High-precision calibration method for measurement control field in long time domain assembly process
By using laser network measurement and particle swarm algorithms in aerospace assembly site, the parameters between the nominal reference and the measured reference are iteratively corrected, which solves the problem of poor accuracy and reliability of measurement control field and realizes high-precision assembly data transmission.
Patent Information
- Application Number
- CN202411199793.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-29
AI Technical Summary
In the aerospace assembly site, the measurement and control field has poor accuracy and reliability due to factors such as foundation settlement, temperature difference and large load, resulting in a decrease in the accuracy of quality data transmission during assembly. The existing technology has failed to effectively solve the conversion between nominal benchmarks and measured benchmarks. Problem of inaccurate parameters.
During the long-term assembly process, the actual coordinates of ERS points are obtained through laser network measurement, and a movable directional point is attached. Based on the least squares principle and particle swarm algorithm, the rotation matrix and translation between the nominal reference and the measured reference are iteratively corrected. Vector, set the coordinate deviation threshold, build a conversion parameter optimization solution model, and improve the accuracy calibration of the measurement control field.
It realizes high-precision calibration of the measurement control field, reduces the unified reference error, ensures high-precision transmission of quality data during assembly, and has good versatility and application prospects.
Smart Images

Figure CN119087907A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of digital measurement and assembly, and relates to a high-precision calibration method for measuring a control field in a long-time domain assembly process. Background Art
[0002] Digital measurement-driven in-situ assembly has become one of the mainstream manufacturing modes for large aerospace components, in which the precise measurement and transmission of geometric information is the core of high-precision assembly of aerospace equipment. At present, the observation data of each local measurement system in the aerospace assembly site, the posture data of the execution tooling / equipment, etc., all need to use the "measurement control field" as a bridge to transmit the above data to the global coordinate system. Therefore, the accuracy and reliability of the measurement control field directly affect the assembly quality of the final product. However, the ground of the aerospace assembly workshop is prone to non-uniform deformation due to factors such as load and natural settlement of the foundation, which leads to the inaccuracy of the original calibration information in the measurement control field, and the temperature difference during the long-term cross-seasonal assembly process causes the ground to expand and contract. The above factors jointly cause the control points (also known as ERS points) in the measurement control field to shift significantly, and the global information (nominal coordinate value) they carry becomes invalid. At this time, if the nominal coordinates of the control points in the measurement control field are still used to solve the conversion parameters between different benchmarks, it is easy to cause a large benchmark unification error, resulting in a serious decrease in the accuracy of quality data transmission during the assembly process. In summary, in order to solve the problem of poor accuracy and reliability of the measurement control field due to factors such as foundation settlement, temperature difference, and large load, a high-precision calibration method for the measurement control field in the long-time domain assembly process is proposed to ensure the accurate transmission of quality data during the assembly process.
[0003] The patent "A method for point calibration in an aircraft assembly measurement field" with patent number CN20201 0353894.8 discloses a method for point calibration in an aircraft assembly measurement field. The method first collects data of orientation points and measuring station points through a laser tracker, and then calculates the coordinates of ERS points and TB points through an adjustment algorithm based on the coordinates of the orientation points and the measuring station points, and converts them to the coordinate system of the measurement field to complete the point calibration; the patent "A method, device, equipment and medium for automatic calibration of a measurement reference network" with patent number CN202410080738.7, the invention obtains the coordinates of ERS points and uses a comprehensive error assessment method for transfer to transfer the laser tracker; the unified spatial measurement network technology is used to convert the coordinate values of ERS points into theoretical coordinates of ERS points in the measurement reference network.
[0004] The above method only considers re-measurement of the ERS point and TB point coordinates of the assembly site to complete the point calibration in the assembly measurement field, and does not correct the ERS point coordinates under the nominal reference. The conversion parameters of the nominal reference and the measured reference obtained are still inaccurate, and there is a large reference unification error. Therefore, the present invention proposes a high-precision calibration method for the measurement control field of a long-term aviation assembly site. When obtaining the measured coordinates of the ERS points of the measurement control field, a movable orientation point is added to form an orientation constraint, and then the nominal coordinates of the ERS points are iteratively corrected and the rotation matrix and translation vector between the nominal reference and the measured reference are solved, and all ERS points in the measurement control field are globally assigned again, so as to realize the calibration of the measurement control field accuracy in the long-term aviation assembly process, reduce the reference unification error caused by factors such as foundation settlement, and ensure the high-precision transmission of quality data in the assembly process, which has good versatility and broad application prospects. Summary of the invention
[0005] The purpose of the present invention is to provide a high-precision calibration method for measuring a control field in a long-time domain assembly process.
