Adaptive machining method for ceramic matrix composite parts

Abstract

The application provides a self-adaptive machining method for a ceramic matrix composite part, and belongs to the technical field of composite material precision machining. The method comprises the following steps: clamping a blank and establishing a machining coordinate system; selecting a theoretical measurement point set and measuring an actual point set; solving initial rigid body transformation parameters; establishing a registration model containing a margin constraint and a direction constraint, and solving a registered numerical model space coordinate system; correcting a measurement program based on the coordinate system and re-measuring; verifying the margin constraint condition, and if the condition is not satisfied, iterating until the condition is satisfied, and then performing single-face machining; after turning over, selecting a registration with a tolerance constraint or a mode based on high point margin calculation to perform machining according to whether the machined surface can be measured. Through the registration model with the margin constraint and the closed-loop verification iteration mechanism, the application effectively avoids overcutting, realizes intelligent compensation for the deformation of the blank, and improves the machining precision and the yield.

Description

Technical Field

[0001] This invention relates to the field of precision machining technology for composite materials, and more specifically, to an adaptive machining method for ceramic matrix composite parts. Background Technology

[0002] Due to the high cost of materials and limitations in forming processes, ceramic matrix composite (CMC) parts typically have a minimum allowance of about 0.3mm per side for the blank. Furthermore, these parts often have complex curved surfaces without a standard, regular shape. Currently, the process relies heavily on tooling fit and manual clamping experience. Because CMC parts contain numerous coarse fiber bundles, the degree of densification is difficult to control, and the parts are prone to deformation during forming. Therefore, traditional manual methods struggle to ensure reasonable allowance distribution, easily leading to inaccurate machining positions and out-of-tolerance contours.

[0003] Traditional CNC machining methods rely on a fixed theoretical numerical model coordinate system for positioning and programming. However, due to the significant deviation between the actual shape of the CMC blank and the theoretical numerical model, directly using the theoretical coordinate system for machining can easily lead to insufficient allowance in local areas, damaging the part (overcutting), or excessive allowance preventing proper machining. Existing registration algorithms are mainly used for registration and alignment of near-net-shape / partially formed parts made of metal materials. These algorithms measure and register areas that are smooth, dense, and have uniform blank allowance distribution / or are the final surface shape, resulting in less error interference during registration. However, they cannot meet the application requirements of ceramic matrix composite parts. Summary of the Invention

[0004] (a) Technical problems to be solved The technical problem to be solved by this invention is to provide a positioning and processing method that can effectively avoid overcutting, intelligently compensate for blank deformation, and achieve high-precision adaptive machining, given the characteristics of small blank allowance, uneven distribution, and irregular reference surface of ceramic matrix composite parts.

[0005] (II) Technical Solution To achieve the above objectives, the technical solution adopted by the present invention is as follows: This invention provides an adaptive processing method for ceramic matrix composite parts, comprising the following steps: Step S1: Clamp and fix the ceramic matrix composite blank and establish a machining coordinate system. By clamping and fixing the ceramic matrix composite blank, the relative movement between the workpiece and the machine tool is eliminated, ensuring the stability of the machining process. Establishing a machining coordinate system provides a unified reference origin for subsequent measurement programs, ensuring that all measurement data are collected and processed in the same coordinate system. This avoids the accumulation of errors caused by inconsistencies in coordinate systems and ensures the repeatability and accuracy of the measurement data.

[0006] Step S2: Select the set of theoretical measurement points in the theoretical coordinate system. {P i ′ } The actual shape and position of the ceramic matrix composite blank clamped under the tooling are measured sequentially in the machining coordinate system to obtain the actual measurement point set corresponding to the theoretical measurement point set. {P i } ,in i The key function of this step is to number the points; it is to obtain the actual shape data of the blank. The theoretical measurement points selected in the theoretical coordinate system represent the ideal positions of the part design, while the set of points actually measured in the machining coordinate system reflects the true deformation state of the blank. By comparing these two sets of points, the degree and distribution pattern of the blank's deformation can be quantitatively evaluated, providing the necessary input data for subsequent adaptive registration.

