A composite component model reconstruction method considering manufacturing errors

Abstract

The present invention discloses a composite component model reconstruction method considering manufacturing errors, which belongs to the technical field of manufacturing quality prediction and control; the specific steps are: obtaining part surface sampling data; constructing a part manufacturing error function; realizing modal decomposition of manufacturing errors by discrete cosine forward transform; error modal identification; realizing model reconstruction by discrete cosine inverse transform, and obtaining reconstruction data of the upper surface of the part; judging whether the reconstruction accuracy is met, and if so, proceeding to the next step; if not, adjusting the energy compaction degree and reconstruction contribution index, and repeating steps 4-6; importing the part reconstruction data f′(m, n) into a three-dimensional software to construct a three-dimensional solid model. The composite model component reconstruction method can simply and effectively realize the decomposition and identification of component surface manufacturing errors, and has high reconstruction accuracy, which is conducive to the statistical analysis and expression of manufacturing errors of multiple components in the same batch.

Description

Technical Field

[0001] The invention belongs to the technical field of manufacturing quality prediction and control, and in particular relates to a composite material component model reconstruction method taking manufacturing errors into consideration. Background Art

[0002] The manufacturing error of composite components is an important cause of assembly errors and is also an unavoidable error in the manufacturing process. In order to meet the assembly requirements of products and ensure the assembly performance of products, it is necessary to control the manufacturing errors of components and analyze the effects of various errors on assembly quality. With the continuous improvement of product precision requirements, the precision analysis method that simplifies the surface of parts into an ideal geometric surface and ignores the influence of surface morphology on assembly is no longer applicable. Therefore, it is necessary to consider multiple manufacturing errors such as dimensional error, orientation error and surface shape error at the same time, so as to establish a reconstruction model of composite components that considers the coupling of multiple manufacturing errors, so as to ensure the accuracy of the prediction of composite component assembly precision, and lay the foundation for the subsequent virtual assembly simulation, especially the numerical simulation of assembly deformation and assembly stress.

[0003] In the prior art, a method for calculating the actual assembly error is provided, which takes into account the shape error of the mating surface of the parts and the assembly deformation error. This method can significantly improve the accuracy of manufacturing quality prediction, but the actual three-dimensional digital model is not obtained, and the subsequent assembly performance prediction cannot be carried out. There is also a method for obtaining the surface real assembly error, which takes into account the part size error, positioning orientation error, and deformation error caused by stability and stress load during assembly. This method can predict the actual gap between the mating surfaces and improve the accuracy of assembly error prediction, but the influence of the actual surface morphology is not considered.

[0004] At present, most reconstruction models of composite components only consider the size error and orientation error of the parts or one of them, and rarely consider the surface morphology characteristics of the components and the above-mentioned multi-manufacturing error coupling, which leads to low accuracy of the reconstruction model of composite components. At the same time, the current component model reconstruction method is difficult to generate a solid model with the help of CAD software, and it is impossible to perform subsequent finite element simulation analysis of the assembly process, which seriously affects the stability and accuracy of assembly performance prediction. Therefore, a simple and fast composite component model reconstruction method with high reconstruction accuracy and suitable for solid model construction is needed. Summary of the invention

[0005] Technical issues to be solved:

[0006] In order to avoid the shortcomings of the prior art, the present invention provides a method for reconstructing a composite material component model taking manufacturing errors into consideration. The method comprehensively considers various manufacturing error coupling situations, is simple to operate, and has high reconstruction accuracy and efficiency. The composite material component model can be reconstructed with the least amount of data possible on the basis of ensuring reconstruction accuracy, and a solid component model can be generated with the help of CAD software.

[0007] The technical solution of the present invention is: a composite material component model reconstruction method considering manufacturing errors, the specific steps are as follows:

[0008] Step 1: Obtain the part surface sampling data and obtain the actual measurement value data set f(m,n) of the part surface sampling points s ;

[0009] Step 2: Construct part manufacturing error function:

[0010] f(m,n)=f(m,n) s -f(m,n) m (1)

[0011] Among them, f(m,n) m is the nominal value of the data sampling point, f(m,n) s is the actual measurement value of the data sampling point, f(m,n) is the error value at the node, if it is a negative value, it means that the data sampling point is lower than the nominal feature point of the component surface, otherwise it is higher than the nominal feature point of the component surface;

