Mirror milling processing and measurement control process method, and system thereof

Abstract

To provide a mirror milling processing and measurement control process method and processing system, including a thin-walled component processing and thickness measurement control process and a thin-walled component processing contour detection process.SOLUTION: A thin-walled component processing path program transplantation process comprises: aligning, according to an actual positioning hole in point cloud data obtained by scanning the thin-walled component and a theoretical positioning hole on a theoretical triangle mesh curved surface generated based on a design curved surface, an actual triangle mesh curved surface with the theoretical triangle mesh curved surface; calculating, for a plurality of cutter location points in a cutter location file, geodesic information between each of the cutter location points and the theoretical positioning hole; transplanting, using the geodesic information, the cutter location points to the actual triangular mesh curved surface to form a transplantation processing path.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the field of mirror milling technology, and particularly to a mirror milling machining and measurement control method and system thereof. [Background technology]

[0002] Mirror milling is a processing method for large, thin-walled aircraft body sheets. The difference from conventional milling is that a support and measuring device is applied to the back of the sheet during the processing process, and the milling tool and the support and measuring device aim at the sheet from both sides and process it synchronously. This achieves better product consistency than conventional body sheet processing using chemical etching.

[0003] In the existing technology, there are already technical solutions for milling thin-walled parts such as aircraft body sheets.

[0004] For example, Chinese Patent CN201410532797.X discloses a method and apparatus for mirror milling aircraft body sheets. The apparatus mainly includes a floor frame, a milling device, a vertical conversion device, a clamping device, a flexible suction device, a jacking device, a measuring device, and a thickness measuring device. In this invention, the vertical conversion device is used to clamp the body sheet in a vertical position, thereby improving the clamping efficiency of the body sheet. Furthermore, the addition of a vacuum suction device further improves the positioning accuracy of the body sheet and prevents deformation due to gravity. On-board measurement based on a laser position sensor detects the actual curvature of the body part before processing, and the body processing tool trajectory can be adjusted based on the actual clamping state. This enables self-adaptive numerically controlled processing. The jacking device can support the body part from the back during the body sheet milling process. This avoids vibration during processing and improves processing stability. Furthermore, real-time thickness monitoring during processing allows for thickness compensation. An integrated control system ensures that the various units on the milling bed work together and do not interfere with each other.

[0005] Chinese Patent CN201910811713.9 also discloses a device for detecting milling defects in body sheets. It is related to the field of body sheet processing and aims to detect and locate milling defects in body sheets in real time. The device is installed on a milling bed and includes a PC end, an infrared camera, a Time of Flight (TOF) depth camera, and a magnetic field motion module. The magnetic field motion module and milling tool are connected to the main shaft of the milling bed via a milling head, transmitting the torque of the milling bed. During the body sheet milling process, the main shaft of the milling bed rotates the milling tool and magnetic field motion module, causing the body sheet to cut through the magnetic force of the magnetic field motion module, generating an induced inductance current inside. The infrared camera is used to detect the surface temperature distribution of the body sheet. The Time of Flight (TOF) depth camera is used to reconstruct the three-dimensional shape of the body sheet. The PC end is used to obtain and analyze a three-dimensional temperature distribution map based on the temperature distribution and three-dimensional shape, and to detect internal defects in the body sheet.

[0006] However, in the actual implementation process, the inventors found that due to the constraints of the physical properties of the thin-walled part itself, certain deformations may easily occur during the processing process, such as the central region being raised or depressed. As a result, during the milling process of the thin-walled part, the actual curved surface of the thin-walled part may not be completely consistent with the designed curved surface and tool path, which further affects the processing accuracy. Summary of the Invention

[0007] To solve the above problems in the existing technology, a mirror milling and measurement control method is currently provided, and a machining system for implementing the machining method is also provided.

[0008] The specific technical proposal is as follows:

[0009] The mirror milling and measurement control method includes a thin-walled part clamping and transporting process, a thin-walled part measurement and point cloud acquisition process, a thin-walled part machining path program transfer process, a thin-walled part machining path post-processing process, a thin-walled part machining and thickness measurement control process, and a thin-walled part machining contour detection process.

[0010] The process of transferring machining path programs for thin-walled parts involves: Based on the actual positioning holes in the point cloud data obtained by scanning the thin-walled part and the theoretical positioning holes on the theoretical triangular mesh surface generated based on the design surface, the actual triangular mesh surface corresponding to the point cloud data is matched with the theoretical triangular mesh surface; For a plurality of tool position points in the tool file, calculate geodesic information between each tool position point and the theoretical positioning hole; It includes transplanting the tool position points onto the actual triangular mesh surface based on the geodesic information, and forming the transplanted machining path.

[0011] In addition, in the clamping and conveying process of the thin-walled parts, a clamping process device is used to clamp the thin-walled parts, and the clamping process device has a mouth-shaped process device frame, in which a number of moving columns are distributed; The process equipment frame and the moving support are fitted with clamps that clamp and secure the thin-walled components. Distributed within the process equipment frame are a plurality of jacking devices which move from the rear to the front of the thin wall components to support the thin wall components.

[0012] In addition, the clamping and conveying process for thin-walled parts requires Lifting the thin-walled part onto a process equipment frame and projecting a laser projection onto the frame; The thin-walled part is positioned within the range of the laser projection; Adjust the jacking device and clamping device according to the laser projection and the process device program, and adjust the thin-walled parts to the intended processing state; The thin-walled parts are clamped in order and scanned to obtain point cloud data.

[0013] Furthermore, in the thin-walled part measurement and point cloud acquisition process, a line laser is driven by the machine tool to scan the thin-walled part and construct point cloud data.

[0014] In addition, the process of extracting the tool position points that calculate the tool file involves: Extract the center coordinates of the theoretical positioning holes on the theoretical triangular mesh surface, and extract the information of each tool position point from the tool file; The tool position point information includes coordinates of the tool position point, For each tool position point, project the tool position point onto a theoretical grid surface of the theoretical triangular mesh surface to generate a first projection point and a projection length between the first projection point and the tool position point; The method includes a process of calculating the area coordinates of the first projection points relative to the projection point grid on which the first projection points are located, and calculating the geodesic lengths of the area coordinates for each of the center coordinates of the theoretical positioning holes to obtain the geodesic information.