[0007] Step S3: Set up the theoretical measurement points {P i ′ } Set of actual measurement points {P i } Perform coordinate transformations and solve the rotation matrix based on rigid body transformation relationships. R 0 and translation matrix T 0; This step achieves a fast coarse registration. Without any prior information, directly solving a constrained optimization model can easily lead to local optima or slow convergence. By first performing an unconstrained rigid body transformation, the macroscopic deviation between the theoretical and actual point sets can be quickly reduced, yielding an initial solution close to the global optimum. This initial solution provides a good starting point for the next step of constrained fine registration, significantly improving the convergence speed and stability of the optimization algorithm and avoiding registration failures caused by unreasonable initial values.

[0008] Step S4: Based on the theoretical measurement point set {P i ′ } Set of actual measurement points {P i} Rotation matrix R 0. Translation matrix T 0 and single-sided theoretical margin ε 0. Establish a registration model with residual constraints and solve it to obtain the registered numerical model space coordinate system; Step S5: Correct the measurement program based on the registered digital model space coordinate system, and execute the corrected measurement program in the machining coordinate system to obtain a new set of actual measurement points. {P i ′′ } This step constitutes a closed-loop feedback loop. After registration, the theoretical coordinate system has shifted, and the original measurement procedure is no longer applicable to the new coordinate system. By correcting the measurement procedure and re-measuring, a new set of actual measurement points can be obtained. This set of data is collected in the registered coordinate system and can truly reflect the degree of agreement between the theoretical model and the actual position of the blank.

[0009] Step S6: For the new set of actual measurement points {P i ′′ } The margin constraint condition is verified. If the condition is met, the ceramic matrix composite blank is machined on one side using the registered digital model space coordinate system to obtain the machined surface. If the condition is not met, the measurement point is reselected or the registration calculation and verification are performed again based on the obtained transformation parameters until the condition is met, and then the single-sided machining is performed. This step verifies the margin constraint condition of the new actual measurement point (checking whether the directed distance does not exceed the limit). ε By checking the coordinate system (including whether the orientation is correct), we can determine whether the current registration result truly meets the processing requirements. If the verification passes, processing can proceed with confidence; if it fails, two iterative strategies are available: one is to reselect new measurement points and start registration from scratch, and the other is to recalculate the registration based on the existing transformation parameters for a new set of points. This closed-loop iterative mechanism ensures that the final coordinate system used is a fully verified and reliable result, avoiding the risks of blind processing and significantly improving the success rate and stability of processing.

[0010] Step S7: Flip the ceramic matrix composite blank so that the unmachined surface faces upwards; if the machined surface can be measured, measure the machined surface, establish a rigid body transformation model with tolerance constraints based on the measurement data, register it, and then machine it; if the machined surface cannot be measured, measure the unmachined surface, and calculate the maximum allowance of the ceramic matrix composite part based on the high points of the measured point set. ε 2. Combined with single-sided tolerance ε 3. Process the unprocessed surface.

[0011] After flipping, the positional relationship of the parts changes fundamentally, making it crucial to ensure the relative positional accuracy between the unmachined and machined surfaces. This invention provides two adaptive modes: When the machined surface can be measured, a rigid body transformation model with tolerance constraints is established to control the positional deviation of the machined surface within the allowable tolerance range, thus ensuring that the machined surface is not damaged secondaryly after flipping, while providing an accurate positioning reference for the unmachined surface. When the machined surface cannot be measured (e.g., obscured by tooling), the unmachined surface is measured instead. The maximum allowance is calculated based on the high points of the measured point set, and the cutting depth is directly determined by combining the single-surface tolerance. This method does not rely on the data of the machined surface and can handle various complex clamping conditions.

[0012] These two modes cover the vast majority of actual production scenarios, requiring no modification to existing equipment, and are highly versatile and practical.

[0013] Preferably, in step S4, the objective function of the margin constraint registration model is: .

