and polishing on the small-scale errors on the mirror surface.

[0043] After the targeted correction of small-scale errors, the mirror surface can continue to be ground and polished at the full aperture.

[0044] When extracting small-scale error surface shape, first read the original residual surface shape Z1(x,y), such as Figure 2As shown. Perform wavelet decomposition on it, the size of the residual matrix is ​​500*500, set the number of decomposition layers to 4 layers, the coefficient matrix size of the 3rd and 4th layers is the smallest, and is selected as the reconstruction coefficient. Perform inverse wavelet transform on the reconstructed coefficients, reconstruct the surface shape, extract the positive part h1(x,y), and calculate the residual Z2(x,y). Set the threshold δ to 0.01, repeat the wavelet decomposition, surface reconstruction and positive value extraction, and stop the iteration when the maximum value of the extracted positive part is less than δ. Iterate 24 times in total and calculate the small-scale error surface shape H(x,y), as shown Figure 3 shown.

[0045] When extracting the small-scale error area, first binarize the small-scale error surface H(x,y), and set the binarization threshold to 0.65 (normalized threshold) based on the small-scale error distribution. Then perform the opening and closing operations, and select the circle as the structural element with a radius of 1. Finally, the binary map of the small-scale error area is obtained, as shown in Figure 4 shown.

[0046] right Figure 4 The regions generate processing paths within the region and plan processing paths between regions respectively, and integrate them to obtain the final processing path, such as Figure 5 shown.

[0047] In summary, the CNC machining method for small-scale errors of mirrors of the present invention performs wavelet decomposition, surface reconstruction, positive value extraction, and iteration for small-scale errors occurring in the mirror finishing stage, so as to realize the extraction of small-scale error surface shapes. After the surface shape extraction, binarization and opening and closing operations are performed to realize the extraction of small-scale error areas. Based on this, a grinding and polishing tool is selected, a machining path is generated, and small-scale errors are ground and polished. The present invention realizes targeted correction of small-scale errors, optimizes the grinding and polishing effect, and improves the machining efficiency of the mirror.

[0048] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A CNC machining method for small-scale errors of a mirror surface, wherein the small-scale errors of the mirror surface are first extracted and the area where the small-scale errors are located is individually corrected, and the mirror surface after the small-scale errors are corrected is then subjected to full-aperture grinding and polishing; characterized in that: The method for extracting and correcting the small-scale error includes: Step 1: Read the residual surface Z1(x, y), perform n-layer wavelet decomposition on it, obtain the coefficient matrix of each layer, and select the m coefficient matrices with the smallest size as the reconstruction coefficients; Step 2: Perform inverse wavelet transform on the reconstruction coefficients obtained in step 1, reconstruct the surface shape, extract the positive part h1(x, y), and calculate the residual Z2(x, y). The formula is as follows: Z2(x,y) = Z1(x,y) - h1(x,y); Step 3: Repeat steps 1 and 2 for the residual Z2(x, y) obtained in step 2, and determine whether the maximum value of the extracted positive part is less than the threshold δ. If it is less than the threshold δ, stop the iteration and calculate the small-scale error surface shape H(x, y). The formula is as follows: H(x,y) = ; Where k is the number of iterations; Step 4: Binarize, open and close the small-scale error surface H(x, y) obtained in step 3 to obtain the area where the small-scale error is located; Step 5: Generate intra-region processing paths and plan inter-region processing paths for the regions extracted in step 4, and integrate them to obtain the final processing path; Step 6: Refer to the area size extracted in step 4, select a suitable grinding and polishing tool, set the processing path obtained in step 5 on the computer, and control the CNC machine tool or robotic arm to use the grinding and polishing tool to perform local grinding and polishing on the small-scale errors on the mirror surface.

2. A CNC machining method for a mirror surface with a small scale error according to claim 1, characterized in that: In step 1, the number of wavelet decomposition layers is not a fixed value. It is preliminarily set in combination with the size of the residual matrix and then adjusted according to the obtained reconstructed surface shape. The number m of coefficient matrices selected as reconstruction coefficients is not a fixed value and is set and modified in combination with the reconstructed surface shape.

3. A CNC machining method for a mirror surface with a small scale error according to claim 1, characterized in that: In step 3, the threshold δ is not a fixed value, but is set according to the change of the maximum value of the positive part extracted in each round of iteration.

4. A CNC machining method for a mirror surface with a small scale error according to claim 1, characterized in that: In step 4, the threshold of the binarization process and the size of the structural element of the opening and closing operations are set according to the small-scale error area and then adjusted according to the extraction results.

5. The method for numerically controlling a mirror surface with a small scale error according to claim 1, characterized in that: In step 6, after the small-scale error is corrected, the mirror surface is subjected to full-aperture grinding and polishing.

Images