[0015] In addition, the process of forming a transplant processing path involves: Extract the center coordinates of the actual positioning hole from the point cloud data, and project the tool position point onto the actual grid plane corresponding to the point cloud data to obtain a second projected point; For the second projection point, calculate a predicted projection geodesic length relative to the center coordinate of the actual positioning hole, and then calculate a geodesic deviation value based on the geodesic information and the predicted projection geodesic length; According to the above geodesic deviation value and the preset geodesic deviation range, the second projection point is iteratively processed until it satisfies the geodesic deviation range, and then it is used as the actual transplantation point; A process for generating a transplant processing path based on the actual transplant points is included.

[0016] In addition, the post-processing process of the machining path of the thin-walled parts includes the process of creating a new implantation tool position file based on the implantation machining path and performing simulation inspection; In the thin-walled part machining and thickness measurement control process, the imported tool position file is used to machine the thin-walled part.

[0017] Additionally, the thin-walled part machining and thickness measurement control process uses a jacking device to support the sunk area of ​​the thin-walled part.

[0018] In addition, in the processing and thickness measurement control process of thin-walled parts, an ultrasonic detection head is used to measure the thickness of the above-mentioned thin-walled parts in real time, and the real-time thickness is used in the mirror milling process.

[0019] In addition, the ultrasonic detection head is equipped with a number of eddy current sensors distributed around it, which generate eddy currents during the measurement process of the ultrasonic detection head, allowing the ultrasonic detection head to measure the eddy current intervals for the thin-walled parts; In the mirror milling process, the ultrasonic detection head controls the distance to the thin-walled part based on the eddy current spacing.

[0020] In addition, during the measurement process of the ultrasonic detection head, the ultrasonic detection head is controlled to face the back normal of the thin-walled part through the eddy current normal holding process; The above eddy current normal holding process involves: Multiple eddy current sensors capture the distance from the back of the thin-walled part, A machine tool coordinate system is constructed based on the eddy current distribution generated by the eddy current sensor, and the back surface distance is transferred to the coordinate system; The center of the above eddy current distribution coincides with the origin of the coordinate system, Calculate the eddy current normal vector based on the back surface distance in the coordinate system; The process includes adjusting the orientation of the ultrasonic detection head based on the eddy current normal vector so that the eddy current normal vector coincides with the back surface normal vector of the thin-walled part.

[0021] In addition, the ultrasonic detection head is a water-immersion ultrasonic detection head, and a spray nozzle is installed outside the coupling part of the ultrasonic detection head, and the nozzle discharges a water flow that fills the area between the coupling part and the thin-walled part during the measurement process of the detection head; There is a water pressure sensor in front of the nozzle to measure the real-time water pressure value, and during the milling process, the real-time water pressure value is compared with the water pressure reference value to adjust the water flow pressure and support the thin-walled parts through water pressure.

[0022] In addition, in the detection process of the machining contour of the thin-walled part, the thin-walled part after machining is scanned to obtain the post-machining point cloud data, and then the post-machining scan curved surface is generated based on the post-machining point cloud data; Recognizing the characteristic area of ​​the design surface and obtaining the boundary of the characteristic area; The boundary is projected onto the processed scanned surface, and a geodesic line is verified between the first feature point of the boundary of the feature area and the second feature point of the processed scanned surface.

[0023] A processing system for carrying out the above mirror milling and measurement control method.

[0024] The processing system also includes a clamping process device, which clamps the thin-walled part, The clamping process device is equipped with a jack device, which moves simultaneously with the milling tool during the milling process to support the thin-walled parts; The top of the above jack device is equipped with an ultrasonic detection head for measuring the thickness of the thin-walled parts from the back side in the mirror milling process; The ultrasonic detection head is surrounded by eddy current sensors, and the processing system uses these sensors to keep the ultrasonic detection head at a constant normal and distance from the thin-walled part; a spray nozzle is provided on the outside of the coupling part of the ultrasonic detection head, and the nozzle discharges a water flow that fills the area between the coupling part and the thin-walled part during the measurement process of the ultrasonic detection head; There is a water pressure sensor in front of the nozzle to measure the real-time water pressure value, and in the mirror milling process, the real-time water pressure value is compared with the water pressure reference value to adjust the water flow pressure.

[0025] The above technical solutions have the following advantages or beneficial effects:

[0026] In existing body processing methods, the deformation of the body itself during the actual processing process affects processing accuracy. In this method, the thin-walled part is clamped and then laser scanned to obtain point cloud data. Positioning holes are then added to the preliminary processing and design process of the thin-walled part. Based on this, the tool position points are projected onto the design surface, and geodesic information is obtained between the tool position points and the theoretical positioning holes on the design surface. Then, based on the geodesic information, the corresponding tool position points can be accurately mapped onto the actual processing surface in the point cloud data to form a transplanted processing path. This reduces the impact of body deformation on the tool path and improves processing accuracy. [Brief explanation of the drawings]

[0027] The embodiments of the present invention will be more fully described with reference to the accompanying drawings, which are used for illustrative purposes only and do not constitute limitations on the scope of the invention.

[0028] [Figure 1] 1 is an overall schematic diagram of an embodiment of the present invention; [Figure 2] FIG. 1 is a diagram illustrating a process for transferring a thin-walled part machining path program according to an embodiment of the present invention. [Figure 3] 1 is a schematic diagram of a mirror milling system according to an embodiment of the present invention; [Figure 4] 1 is a diagram of a clamping process apparatus according to an embodiment of the present invention; [Figure 5] FIG. 2 is a diagram of a point cloud acquisition process according to an embodiment of the present invention. [Figure 6] FIG. 10 is a diagram of a tool position point extraction process according to an embodiment of the present invention. [Figure 7] FIG. 2 is a diagram of a geodesic calculation process according to an embodiment of the present invention. [Figure 8] FIG. 1 is a diagram of a path migration process according to an embodiment of the present invention. [Figure 9] FIG. 10 is a diagram of an eddy current normal preserving process according to an embodiment of the present invention. [Figure 10] 1 is a diagram of a spray nozzle according to an embodiment of the present invention. [Figure 11] 1 is a diagram of a water supply system according to an embodiment of the present invention; [Figure 12] FIG. 10 is a diagram of a thin-walled part machining contour detection process according to an embodiment of the present invention. [Figure 13] 1 is a diagram of a processing system according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0029] In the following, the technical solutions of the embodiments of the present invention will be clearly and completely explained in combination with the figures of the embodiments of the present invention. Obviously, the embodiments described here are only a part of the present invention, and are not all of it. Based on the embodiments of the present invention, ordinary skilled artisans in this field should understand that all other embodiments obtained without creative efforts also fall within the scope of protection of the present invention.