[0014] Preferably, the margin-constrained registration model includes the following constraints: First constraint: The set of theoretical measurement points {P i ′} After coordinate transformation, it matches the actual set of measured points. {P i ′′ } distance ′ Not less than the theoretical margin of one side ε 0; The first constraint requires that after coordinate transformation, the theoretical measurement point is located inside the material of the actual measurement point, and the directional distance between the two is not less than the preset single-sided theoretical margin. ε 0. Traditional registration only pursues the best fit of geometric shape without considering machining allowance. It may "attach" the theoretical model to the surface of the blank or even embed it in the external space, resulting in cutting through the blank during actual processing. However, the allowance constraint of this invention forces the theoretical point to "sink" into the interior of the blank with a safety margin, thereby fundamentally avoiding overcutting and scrapping, and ensuring the yield rate.

[0015] Second constraint: The set of theoretical measurement points {P i ′ } After coordinate transformation, it matches the actual set of measured points. {P i ′′ } directed vectors The first constraint points to the non-material side. The second constraint prevents physically unreasonable "flipping" during registration—for example, the theoretical measurement point being mistakenly moved to the opposite side of the actual measurement point, leading to a completely incorrect subsequent machining direction. This constraint guarantees the physical correctness of the registration result, ensuring that the theoretical model is always on the correct side of the blank.

[0016] By combining these two constraints, this step can automatically find an optimal rigid body transformation under the premise of meeting the machining allowance requirements, and intelligently distribute the limited blank allowance among the machining areas. This ensures that all areas have enough material to be removed, while minimizing geometric deviations, thus achieving a balance between accuracy and safety.

[0017] Preferably, the mathematical expression for the first constraint is: ,in, n i =[a,b,c] This is the unit normal vector of the theoretical measurement point.

[0018] Preferably, the mathematical expression for the second constraint is: ,in, n z =[0,0,1] The Z-axis unit vector is used as the direction criterion, thus achieving directional constraints simply and effectively. In actual machining scenarios, parts are typically clamped with their principal planes parallel to the XY plane, and the Z-axis direction is the tool feed direction. Therefore, this constraint ensures that the theoretical measurement point is always located below the actual measurement point (i.e., close to the inside of the workpiece), consistent with the actual machining direction, preventing the theoretical model from flipping to the outside of the workpiece due to registration errors.

[0019] Preferably, after step S1 and before step S2, the method further includes: performing a grouting treatment on the measuring surface of the ceramic matrix composite blank to fill surface pores and fiber undulation defects. The surface of the ceramic matrix composite blank naturally contains numerous pores and fiber undulation defects. These defects can cause the probe to generate false readings during contact measurement; the probe may fall into pores or get stuck between fiber gaps, causing the measurement point position to deviate from the true material surface. The grouting treatment, by filling these surface defects, provides the probe with a continuous and smooth measuring surface, thereby significantly improving the authenticity and accuracy of the measurement data.

[0020] Preferably, in step S6, the verification of the margin constraint condition includes the following steps: Calculate the new set of actual measurement points {P i ′′ } With the set of theoretical measurement points {Pi ′ } Distance between g i ′ And determine whether the following conditions are met simultaneously: g i ′ ≤ ε 0, and the theoretical point on the upper surface P i ′ After coordinate transformation, at the measured point P i ′′ Below.

[0021] Preferably, in step S6, if the condition is not met, the iteration is performed in any of the following ways: Method 1: Reselect the theoretical measurement point and the actual measurement point, and repeat steps S2 to S6; Method 2: Based on the obtained rotation matrix R 0 and translation matrix T 0. Repeat steps S4 to S6 for the new set of actual measurement points and the new set of theoretical measurement points.

[0022] Preferably, in step S7, the surface of the flipping fixture used for flipping clamping is provided with an adhesive groove to achieve a tight fit between the machined surface and the flipping fixture. During flipping clamping, adhesive needs to be applied between the machined surface and the fixture to ensure a tight fit. Without an adhesive groove, excess adhesive will overflow under pressure, potentially causing an uneven adhesive layer between the machined surface and the fixture, affecting positioning accuracy. The adhesive groove design provides space to accommodate excess adhesive, making the adhesive layer thickness uniform and controllable, while increasing the bonding area and improving the stability and repeatability of the flipping clamping.

[0023] Preferably, in step S7, the mathematical expression of the rigid body transformation model with tolerance constraints is: ,in, n j The unit normal vector of the theoretical measurement point. ε 1 represents a single-sided tolerance. R 0 ′ Let be a rotation matrix. T 0 ′ It is a translation matrix.