[0012] Step 3: The modal decomposition of manufacturing error is achieved through discrete cosine transform. The formula is as follows:

[0013]

[0014]

[0015]

[0016] Among them, C(u,v) is the transformation coefficient matrix after forward transformation, g(m,n,u,v) is the kernel function of discrete cosine forward transformation, m and n are the data sampling point numbers, M and N are the total number of data points, and u and v are the sampling frequency values;

[0017] Step 4: Error mode identification:

[0018]

[0019]

[0020] Among them, E is the energy compaction degree, 0≤E≤100%, the larger the E, the smaller the modal set Ω eThe more error modes are included, when E = 100%, all error modes are included in the mode set Ω1; β is the reconstruction contribution, which is used to judge the contribution of the error mode to the entire manufacturing error model; B is the mode contribution;

[0021] Step 5: Reconstruct the model through inverse discrete cosine transform to obtain the reconstructed data of the upper surface of the part:

[0022]

[0023] Step 6: Determine whether the reconstruction accuracy is met. If it is met, proceed to the next step. If it is not met, adjust the energy compaction degree and reconstruction contribution index and repeat steps 4-6.

[0024] Step 7: Import the part reconstruction data f′(m, n) into the 3D software to construct a 3D solid model.

[0025] A further technical solution of the present invention is: in step 1, a three-coordinate measuring instrument is used to measure the surface of the part; when the composite flat plate component is solidified, the lower surface of the flat plate component is constrained by the solidification mold and has a smooth surface, which has little impact, so only the upper surface needs to be measured to obtain the part surface data set f(m,n) s .

[0026] A further technical solution of the present invention is: in the step 2, a grid model is generated based on the nominal features of the upper surface of the part according to the set sampling frequency of the three-dimensional coordinate measuring machine, the data sampling points are set as grid nodes, and the normal distance of the grid nodes is used to characterize the part manufacturing error, that is, the difference between the grid nodes of the actual surface and the ideal surface is used to characterize it, thereby obtaining a manufacturing error function.

[0027] A further technical solution of the present invention is: in the step 3, the obtained component surface manufacturing error field f(m,n) is subjected to a discrete cosine transform through a transformation algorithm with the aid of MATLAB software to obtain a transformation coefficient matrix C(u,v), which includes M×N transformation coefficients. Each transformation coefficient in the matrix can correspond to a manufacturing error mode after inverse transformation. Each manufacturing error mode is related to the error source of the composite material processing process. Therefore, there are a total of M×N error modes, which realizes the modal decomposition of the component surface manufacturing error.

[0028] A further technical solution of the present invention is: in step 4, a modal identification criterion is used to realize the error modal identification with large reconstruction contribution.

[0029] A further technical solution of the present invention is: in step 4, in order to ensure that the overall model reconstruction error accuracy meets the assembly requirements and to screen out the error modes with less influence on the reconstruction, the energy compaction degree E should be greater than 99%, and the single error mode contribution degree should be greater than 1 / 10MN; therefore, given the energy compaction degree E and the mode contribution degree B, a large part of the error modes with less influence will be deleted from the M×N error modes, thereby achieving the purpose of simplifying the data storage and calculation difficulty under the condition of ensuring the reconstruction accuracy. If the restriction requirements on the energy compaction degree E and the mode contribution degree B cannot be met at the same time, the energy compaction degree E should be guaranteed to meet the requirements first, and the requirements on the mode contribution degree B should be appropriately relaxed, that is, more error modes are identified to ensure the reconstruction accuracy, but the effect of simplifying the data will be weakened, wherein the identified k error modes are included in the mode set Ω:

[0030] Ω=Ω1∩Ω2 (6).

[0031] A further technical solution of the present invention is: in step 5, the identified manufacturing error modes are superimposed to obtain a new transformation coefficient matrix C(u, v)′, and then an inverse discrete cosine transform is performed to obtain the reconstructed data of the upper surface of the part.

[0032] A further technical solution of the present invention is: in step 6, the reconstruction error rate is used to characterize the reconstruction accuracy of the component model. Since the reconstruction error rate is 0% when the modal number is not identified and deleted, if the reconstructed surface does not meet the required reconstruction accuracy, the energy compaction degree E and the error reconstruction contribution degree B are readjusted to meet the accuracy requirements:

[0033]

[0034]

[0035] Where e is the root mean square error, η is the reconstruction error rate, f(m,n) is the error value of the point at (m,n) on the measured plane, and f(m,n)′ is the error value of the point at (m,n) on the reconstructed plane.