[0030] It should be noted that, unless there is a conflict, the embodiments and features of the embodiments of the present invention can be combined with each other.

[0031] The present invention will be described in more detail below in combination with figures and specific examples, which are not to be construed as limitations of the present invention.

[0032] The present invention provides As shown in Figure 1, the mirror milling and measurement control method includes a thin-walled part clamping and conveying process, a thin-walled part measurement and point cloud acquisition process, a thin-walled part machining path program transfer process, a thin-walled part machining path post-processing process, a thin-walled part machining and thickness measurement control process, and a thin-walled part machining contour detection process, As shown in Figure 2, the transfer process of the machining path program for thin-walled parts is as follows: Based on the actual positioning holes in the point cloud data obtained from the scan of the thin-walled part and the theoretical positioning holes on the theoretical triangular mesh surface generated based on the design surface, the actual triangular mesh surface is matched with the theoretical triangular mesh surface; For a plurality of tool position points in the tool file, calculate geodesic information between each tool position point and the theoretical positioning hole; and using the geodesic information to graft the tool position points onto the actual triangular mesh surface to form a grafted machining path.

[0033] Specifically, in response to the problem of the existing mirror milling method affecting machining accuracy due to deformation of the thin-walled part itself during the actual machining process, this embodiment improves the tool transfer process for thin-walled parts. Here, multiple positioning holes are pre-formed on the thin-walled part through punching or a similar process, and corresponding positioning holes exist on the design surface during the thin-walled part design process. For the sake of distinction, the positioning holes on the design surface are referred to as theoretical positioning holes, and the actual positioning holes on the thin-walled part are referred to as actual positioning holes. These have a one-to-one correspondence. Typically, positioning holes are formed at the four corners around the main body of the thin-walled part, with a total number of four. However, the number and location may vary depending on the type of thin-walled part.

[0034] Before the tool path is transferred, a number of tool position points are obtained for the tool file obtained after the tool path is designed, and the tool position points correspond to the tool positions for milling the thin-walled part to match the design surface during the machining process. To achieve a more accurate tool position transfer process, the geodesic information for each tool position point is calculated in advance by referencing the positioning holes on the design surface. Then, based on the geodesic information, the tool position points can be mapped by referencing the actual positions of the positioning holes on the measured point cloud data, thereby achieving a more accurate transfer process.

[0035] Here, the point cloud data is obtained by scanning the thin-walled part after clamping it. Because the thin-walled part is flexible, it needs to be pre-clamped and jacked up to the desired processing state. Next, an appropriate scanning device, such as a line laser scanner, is used to scan the entire curved surface of the thin-walled part, and a point cloud of the reflected laser is obtained and used. This point cloud data contains the position information of each point on the thin-walled part in 3D space, and through further analysis, holes in the thin-walled part, such as the actual positioning holes, can be easily obtained.

[0036] Based on the above-mentioned transplantation process, the tool position points in the tool file are transplanted in order onto the actual curved surface of the thin-walled part, and then the tool information, such as the tool normal, sequence, number of rows, etc., is combined and processed to generate a corresponding transplanted machining path. Based on the transplanted machining path, in the subsequent milling process, the milling tool machining and the operation of the jack device and measuring device are controlled by the appropriate milling process to achieve a better machining effect.

[0037] In actual implementation, the above-mentioned machining method is mainly realized by a specific milling system, which includes a jacking device and a measuring device that can be synchronized with the corresponding multi-axis machining center and milling tool. Figure 3 shows a typical milling system, including a main shaft A001 for milling thin-walled parts, a scanning device A002 for acquiring the curved surface of the thin-walled part, a jacking device A003, a measuring device A004, and a computer A005. Here, the computer A005 is a milling bed control system or an equivalent device, and is configured with a specific computer program for realizing specific functions, especially the program related to the above-mentioned machining method.

[0038] In order to achieve a better measurement effect, the measuring device can be installed at the end of the jacking device and operate simultaneously with the jacking device. For example, if the measuring device is an ultrasonic detection head, it is located at the end of the jacking device and measures the real-time thickness of the thin-walled part from the back when the jacking device jacks up the thin-walled part. This can be used in conjunction with the milling tool on the front to accurately determine how much of the current thin-walled part has been removed and whether the thickness meets the actual processing requirements.

[0039] Before starting the actual measurement, a suitable pre-treatment process should be carried out, especially on thin-walled parts, which may include heat treatment of the material, cold rolling, stretching, rough cutting or similar machining processes, so that the thin-walled parts can be formed by milling.

[0040] In addition, before starting actual processing, corresponding design work should also be carried out for thin-walled parts. This involves using 3D design software to design the desired curved surface of the thin-walled part, such as the skin surface, to obtain the design surface. The design surface is actually represented as a computer program model file, which can be read and displayed by specific 3D industrial design software, and then corresponding processing drawings, etc. can be output as needed.

[0041] A tool file is a feasible machining solution designed based on the milling system's machining parameters after determining the pre-machining raw material parameters and design surface of a thin-walled part. It includes the tool location points that need to be machined to match the thin-walled part to the design surface, the trajectory composed of the tool location points, and tool information such as the tool normal, number of rows, and spindle rotation speed. The tool file is a file in a specific format, such as a CLS / APT file, that can be read and executed by a CNC machine to machine the workpiece. However, in the scenario targeted by this invention, a directly generated tool file usually has a certain offset in 3D space between the actual tool location points and the designed expected tool location points due to factors such as the clamping accuracy of the thin-walled part and material deformation. Therefore, this invention involves a process of remapping the tool location points to the actual machining surface. After this process is completed, a modified tool file is generated, and the tool file can be used to perform a more accurate milling process.

[0042] It should be noted that the tool position point transfer process should be performed after the thin-walled part is first clamped and scanned, and the thin-walled part should be milled immediately after the transfer is completed, which can avoid introducing new offsets during the repeated clamping of the thin-walled part.