[0024] (III) Beneficial Effects The above-described technical solution of the present invention has at least the following advantages: This invention, through triple safeguards of margin constraints, directional constraints, and verification iterations, fundamentally eliminates the possibility of cutting through the blank due to improper registration, making it particularly suitable for machining ceramic matrix composite parts. This invention can control the adjustment of registration calculations based on constraints, achieving deformation and defect compensation for near-net-shape ceramic matrix composite part blanks, and realizing a complete encapsulation effect of the near-net-shape composite blank on the part, which is beneficial for high-quality machining. This invention reduces the accuracy requirements for part clamping, lowers the requirements for operator experience and skill, reduces interference from manual operations, and improves the efficiency of manual operations, making it suitable for large-scale mass production. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic flowchart of an adaptive processing method for ceramic matrix composite parts provided in an embodiment of the present invention.

[0027] Figure 2 This is a schematic diagram of the structure of the ceramic matrix composite material blank provided in the embodiment of the present invention.

[0028] Figure 3 This is a schematic diagram of the measurement trajectory on a ceramic matrix composite blank provided in an embodiment of the present invention.

[0029] Figure 4 This is an example of the registration process of the adaptive processing method for ceramic matrix composite parts provided in the embodiments of the present invention.

[0030] Figure 5 This is an example of the registration result of the adaptive processing method for ceramic matrix composite parts provided in the embodiments of the present invention.

[0031] Figure 6 This is an example of the registration algorithm output coordinates of the adaptive machining method for ceramic matrix composite parts provided in this embodiment of the invention.

[0032] Figure 7 This is a schematic diagram of the adhesive tank provided in an embodiment of the present invention.

[0033] Figure 8 This is an example of probe swing angle measurement in an adaptive machining method for ceramic matrix composite parts provided in this embodiment of the invention. Detailed Implementation

[0034] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0035] The specific implementation of the present invention will be described in more detail below with reference to specific embodiments: like Figure 1 As shown, the present invention provides an adaptive processing method for ceramic matrix composite parts, comprising the following steps: Step S1: Clamp and fix the ceramic matrix composite blank and establish a machining coordinate system; the structure of the ceramic matrix composite blank provided in this embodiment is as follows: Figure 2 As shown; based on the measurement process requirements, a machining coordinate system is established on the machined part. OXYZ-1 ; Step S1-1: Filling the pits and defects on the surface of the ceramic matrix composite blank. In this example, filler is used to fill the holes, fiber undulations and other defects on the surface of the ceramic matrix composite blank. Step S2, in the theoretical coordinate system OXYZ-0 Select the theoretical measurement point set below {P i ′ } , P i ′ =[x i ′ ,y i ′ ,z i ′ ] T The actual shape and position of the ceramic matrix composite blank clamped under the tooling were measured sequentially to obtain the actual measurement point set. {P i } (Based on machining coordinate system) OXYZ-1 (to perform calibration) P i =[x i ,y i ,z i ] T ,in i The points are numbered, and the measurement trajectory is as follows: Figure 3 As shown; Step S3: Set up the theoretical measurement points{P i ′ } Set of actual measurement points {P i } Perform coordinate transformations based on rigid body transformation relationships and solve for the rotation matrix. R 0 and translation matrix T 0; The mathematical expression is as follows: (1) The residual constraint conditions for the residual constraint registration model are: ① Theoretical measurement point set {P i } After coordinate transformation, it matches the actual set of measured points. {P i ′′ } distance Not less than the theoretical margin of one side ε 0; The mathematical expression is as follows: (2) In the formula, g i The distance between the theoretical measurement point and the actual measurement point is denoted as , where . n i =[a,b,c] This is the unit normal vector of the theoretical measurement point.