[0036] A further technical solution of the present invention is: in the step 7, CATIA software is imported, and a digital editor module is used to import a manufacturing error function set; a fast surface reconstruction module is used to fit the error function set to complete the surface modeling function; a general surface modeling module is used to stretch the surface according to the corresponding thickness of the component to realize the construction of a three-dimensional solid model.

[0037] Beneficial Effects

[0038] The beneficial effects of the present invention are:

[0039] 1. This composite material model component reconstruction method can simply and effectively realize the decomposition and identification of component surface manufacturing errors, and the reconstruction accuracy is high, which is conducive to the statistical analysis and expression of manufacturing errors of multiple components in the same batch.

[0040] 2. The error modes are identified through energy compaction and error contribution indicators. On the basis of ensuring reconstruction accuracy, less data is used to characterize the surface manufacturing errors of components, which simplifies the difficulty of data storage and calculation and greatly improves the calculation efficiency.

[0041] 3. By adjusting the modal identification criteria, different requirements for the reconstruction accuracy of composite component models in different assembly scenarios can be met.

[0042] 4. The reconstructed data can be directly imported into CATIA software for 3D solid model construction, laying the foundation for finite element simulation analysis of composite component assembly. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Schematic diagram of a composite plate component including manufacturing errors;

[0044] Figure 2 A schematic diagram of the measured surface of a composite material component taking into account manufacturing errors provided by the present invention;

[0045] Figure 3 A schematic diagram of a reconstructed surface of a composite material component taking manufacturing errors into consideration provided by the present invention;

[0046] Figure 4 A schematic diagram showing the effect comparison of a composite material component model reconstruction method taking manufacturing errors into consideration provided by the present invention;

[0047] Figure 5 A flow chart of a composite material component model reconstruction method considering manufacturing errors provided by the present invention;

[0048] Figure 6 A three-dimensional solid model diagram of a composite material component taking manufacturing errors into consideration is provided by the present invention. DETAILED DESCRIPTION

[0049] The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0050] Reference Figure 1 In the present invention, a composite material flat plate component is taken as an example, the lower surface of the entire flat plate component is an ideal surface, and the upper surface is a non-ideal surface containing manufacturing errors. Figure 5 As shown, it is a flow chart of a composite component model reconstruction method considering manufacturing errors provided by the present invention, which comprises the following steps in sequence:

[0051] Step 1: Use a three-coordinate measuring machine to measure the surface of the part. In this example, 6×6 sampling measurement points are selected to obtain a part surface data set.

[0052] Step 2: Generate a grid model based on the nominal features of the upper surface of the component according to the sampling frequency of the measuring instrument mentioned above, and set the data sampling points as grid nodes. Table 1 shows the manufacturing error of the part represented by the normal distance of the grid nodes, that is, the difference between the actual surface and the ideal surface at the grid nodes:

[0053] f(m,n)=f(m,n) s -f(m,n) m (1)

[0054] Table 1

[0055]

[0056]

[0057] Step 3: With the help of MATLAB software, the obtained component surface manufacturing error field f(m,n) is subjected to discrete cosine positive transformation through transformation algorithm to obtain the transformation coefficient matrix C(u,v), which contains 36 transformation coefficients. Each transformation coefficient in the matrix can correspond to a manufacturing error mode after inverse transformation. Therefore, there are 36 error modes in total, realizing the modal decomposition of component surface manufacturing error, as shown in Figure 2. Figure 2 As shown:

[0058]

[0059]

[0060]

[0061] Among them, C(u,v) is the transformation coefficient matrix after forward transformation, g(m,n,u,v) is the kernel function of discrete cosine forward transform, m and n are the data sampling point numbers, and u and v are the sampling frequency values.