[0043] In one embodiment, the clamping process uses a clamping process apparatus as shown in Figure 4 to clamp the thin-walled part. The clamping process apparatus has a square-shaped process apparatus frame 101, in which a plurality of moving columns 102 are distributed. The process equipment frame 101 and the moving support 102 are fitted with clamps 103 for clamping and fixing thin-walled components. Distributed on the process equipment frame 101 are a plurality of jack devices 104 that move from the rear to the front of the thin-walled components to support the thin-walled components.

[0044] Specifically, to achieve a better clamping effect, in this embodiment, the above-mentioned clamping process device is used to clamp the thin-walled part before scanning the thin-walled part. Here, the clamping process device has a square-shaped process device frame 101, which is a hollow frame structure with a front and a back. When it is necessary to clamp the thin-walled part, the process device frame 101 is placed with the front side up, and the moving support 102 and the jack device 104 are positioned near the back side of the frame 101. The process device frame 101 has multiple clamps 103 distributed around the circumference, which clamp and fix the thin-walled part around the circumference.

[0045] On the other hand, since the thin-walled part itself has an irregular structure and may have an edge that is raised in three-dimensional space, an additional clamp 103 is set on the process equipment frame 101 via a movable support 102. The process equipment frame 101 has a plurality of guide rails arranged on its rear side, and the movable support 102 is set movably on the process equipment frame 101 via the guide rails, and its position can be adjusted as needed, allowing the clamp 103 to effectively clamp the edge of the curved part.

[0046] In addition, in order to maintain the shape of the thin-walled part, a plurality of jack devices 104 are arranged within the process equipment frame 101, and the jack devices 104 support the back of the thin-walled part according to a set jack program, so that the thin-walled part maintains a specific curved state, allowing subsequent operations such as scanning, measuring, milling, etc.

[0047] In addition, in order to maintain the shape of the thin-walled part, the frame 101 is equipped with a plurality of jacking devices 104 which are distributed inside, and which jack up the back of the thin-walled part based on a set jacking program, maintain a specific bending state, and facilitate subsequent scanning, measuring, milling, and other operations.

[0048] To achieve a good jacking effect, a plurality of jacking devices 104 can be arranged in a rectangular shape inside the rectangular process equipment frame 101, and the height of each jack can be controlled individually to support and maintain the shape of the thin-walled parts.

[0049] Similarly, the number of the above clamps 103 and moving columns 102 can be correspondingly disassembled, installed, adjusted in position and angle based on the shape, size and other parameters of the thin-walled parts to be actually processed.

[0050] In one embodiment, as shown in FIG. 5, the clamping and conveying process for thin walled parts includes: The thin-walled part is hoisted onto the process equipment frame, and laser projection is performed onto the process equipment frame. The thin-walled part is positioned within the range of the laser projection; Adjusting the jacking device and clamps to adjust the thin-walled part to the intended processing state based on the laser projection and the process device program; The steps include clamping the thin-walled parts in order and then scanning them to obtain point cloud data.

[0051] Specifically, by using the clamping process device as the center, the object can be clamped and scanned to obtain more accurate point cloud data.

[0052] Specifically, before lifting the thin-walled part for work, a corresponding jack program is weaved in advance according to the design curved surface, and the jack parameters required to bring the thin-walled part into the planned processing state are adopted, and then the number of jack devices actually equipped on the process equipment frame is combined to perform matrix processing to generate matrix point pixels, and then the corresponding process equipment program for controlling the jack devices is generated.

[0053] In addition, laser projection assistance is also introduced into the clamping process of thin-walled parts to achieve a better clamping instruction process. This laser projection assistance is realized by a laser projector, which projects a laser pattern onto the object surface to indicate each position in the clamping process, including where the edge of the thin-walled part should be located and where the clamp will clamp the thin-walled part.

[0054] To achieve this process, a laser projection programming process must be performed in advance. This involves planning the front of the process equipment frame and determining the positioning markers that the laser projection depends on for the projection process, such as the positioning markers on the process equipment frame and the effective positioning hole locations of thin-walled parts. After determining the relative references, further planning is performed to obtain the corresponding clamp lines. Then, the laser projection programming is performed using Projector, and the corresponding laser projection programming file is converted for subsequent calling.

[0055] After the initial preparation process is complete, the thin-walled part can be lifted by crane. Specifically, the processing equipment frame is placed horizontally, and then the thin-walled part is lifted onto the processing equipment frame by crane and dropped. In this process, the processing equipment program is used to adjust the height of the jacking device in advance. After the thin-walled part is dropped, the laser projector is controlled based on the laser projection programming file to project the image. Specifically, the center mark of the laser projection is first aligned with the target point on the surface of the thin-walled part or the surface of the processing equipment frame. Then, the effective positioning hole of the thin-walled part is projected. Based on the projection position, the orientation of the thin-walled part and the height of the jacking device are fine-tuned until the actual positioning hole matches the projected positioning hole mark. This indicates that the thin-walled part has reached the intended processing state and can be clamped. During the clamping process, the clamps should be tightened in order according to the corresponding engineering standards to avoid placing excessive stress on the thin-walled part.

[0056] Finally, after the thin-walled part is clamped, the process moves to the thin-walled part measurement and point cloud acquisition process, where a scan is performed to obtain point cloud data. In one embodiment, a line laser is used as the scanning tool. Specifically, a line laser scanning program is developed in advance, centered on the design surface and capable of completely scanning the design surface under the same coordinate system as the design surface. The line laser is then driven through the milling bed to scan the entire front surface of the thin-walled part, and the reflected signals are collected as scan data. Based on this scan data, a kinematic algorithm for the milling bed is combined to reconstruct the point cloud data in the space of the curved surface corresponding to the surface of the thin-walled part, which can then be easily processed and output as point cloud data.

[0057] In one embodiment, in the thin-walled part machining and thickness measurement and positioning process, a jacking device is used to support the sunk area of ​​the thin-walled part.

[0058] Specifically, after selecting the above clamping process device to perform clamping, in order to achieve better milling accuracy, after scanning the thin-walled part, the scanning process device frame is directly turned over and moved to the processing position for the subsequent milling process. At this time, a jack device is connected to the mirror milling system, controlled during the processing process, and moved simultaneously with the front milling cutter to support the sunken area of ​​the thin-walled part during the processing process, thereby achieving better processing results.