[0036] ② Theoretical measurement point set {P i ′ } After coordinate transformation, it matches the actual set of measured points. {P i ′′ } directed vectors Pointing to the non-material side. For example, in the vertical direction, the theoretical measurement point on the upper surface of the blank. P i ′ After coordinate transformation, at the measured point P i Below; the mathematical expression is as follows: (3) In the formula, h i Used to determine the location of theoretical points n z =[0,0,1] It is the unit vector along the Z-axis; Establish residual constraint equations to transform the distance and constraint model into a system of linear equations; use the Lagrange optimization method to solve the constraint problem. Step S4: Based on the actual set of measurement points {P i } and the theoretical measurement point set {P i ′ } and single-sided theoretical margin ε 0. Establish a registration model with margin constraints and solve it to obtain the registered theoretical coordinate system. OXYZ-0 ′ The solution model is shown in equations (4) to (6), and the registration process is as follows: Figure 4 As shown; (4) (5) (6) In the formula, g i The distance between the theoretical measurement point and the actual measurement point. h i Used to determine the location of theoretical measurement points. n i =[a, b,c] The unit normal vector of the theoretical measurement point. ε 0 represents the theoretical margin for a single surface. n z =[0,0,1] It is the unit vector along the Z-axis.

[0037] Step S5: Based on the registered digital model space coordinate system OXYZ-0 ′ Perform measurement program corrections and obtain a new set of actual measurement points after correction. {P i ′′ } ; Step S6: For the new set of actual measurement points {P i ′′ } and theoretical measurement point set {P i ′ } Perform margin constraint verification and calculate the actual set of measurement points. {P i ′′ } For the theoretical measurement point set {P i′ } distance g i ′ With single-sided theoretical margin ε The relationship between 0 and 1, the registration result is as follows Figure 5 As shown, the output registration coordinates are as follows: Figure 6 As shown, the verification process for the margin constraint is as follows: ①If g i ′ ≤ ε 0 and the theoretical measurement point on the upper surface P i ′ After coordinate transformation, at the actual measurement point P i ′′ Below this, the margin constraint condition is met, and a reference coordinate system can be used. OXYZ-0 ′ Processing; ②Otherwise: 1) Reselect the theoretical measurement points and actual measurement points, and repeat steps S2 to S6 to perform measurement registration and verification; 2) Based on the obtained rotation matrix R 0 and translation matrix T 0. Repeat steps S4 to S6 for the new set of actual measurement points and the new set of theoretical measurement points. Step S6-1: Complete the measurement, registration, and machining of the upper surface of the ceramic matrix composite blank to obtain the machined surface; Step S7: Flip and clamp the ceramic matrix composite part (ceramic matrix composite blank) that has been machined on one side. The tooling surface after flipping has a glue-receiving groove, such as... Figure 7 As shown, a tight fit is achieved between the machined surface of the ceramic matrix composite part and the flipping fixture; after flipping, the ceramic matrix composite part is adaptively processed using two methods depending on the actual situation: ① If the machined surface can be measured after clamping, based on the second theoretical coordinate system OXYZ-2 Take a set of measurement points on the surface area of ​​the machined surface. {Pj} Measure the machined surfaces sequentially, such as Figure 8 As shown, the actual location data set was obtained. {Pj} (Based on the second actual coordinate system) OXYZ-3 (to perform calibration), among which j Number the points and establish rigid body transformation relationships (refer to equation (1)). Constraint relationships are referenced to equation (7): (7) In the formula, gj The distance between the theoretical measurement point and the actual measurement point. n j The unit normal vector of the theoretical measurement point. ε 1 represents a single-sided tolerance. R 0 ′ Let be a rotation matrix. T 0 ′ It is a translation matrix; ② If the surface of the machined part cannot be measured after clamping, the theoretical coordinate system should be used as a reference. OXYZ-2 Take a set of measurement points for the unprocessed surface area {Pj} The actual surface was measured sequentially to obtain the set of measured points. {Pj} (Based on the actual coordinate system OXYZ-3), where j is the point number, the maximum allowance ε2 of the ceramic matrix composite part is calculated based on the high point of the measured point set. The maximum allowance ε2 represents how much material needs to be removed from the thickest part of the unprocessed surface in the current blank state. Combined with the single-sided tolerance ε3 of the part, machining is performed, that is, the actual amount removed during machining is the difference between ε2 and ε3.

[0038] Complete the parts registration and machining.