[0062] Step 4: Use energy compaction and error reconstruction contribution index (modal identification criterion) to identify error modes with large reconstruction contribution:

[0063]

[0064]

[0065] In this example, the energy compaction degree E is given as 99% and the modal contribution B is given as 0.1%. 12 error modes with less influence will be deleted from 36 error modes, and the deletion rate is 33%, which greatly simplifies the difficulty of data storage and calculation and improves the calculation rate. The identified error modes are included in the modal set Ω:

[0066] Ω=Ω1∩Ω2 (6)

[0067] Step 5: Superimpose the identified error modes to obtain a new transformation coefficient matrix C(u,v)′, see Table 2, and then reconstruct the component surface model through inverse discrete cosine transform, as shown in Figure 3 As shown:

[0068]

[0069] Table 2

[0070] 1 2 3 4 5 6 1 0.1220 0 0.0441 -0.1293 -0.0127 0 2 0.0347 0.1303 0.1593 -0.1046 0 -0.0246 3 -0.0600 0.0936 0.0708 0.1029 0 0.0722 4 -0.0553 -0.0447 -0.0118 0.0393 -0.0130 0 5 -0.0297 0 -0.0160 0 0.0298 -0.0152 6 0 0 -0.0119 0 0 0

[0071] Step 6: Different assembly scenarios have different requirements for the reconstruction accuracy of composite component models. In this example, the reconstruction error rate is used to characterize the reconstruction accuracy of the component model. When the energy compaction degree E is 99% and the modal contribution B is 0.1%, Figure 2-3 As shown, the root mean square error e of this example is 0.004 mm, the reconstruction error rate is 6.6%, and the data reduction rate is 33%, which meets the reconstruction accuracy requirements and data simplification effects.

[0072] like Figure 4 As shown in the figure, under the condition of the same modal number, the DCT-based composite component model reconstruction method proposed in the present invention has higher accuracy than the previous FFT reconstruction method. At the same time, when the modal number is reduced to 20 and the data deletion rate reaches 44%, the reconstruction accuracy of the model reconstruction method proposed in the present invention decreases slowly, and the reconstruction error rate of about 10% can still be guaranteed, which is much higher than the reconstruction accuracy of the FFT reconstruction method. At the same time, combined with the energy compaction curve and the DCT reconstruction error rate curve, the energy compaction is maintained above 90% in a large range, and the reconstruction error rate is maintained at a low level within this range, that is, the reconstruction accuracy is high. Therefore, on the basis of ensuring the reconstruction accuracy, less data is used to characterize the surface manufacturing error of the component, which greatly simplifies the difficulty of calculation and improves the calculation rate. The effect of using energy compaction as a manufacturing error modal identification criterion is significant.

[0073] Step 7: Figure 6As shown in the figure, the reconstruction data f(m, n)′ of the composite component is imported into the 3D software to build a 3D solid model. Specifically, the manufacturing error function set is imported using the Digitized Shape Editor (DSE module); the error function set is fitted using the Quick Surface Reconstruction (QSR module) to complete the surface modeling function; the Gennrative Shape Design (GSD module) is used to stretch the surface according to the thickness of the component of 3.6 mm to realize the construction of the 3D solid model.

[0074] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and intent of the present invention.

Claims

1. A composite component model reconstruction method considering manufacturing errors, characterized in that The specific steps are as follows: Step 1: Obtain the part surface sampling data and obtain the actual measurement value data set f(m,n) of the part surface sampling points s ; Step 2: Construct part manufacturing error function: f(m,n)=f(m,n) s -f(m,n) m (1) Among them, f(m,n) m is the nominal value of the data sampling point, f(m,n) s is the actual measured value of the data sampling point, f(m,n) is the error value at the node, if it is a negative value, it means that the data sampling point is lower than the nominal feature point of the component surface, otherwise it is higher than the nominal feature point of the component surface; Step 3: The modal decomposition of manufacturing error is achieved through discrete cosine transform. The formula is as follows: Among them, C(u,v) is the transformation coefficient matrix after forward transformation, g(m,n,u,v) is the kernel function of discrete cosine forward transformation, m and n are the data sampling point numbers, M and N are the total number of data points, and u and v are the sampling frequency values; Step 4: Error mode identification: Among them, E is the energy compaction degree, 0≤E≤100%, the larger the E, the smaller the modal set Ω e The more error modes are included, when E = 100%, all error modes are included in the mode set Ω1; β is the reconstruction contribution, which is used to judge the contribution of the error mode to the entire manufacturing error model; B is the mode contribution; Step 5: Reconstruct the model through inverse discrete cosine transform to obtain the reconstructed data of the upper surface of the part: Step 6: Determine whether the reconstruction accuracy is met. If it is met, proceed to the next step. If it is not met, adjust the energy compaction degree and reconstruction contribution index and repeat steps 4-6. Step 7: Import the part reconstruction data f′(m, n) into the 3D software to construct a 3D solid model.