[0059] In one embodiment, as shown in FIG. 6, the process of extracting the calculated tool position points of the tool file includes: Extract the center coordinates of the theoretical positioning holes on the theoretical triangular mesh surface, and extract the information of each tool position point from the tool position file. The sword position point information includes the coordinates of the sword position point, Projecting each of the tool position points onto a theoretical grid surface of the theoretical triangular mesh surface to generate a first projection point and a projection distance between the first projection point and the tool position point; The process includes calculating the area coordinates of each of the first projection points relative to the projection point grid where the first projection point is located, and calculating the geodesic length relative to the area coordinates for each of the center coordinates of the theoretical positioning holes to obtain geodesic information.

[0060] Specifically, after obtaining the design surface and tool file, the above process can obtain the geodesic information for each tool position point relative to the theoretical positioning hole coordinates, which is used as a reference for subsequent tool position point transplantation.

[0061] Here, the design surface is first discretized at equal intervals to form a design surface point cloud, which is then reconstructed into a point cloud data similar to the point cloud data. The interval setting can be selected based on the scanning interval of the line laser. The discretization process involves obtaining the parameter range of the design surface, parameterizing the surface, and then modifying the UV values ​​on the UV surface according to the interval to obtain a number of points and adding them to the design surface point cloud. The theoretical positioning hole locations planned within the design surface can then be combined to easily extract the center coordinates of the theoretical positioning holes on the design surface. The design surface point cloud can then be combined to extract the theoretical positioning hole center coordinates after point cloud construction. This process adjusts the theoretical positioning hole center coordinates to match the point cloud data, facilitating subsequent alignment with the actual positioning holes in the point cloud data.

[0062] The tool file is also read in a loop to obtain many tool position points, and the tool position point information mainly includes the tool position point coordinates, normals, and the line number where the tool position point is located. The transplantation process mainly focuses on the deformation of thin-walled parts that affects the position change of the projected tool position point in space, that is, the movement of the tool position point coordinates.

[0063] To achieve better processing efficiency, in this embodiment, the design surface point cloud of the design surface and the point cloud data in the point cloud data obtained by scanning are meshed separately. In one embodiment, a Delaunay triangular meshing algorithm is used to mesh the design surface and the point cloud data, respectively. The meshed surface of the design surface after meshing is called a theoretical grid surface, and the meshed surface of the point cloud data after meshing is called an actual grid surface.

[0064] Next, the tool position point is projected onto the theoretical grid surface along the surface normal, which always falls within a triangular grid on the theoretical grid surface. The projection length between the projection point and the tool position point, as well as the triangular grid where the projection point is located, are recorded as the projection point grid. Within the projection point grid, connections are made from the projection point toward the vertices of the projection grid to obtain many grid subunits within the projection point grid. The triangular grid has three grid subunits, and the area coordinates of each grid subunit are calculated. The geodesic length of each theoretical positioning hole for the projection point can be obtained by interpolating the geodesic length from the positioning hole coordinate to the three vertices and the area coordinates. Finally, the geodesic lengths for each theoretical positioning hole for the projection point are summed to obtain the complete geodesic information, which is used for subsequent transplantation of the tool position point.

[0065] Specifically, referring to Figure 7, three grid subunits are set in a triangular grid with three vertices A, B, and C, and their central positions are the projection point p. The area coordinates of the grid subunits are S1, S2, and S3, respectively, where S1+S2+S3=1. Then, the geodesic length d between the external theoretical positioning hole pos and the projection point is d=d1*S1+d2*S2+d3*S3.

[0066] In one embodiment, as shown in FIG. 8, the path implantation process of the implantation machining path includes: Extract the center coordinates of the actual positioning hole from the point cloud data, and project the tool position point onto the actual grid plane corresponding to the point cloud data to obtain a second projected point; For the second projection point, calculate a predicted projection geodesic length relative to the center coordinate of the actual positioning hole, and then calculate a geodesic deviation value based on the geodesic information and the predicted projection geodesic length; According to the geodesic deviation value and the preset geodesic deviation range, the second projection point is iteratively processed until it satisfies the geodesic deviation range, and then used as the actual transplantation point; A process is included to generate a transplant processing path based on the actual transplant point.

[0067] Specifically, after the geodesic line information is determined, the tool position points can be moved to the point cloud data based on the geodesic line information.

[0068] Here, the point cloud data has been scanned with a line laser to extract point cloud data related to the actual positioning hole, and then a boundary recognition algorithm is used to recognize the point cloud, obtain the edge portion in the point cloud data, recognize the hole portion in the edge based on curvature recognition, and perform fitting based on the point cloud data of that portion to obtain the center coordinates of the actual positioning hole.

[0069] Similarly, the actual grid surface is obtained in advance through the meshing process described above. The tool position points are projected onto the actual grid surface in the normal direction of the surface, and they must fall within the triangular grid of the actual grid surface.

[0070] At this point, the projection length between the predicted projection point and the tool position point is recorded, and the triangular grid where the predicted projection point exists is recorded as the projection grid. Within the projection grid, connections are made from the predicted projection point to each vertex to form multiple subunits within the grid. The triangular projection grid has three grid subunits, and the area coordinates of each grid subunit are calculated. This allows the geodesic length to be interpolated using the geodesic length from the tool position point to the three vertices and the area coordinates. The predicted projection geodesic length is then calculated by combining the position information of the predicted projection point grid with the relative positional relationship with the actual positioning hole.

[0071] Based on this predicted projected geodesic length, the previously collected geodesic information is compared to determine whether the current predicted projected point matches the expected projected position for the actual positioning hole. That is, it is determined whether the difference between the predicted projected geodesic length and the geodesic information is within an acceptable range. If not, an optimization algorithm is used to adjust the position of the predicted projected point until it falls within the geodesic deviation range, and then the predicted projected point is used as the actual projected point.

[0072] Next, an interpolation fitting method is used to calculate the normal direction of the actual projected point based on the normal directions of the surrounding grid points associated with the actual projected point on the actual grid plane. The actual projected point is offset along its normal direction by the projection length to obtain its actual projected position. Processing is then performed using the actual projected point and the adjusted normal to obtain the projected machining trajectory used in the mirror milling process.