[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive machining method for ceramic matrix composite parts, characterized in that, Includes the following steps: Step S1: Clamp and fix the ceramic matrix composite blank and establish a machining coordinate system; Step S2: Select the set of theoretical measurement points in the theoretical coordinate system. {P i ′ } The actual shape and position of the ceramic matrix composite blank clamped under the tooling are measured sequentially in the machining coordinate system to obtain the actual measurement point set corresponding to the theoretical measurement point set. {P i } ,in i Number the location; Step S3: Set up the theoretical measurement points {P i ′ } Set of actual measurement points {P i } Perform coordinate transformations and solve the rotation matrix based on rigid body transformation relationships. R 0 and translation matrix T 0; Step S4: Based on the theoretical measurement point set {P i ′ } Set of actual measurement points {P i } Rotation matrix R 0. Translation matrix T 0 and single-sided theoretical margin ε 0. Establish a registration model with residual constraints and solve it to obtain the registered numerical model space coordinate system; Step S5: Correct the measurement program based on the registered digital model space coordinate system, and execute the corrected measurement program in the machining coordinate system to obtain a new set of actual measurement points. {P i ′′ } ; Step S6: For the new set of actual measurement points {P i ′′ } Perform margin constraint condition verification. If the condition is met, use the registered digital model space coordinate system to perform single-sided machining on the ceramic matrix composite blank to obtain the machined surface. If the conditions are not met, the measurement points are reselected or the registration calculation and verification are performed again based on the obtained transformation parameters until the conditions are met, and then single-sided processing is performed. Step S7: Flip the ceramic matrix composite blank so that the unmachined surface faces upwards; if the machined surface can be measured, measure the machined surface, establish a rigid body transformation model with tolerance constraints based on the measurement data, register it, and then machine it; if the machined surface cannot be measured, measure the unmachined surface, and calculate the maximum allowance of the ceramic matrix composite part based on the high points of the measured point set. ε 2. Combined with single-sided tolerance ε 3. Process the unprocessed surface.

2. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, In step S4, the objective function of the residual constraint registration model is: .

3. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, The margin-constrained registration model includes the following constraints: First constraint: The set of theoretical measurement points {P i ′ } After coordinate transformation, it matches the actual set of measured points. {P i ′′ } distance Not less than the theoretical margin of one side ε 0; Second constraint: The set of theoretical measurement points {P i ′ } After coordinate transformation, it matches the actual set of measured points. {P i ′′ } directed vectors Pointing to the non-material side.

4. The adaptive machining method for ceramic matrix composite parts as described in claim 3, characterized in that, The mathematical expression for the first constraint is: ,in, n i =[a,b,c] This is the unit normal vector of the theoretical measurement point.

5. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, The mathematical expression for the second constraint is: ,in, n z =[0,0,1] It is the unit vector along the Z-axis.

6. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, The process after step S1 and before step S2 also includes: filling the gaps on the measured surface of the ceramic matrix composite blank to fill the surface pores and fiber undulation defects.

7. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, In step S6, the verification of the margin constraint condition includes the following steps: Calculate the new set of actual measurement points {P i ′′ } With the set of theoretical measurement points {P i ′ } Distance between g i ′ And determine whether the following conditions are met simultaneously: g i ′ ≤ ε 0, and the theoretical point on the upper surface P i ′ After coordinate transformation, at the measured point P i ′′ Below.

8. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, In step S6, if the condition is not met, the iteration shall be performed in any of the following ways: Method 1: Reselect the theoretical measurement point and the actual measurement point, and repeat steps S2 to S6; Method 2: Based on the obtained rotation matrix R 0 and translation matrix T 0. Repeat steps S4 to S6 for the new set of actual measurement points and the new set of theoretical measurement points.

9. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, In step S7, the surface of the flipping fixture used for flipping clamping is provided with a glue-receiving groove to achieve a tight fit between the machined surface and the flipping fixture.

10. The adaptive machining method for ceramic matrix composite parts as described in claim 1, characterized in that, In step S7, the mathematical expression for the rigid body transformation model with tolerance constraints is: ,in, n j The unit normal vector of the theoretical measurement point. ε 1 represents a single-sided tolerance. R 0 ′ Let be a rotation matrix. T 0 ′ It is a translation matrix.

Images