2. A composite material component model reconstruction method considering manufacturing errors according to claim 1, characterized in that: In step 1, a three-coordinate measuring machine is used to measure the surface of the part; when the composite flat plate component is solidified, the lower surface of the flat plate component is constrained by the solidification mold and has a smooth surface, so the impact is small. Only the upper surface needs to be measured to obtain the part surface data set f(m,n) s .

3. A composite material component model reconstruction method considering manufacturing errors according to claim 1, characterized in that: In step 2, a grid model is generated based on the nominal features of the upper surface of the part according to the set sampling frequency of the three-dimensional coordinate measuring machine, the data sampling points are set as grid nodes, and the normal distance of the grid nodes is used to characterize the part manufacturing error, that is, the difference between the grid nodes of the actual surface and the ideal surface is used to characterize it, thereby obtaining a manufacturing error function.

4. A composite material component model reconstruction method considering manufacturing errors according to claim 1, characterized in that: In step 3, the obtained component surface manufacturing error field f(m,n) is subjected to discrete cosine transform by transformation algorithm with the aid of MATLAB software to obtain a transformation coefficient matrix C(u,v), which includes M×N transformation coefficients. Each transformation coefficient in the matrix can correspond to a manufacturing error mode after inverse transformation. Each manufacturing error mode is related to the error source of the composite material processing process. Therefore, there are a total of M×N error modes, which realizes the modal decomposition of the component surface manufacturing error.

5. The method for reconstructing a composite material component model taking into account manufacturing errors according to claim 1, characterized in that: In step 4, a modal identification criterion is used to realize the error modal identification with large reconstruction contribution.

6. The composite material component model reconstruction method considering manufacturing errors according to claim 1, characterized in that: In step 4, in order to ensure that the overall model reconstruction error accuracy meets the assembly requirements and to screen out the error modes with less influence on the reconstruction, the energy compaction degree E should be greater than 99%, and the single error mode contribution degree should be greater than 1 / 10MN; therefore, given the energy compaction degree E and the mode contribution degree B, a large part of the error modes with less influence will be deleted from the M×N error modes, so as to achieve the purpose of simplifying the data storage and calculation difficulty under the condition of ensuring the reconstruction accuracy. If the restriction requirements on the energy compaction degree E and the mode contribution degree B cannot be met at the same time, the energy compaction degree E should be guaranteed to meet the requirements first, and the requirements on the mode contribution degree B should be appropriately relaxed, that is, more error modes are identified to ensure the reconstruction accuracy, but the effect of simplifying the data will be weakened, and the identified k error modes are included in the mode set Ω: Ω=Ω1∩Ω2 (6).

7. The method for reconstructing a composite material component model taking manufacturing errors into consideration according to claim 1, characterized in that: In step 5, the identified manufacturing error modes are superimposed to obtain a new transformation coefficient matrix C(u, v)′, and then an inverse discrete cosine transform is performed to obtain the reconstructed data of the upper surface of the part.

8. The method for reconstructing a composite material component model taking into account manufacturing errors according to claim 1, characterized in that: In step 6, the reconstruction error rate is used to characterize the reconstruction accuracy of the component model. Since the reconstruction error rate is 0% when the modal number is not identified and deleted, if the reconstructed surface does not meet the required reconstruction accuracy, the energy compaction degree E and the error reconstruction contribution degree B are readjusted to meet the accuracy requirements: Where e is the root mean square error, η is the reconstruction error rate, f(m,n) is the error value of the point at (m,n) on the measured plane, and f(m,n)′ is the error value of the point at (m,n) on the reconstructed plane.

9. The composite material component model reconstruction method considering manufacturing errors according to claim 1, characterized in that: In step 7, CATIA software is imported, and a digital editor module is used to import a manufacturing error function set; a fast surface reconstruction module is used to fit the error function set to complete the surface modeling function; a general surface modeling module is used to stretch the surface according to the corresponding thickness of the component to realize the construction of a three-dimensional solid model.

Images