[0073] In one embodiment, the thin-walled part machining and thickness measurement control process uses an ultrasonic detection head to measure the thickness of said thin-walled part in real time, and the real-time thickness is used in the mirror milling process. Specifically, to achieve a more accurate milling process, in this embodiment, an ultrasonic detection head is used to measure the thickness of the thin-walled part in real time during milling to obtain the real-time thickness of the thin-walled part. This is used to provide feedback control for the milling system. Specifically, to achieve a better processing effect, the milling system installs a jacking device on the back of the thin-walled part during the processing process to support and maintain the shape of the thin-walled part. During the milling cutter movement process, the jacking device moves synchronously with the milling cutter. In this system, an ultrasonic detection head can be optionally installed at the tip of the jacking device, which can achieve a more accurate thickness measurement process for the milling cutter's cutting portion. During the measurement process, a coupling agent can be adjusted based on the measurement requirements. In one embodiment, the coupling agent is a water jet injected between the ultrasonic detection head and the thin-walled part through a spray nozzle.

[0074] In one embodiment, the ultrasonic detection head has a plurality of eddy current sensors distributed around the periphery, which generate eddy currents during the measurement process of the ultrasonic detection head, and the ultrasonic detection head measures the eddy current distances for the thin-walled part; In the mirror milling process, an ultrasonic detection head controls the distance to the thin-walled part based on eddy current spacing.

[0075] In this embodiment, the ultrasonic detection head is modified to include multiple eddy current sensors in the circumferential direction, which can generate eddy currents corresponding to the direction of the ultrasonic detection head's direction and receive echo signals. Based on the echo signal intensity, the ultrasonic detection head can easily calculate the eddy current distance for thin-walled parts by combining the pre-defined reflection intensity-distance correspondence function, which can be used as a machine control parameter for the mirror milling system to realize feedback control of the milling cutter cutting process.

[0076] As an optional implementation, there are four eddy current sensors, evenly distributed at the 12 o'clock, 3 o'clock, 6 o'clock, and 9 o'clock positions on the watch face of the ultrasonic detection head.

[0077] In order to realize the above-mentioned interval control process, it is necessary to locate the relationship between the intensity of the eddy current and the interval before processing. A metal locating block is introduced in advance, and the normal of the eddy current sensor is aligned with the surface of the metal locating block, maintaining a certain distance. This distance is the zero point distance. At this time, the eddy current value is read and stored in association with the zero point position. Based on this, the eddy current sensor is moved in the normal direction in specific steps to increase the distance, and the eddy current value is recorded and stored. A linear fit can be performed using each distance and eddy current value collected above to obtain a reflection intensity-distance correspondence function.

[0078] In one embodiment, the measurement process of the ultrasonic detection head includes controlling the ultrasonic detection head to face the back surface normal of the thin-walled part through an eddy current normal holding process; As shown in Figure 9, the eddy current normal holding process involves: Multiple eddy current sensors capture the distance from the back of the thin-walled part, A machine tool coordinate system is constructed based on the eddy current distribution generated by the eddy current sensor, and the back surface distance is transferred to the coordinate system; The center of the eddy current distribution coincides with the origin of the coordinate system, Calculate the eddy current normal vector based on the back surface distance in the coordinate system; The process includes adjusting the orientation of the ultrasonic detection head based on the eddy current normal vector so that the eddy current normal vector coincides with the back surface normal vector of the thin-walled component.

[0079] Specifically, in order to achieve a better measurement effect, in this embodiment, the ultrasonic detection head can be maintained in alignment with the normal direction of the back surface of the thin-walled part through the feedback signal from the eddy current sensor. In this embodiment, the ultrasonic detection head and the eddy current sensor are mounted on a spherical joint, so that their orientation can be freely changed.

[0080] Here, after measuring the real-time thickness through eddy currents, multiple eddy current sensors can obtain the relative distance to the back surface of the thin-walled part according to the process. Three of the eddy current sensors are selected and a machine coordinate system is constructed based on the distribution surface of the eddy currents they generate. This machine coordinate system takes the distribution center of the three eddy currents as its coordinate origin, and maps the measurement distances of the three eddy current sensors onto the machine coordinate system.

[0081] Next, the eddy current normal vector between the three eddy currents is calculated, and its value is n = n1 × n2.

[0082] In this formula, n1 represents the vector of the first and second eddy currents and is the direction from the second eddy current to the first eddy current. n2 represents the vector of the first and third eddy currents and is the direction from the third eddy current to the first eddy current. n represents the eddy current normal vector of the first eddy current and is the direction of the cross product of vectors n1 and n2, that is, it is calculated according to the right-hand rule.

[0083] Based on the above process, the eddy current normal vector can be obtained. The back surface of the thin-walled part has a theoretical normal vector (0,0,1) predetermined in the machine coordinate system. For these two vectors, the vector angle between them can be easily calculated, which corresponds to the deflection angle of the ultrasonic detection head. Then, the rotation of the spherical joint is controlled based on the deflection angle to match these two vectors, completing the normal control process of the ultrasonic detection head.

[0084] In one embodiment, the ultrasonic detection head is a water immersion ultrasonic detection head, and a spray nozzle is installed outside the coupling part of the ultrasonic detection head, and the nozzle discharges a water flow that fills the area between the coupling part and the thin-walled part during the measurement process of the detection head; There is a water pressure sensor in front of the nozzle to measure the real-time water pressure value, and during the milling process, the real-time water pressure value is compared with the water pressure reference value to adjust the water flow pressure and support the thin-walled parts through water pressure.

[0085] Specifically, to achieve better measurement results, a water-immersion ultrasonic detection head is selected in this embodiment. A spray nozzle is installed around the ultrasonic detection head to discharge a water flow that fills the area between the coupling part and the thin-walled part during the ultrasonic detection head measurement process. As shown in FIG. 10, the spray nozzle 201 has a generally U-shaped structure, and a water-immersion ultrasonic detection head 202 is installed at the center bottom. A water inlet is installed around the immersion ultrasonic detection head 202, and the coupling agent flows out from the water inlet to fill the U-shaped structure. An eddy current sensor 203 for distance measurement is installed around the outer periphery of the spray nozzle 201, and the spray nozzle 201 is installed at a regular interval relative to the thin-walled part 204.

[0086] Additionally, the water pressure generated by the spray nozzle 201 is also used to support thin-walled components.

[0087] As shown in FIG. 11, the upstream stage of the spray nozzle includes a series of water supply systems, including a tank 301 as a water source, a pump 302 that supplies a water flow, a defoamer 303 that removes bubbles from the water flow from the pump 302, and a fluid valve 304 that supplies a water flow to the spray nozzle 201.

[0088] During the thickness measurement process, pump 302 operates to draw coupling agent from tank 301, and the output of pump 302 is passed to defoamer 303 to remove bubbles generated in the coupling agent due to the pressure disturbance of pump 302. After that, fluid valve 201 opens and the coupling agent is sprayed out of the spray nozzle. Due to the semi-closed structure of spray nozzle 302 itself, after the water flow contacts the back surface of the thin-walled part, a certain amount of back pressure is formed toward spray nozzle 201, which is directly proportional to the water pressure for thickness measurement.

[0089] During the measurement process, the water pressure for thickness measurement must be kept constant to maintain consistency in the measurement process. To this end, in this embodiment, a water pressure sensor 305 is installed between the fluid valve 304 and the degassing device 303 to measure the back pressure of the discharged water and output the real-time water pressure value. The standard water pressure value for the spray nozzle 201 is obtained in advance through experiments, and the real-time water pressure value is matched with the standard water pressure value to achieve good thickness measurement results.

[0090] In the process of controlling the real-time water pressure value, the difference between the real-time water pressure value and the standard water pressure value is processed based on a PID algorithm, and this process can be achieved by controlling the rotation speed of the pump 302. This can also be achieved by a bypass water discharge method. Specifically, a three-way connector is installed upstream of the fluid valve 304, which is connected to the water inlet of the tank 301 via the electric valve 306, forming a reflow path from the pump 302 to the electric valve 306 and the tank 301. The opening degree of the electric valve 306 varies depending on the difference between the real-time water pressure value and the standard water pressure value. For example, if the water pressure difference is positive, it indicates that the real-time water pressure value exceeds the standard water pressure value, and the electric valve 306 opens more to decrease the real-time water pressure value; if the water pressure difference is negative, it indicates that the real-time water pressure value is lower than the standard water pressure value, and the electric valve 306 opens less to increase the real-time water pressure value.

[0091] In one embodiment, as shown in FIG. 12 , in the process of detecting the machining contour of the thin-walled part, the after-machining thin-walled part is scanned to obtain after-machining point cloud data, and an after-machining scanned surface is generated based on the after-machining point cloud data; Recognizing a characteristic area of ​​the design surface and obtaining the boundary of the characteristic area; The boundary is projected onto the processed scanned surface, and a geodesic line is verified between the first feature point of the boundary of the feature area and the second feature point of the processed scanned surface.

[0092] Specifically, to determine whether the processed thin-walled part matches the design surface, the processed contour of the thin-walled part is inspected after the thin-walled part's processing and thickness measurement control processes are completed. Specifically, the design surface is marked with feature areas and feature points within the feature areas, which can be used for pre-comparison. After processing is completed, the same line laser is used to scan the processed thin-walled part to obtain post-processing point cloud data, from which the scanned surface is reconstructed. Accordingly, a feature boundary recognition algorithm is applied to the design surface to extract the boundaries of the pre-marked feature areas. Because the design surface and the thin-walled part are pre-aligned, the boundaries of the feature areas can then be projected onto the processed scanned surface. Geodesic information is then calculated for the feature points corresponding to the feature areas on the processed scanned surface, and this geodesic information is compared with the geodesic information of the feature points pre-marked on the design surface. This allows us to determine whether the processed contour is as expected.

[0093] A processing system for carrying out the above processing method.

[0094] As shown in FIG. 13, the processing system includes a clamping process device B020 that is used to clamp thin-walled parts.

[0095] The clamping process device B020 incorporates a jack device B021, which moves simultaneously with the milling knife in the milling process and is used to support thin-walled parts.

[0096] To implement the above-described machining method, a corresponding machining system is also provided in this embodiment. The machining system must include at least a mirror milling system B010, a clamping process device B020, a curved surface scanning device B030, a machining process measuring device B040, and a control device B050. The clamping process device B020 includes a jack device B021, which moves simultaneously with the milling cutter during the mirror milling process and is used to support thin-walled parts. The machining process measuring device B040 provides real-time data to the control device B050 during the mirror milling process to achieve feedback control. The control device B050 controls each of the above modules based on a pre-programmed computer program.

[0097] The above are merely preferred embodiments of the present invention, and do not limit the implementation method and protection scope of the present invention. Any solutions obtained by equivalent replacement and obvious changes made by a person skilled in the art using the contents of the description and drawings of the present invention should all be included in the protection scope of the present invention.

Claims

1. A mirror milling and measurement control method, comprising: a thin-walled part clamping and conveying process; a thin-walled part measuring and point cloud acquiring process; a thin-walled part machining path program transfer process; a thin-walled part machining path post-processing process; a thin-walled part machining and thickness measurement control process; and a thin-walled part machining contour detection process; In the above process of transferring the machining path program of the thin-walled part, Based on the actual positioning holes in the point cloud data obtained from the scan of the thin-walled part and the theoretical positioning holes on the theoretical triangular mesh surface generated based on the design surface, the actual triangular mesh surface is matched with the theoretical triangular mesh surface; For a plurality of tool position points in the tool file, calculate geodesic information between each tool position point and the theoretical positioning hole; and using the geodesic information to transfer the tool position points onto the actual triangular mesh surface to form a transfer machining path. Construction method.

2. 2. The mirror milling and measurement control method according to claim 1, wherein the clamping and transporting process of the thin-walled part comprises: securing the thin-walled component using a clamping process device, the clamping process device comprising a square-shaped process device frame having a plurality of moving posts distributed therein; Clamps are attached to the process equipment frame and the moving support column to clamp and secure the thin-walled components; A plurality of jack devices are distributed within the process equipment frame, which move from the rear to the front of the thin wall components to support the thin wall components. Construction method.

3. In the mirror milling and measurement control method as claimed in claim 2, the clamping and transporting process of the thin-walled part includes: suspending the thin-walled component on the process equipment frame and projecting a laser beam onto the process equipment frame; the thin-walled part is positioned within the range of the laser projection; Adjusting the jack device and the clamp based on the laser projection and the process device program to adjust the thin-walled part to a predetermined processing state; The steps include clamping the thin-walled parts in order and then scanning them to obtain point cloud data. Construction method.

4. In the mirror milling and measurement control method as claimed in claim 1, the thin-walled part measurement and point cloud acquisition process includes driving a line laser by a machine tool to scan the thin-walled part and construct point cloud data. Construction method.

5. In the mirror milling and measurement control method as claimed in claim 1, the process of extracting tool position points based on the calculation of the tool file includes: Extracting the center coordinates of the theoretical positioning holes on the theoretical triangular mesh surface, and extracting information on each of the tool position points from the tool position file; The above-mentioned sword position point information includes coordinates of the sword position point, Projecting each of the tool position points onto a theoretical grid surface of the theoretical triangular mesh surface to generate a first projection point and a projection distance between the first projection point and the tool position point; and a process of calculating area coordinates of the first projection points relative to a projection point grid on which the first projection points are located, and calculating geodesic lengths relative to the area coordinates for each of the center coordinates of the theoretical positioning holes, to obtain the geodesic line information. Construction method.

6. In the mirror milling and measurement control method according to claim 1, the process of forming the implanted machining path includes: Extract the center coordinates of the actual positioning hole from the point cloud data, and project the tool position point onto an actual grid plane corresponding to the point cloud data to obtain a second projected point; For the second projection point, calculate a predicted projection geodesic length relative to the center coordinate of the actual positioning hole, and then calculate a geodesic deviation value based on the geodesic information and the predicted projection geodesic length; According to the geodesic deviation value and the preset geodesic deviation range, the second projection point is iteratively processed until it satisfies the geodesic deviation range, and then it is used as the actual transplantation point; This includes a process to generate a transplant processing path based on the actual transplant point. Construction method.

7. In the mirror milling and measurement control method as claimed in claim 1, the post-processing process of the machining path of the thin-walled part includes the process of creating a new implantation tool position file according to the implantation machining path, and performing simulation testing; In the thin-walled part processing and thickness measurement control process, the thin-walled part is processed using the transferred tool position file. Construction method.

8. In the mirror milling and measurement control method as claimed in claim 2, the machining and thickness measurement control process of the thin-walled parts uses a jack device to support the subsidence area of ​​the thin-walled parts. Construction method.

9. In the mirror milling and measurement control method as claimed in claim 1, the processing and thickness measurement control process of the thin-walled part uses an ultrasonic detection head to measure the thickness of the thin-walled part in real time, and the real-time thickness is used in the mirror milling process. Construction method.

10. In the mirror milling and measurement control method as claimed in claim 9, the ultrasonic detection head has a plurality of eddy current sensors distributed around it, which generate eddy currents during the measurement process of the ultrasonic detection head, and the ultrasonic detection head measures the eddy current intervals for the thin-walled parts; In the mirror milling process, the ultrasonic detection head controls the distance to the thin-walled part based on the eddy current distance. Construction method.

11. In the mirror milling and measurement control method as claimed in claim 10, in the measurement process of the ultrasonic detection head, the ultrasonic detection head is controlled to face the back normal of the thin-walled part through an eddy current normal holding process; The eddy current normal holding process includes: Multiple eddy current sensors capture the distance from the back of the thin-walled part, A machine tool coordinate system is constructed based on the eddy current distribution generated by the eddy current sensor, and the back surface distance is transferred to the coordinate system; the center of the eddy current distribution coincides with the origin of the coordinate system, Calculate the eddy current normal vector based on the back surface distance in the coordinate system; and adjusting the orientation of the ultrasonic detection head based on the eddy current normal vector so that the eddy current normal vector coincides with the back surface normal vector of the thin-walled part. Construction method.

12. 10. The mirror milling and measurement control method according to claim 9, wherein the ultrasonic detection head is a water-immersion ultrasonic detection head, and a spray nozzle is installed outside the coupling part of the ultrasonic detection head, which discharges a water flow that fills the area between the coupling part and the thin-walled part during the measurement process of the detection head; There is a water pressure sensor in front of the nozzle, which measures the real-time water pressure value. During the milling process, the real-time water pressure value is compared with the water pressure reference value to adjust the water flow pressure and support the thin-walled parts through water pressure. Construction method.

13. In the mirror milling and measurement control method as described in claim 1, in the process of detecting the machining contour of the thin-walled part, the thin-walled part after machining is scanned to obtain post-machining point cloud data, and a post-machining scanned curved surface is generated based on the post-machining point cloud data; Recognizing a characteristic area of ​​the design surface and obtaining a boundary of the characteristic area; The boundary is projected onto the processed scanned surface, and a geodesic line is verified between the first feature point of the boundary of the feature area and the second feature point of the processed scanned surface. Construction method.

14. A machining system for carrying out the mirror milling and measurement control method according to claim 1.

15. 15. The processing system of claim 14, wherein the processing system includes a clamping process device, the clamping process device clamping the thin-walled component; The clamping process device is provided with a jack device, which moves simultaneously with the milling tool to support the thin-walled parts during the milling process. system.

16. 16. The processing system according to claim 15, wherein an ultrasonic detection head is installed on the top of the jack device to measure the thickness from the back surface of a thin-walled part in a mirror milling process. system.

17. 15. The machining system of claim 14, wherein said machining system includes an ultrasonic detection head for measuring the thickness of thin-walled components in a mirror milling process. system.

18. 18. The processing system according to claim 17, wherein eddy current sensors are installed around the ultrasonic detection head, and the processing system uses these sensors to keep the ultrasonic detection head at a constant normal and distance from the thin-walled part. system.

19. 18. The processing system according to claim 17, wherein a spray nozzle is installed outside the coupling part of the ultrasonic detection head, and the nozzle discharges a water flow to fill the area between the coupling part and the thin-walled part during the measurement process of the ultrasonic detection head; There is a water pressure sensor in front of the nozzle to measure the real-time water pressure value, and in the mirror milling process, the real-time water pressure value is compared with the water pressure reference value to adjust the water flow pressure. system.

Images