[0006] The technical solution of the present invention:
[0007] A high-precision calibration method for a measurement control field in a long-time domain assembly process is disclosed. First, during the long-time domain assembly process of a component, laser network measurement is regularly used to obtain the measured coordinates of ERS points in the measurement control field and the measured coordinates of additional orientation points before and after movement. Secondly, based on the least squares principle, the conversion parameters between the nominal datum and the measured datum of the measurement control field are calculated by measuring the current measured coordinates of a series of ERS points with the same name in the measurement control field and the nominal coordinates of the original calibration, and the coordinates of the ERS points and the orientation points under the measured datum are aligned to the nominal datum to obtain the coordinate deviations of each ERS point under the nominal datum. Then, in combination with the actual process requirements of the assembly site, a coordinate deviation threshold is set, and an orientation strength constraint based on the additional orientation point is constructed, and the constraint is introduced into a solution model with the minimum fitting residual of the ERS points, thereby forming a new conversion parameter optimization solution model with the minimum sum of the coordinate deviations of each ERS point as the goal, thereby improving the conversion parameter solution accuracy of the nominal datum and the measured datum of the measurement control field. Finally, based on the above conversion parameters, all ERS points in the measurement control field are globally assigned again to realize the calibration of the measurement control field accuracy in the long-time domain assembly process, thereby ensuring the high-precision transmission of quality data. The specific steps are as follows:
[0008] The first step is to measure the ERS points and directional points in the control field.
[0009] First, a movable orientation point is added to the measurement control field at the assembly site, and multiple laser trackers are placed in the measurement control field. The measured coordinate values of the ERS points in the measurement control field are obtained through laser networking measurement, and the data set Data is constructed, as shown in formula (1):
[0010]
[0011] In the formula, is the measured coordinate of the i-th ERS point under the measured benchmark, i = 1, 2, … m, m is the number of ERS points in the measurement control field;
[0012] Secondly, obtain the initial measured coordinates q of the additional movable orientation point l =(x l ,y l ,z l ), then control the additional movable orientation point to move in a directional manner, and obtain the measured coordinates of the additional orientation point
[0013] The second step is to measure the coordinate deviation of the ERS point under the nominal reference of the control field.
[0014] set up is the nominal coordinate of the i-th ERS point under the known nominal datum. Based on the least squares principle, the nominal coordinates of the ERS point in the nominal datum and the measured datum are Measured coordinates Solve the transformation parameters between the above two benchmarks, namely the rotation matrix and the translation vector, as shown in equation (2);
[0015]
[0016] In the formula, R L-G and T L-G is the rotation matrix and translation vector for converting the measured datum to the nominal datum, as shown in (3);
[0017]
[0018] Where γ, β, and α represent the rotation angle between the measured reference and the nominal reference respectively; c and s represent the cosine function and the sine function respectively;
[0019] The measured coordinates of the ERS point in the measured benchmark Additional movable orientation point starting point measured coordinates q l And the measured coordinates of the endpoint of the additional movable orientation point Convert to nominal basis, as shown in formula (4);
[0020]
[0021] In the formula, is the measured coordinate of the ith ERS point in the control field under the nominal datum, The starting point coordinates of the additional movable orientation point under the nominal reference. is the end point coordinate of the additional movable orientation point under the nominal datum; then, the deviation between the measured coordinates and the nominal coordinates of the series of ERS points with the same name under the nominal datum and the displacement before and after the orientation point is moved are calculated by formula (5);
[0022]
[0023] In the formula, δ i,0 is the deviation between the measured coordinates and the nominal coordinates of the ith ERS point under the nominal reference, so δ i,0 =(Δx i,0 ,Δy i,0 , Δz i,0 ) T , ε is the displacement of the orientation point before and after the movement under the nominal datum;
[0024] The third step is to solve the transformation parameters based on the iterative correction of the nominal coordinates
[0025] Firstly, a transformation parameter optimization solution model with the goal of minimizing the sum of the coordinate deviations of each ERS point is established. The parameters to be optimized are the nominal coordinate deviation value of the ERS point and the transformation parameter between the nominal datum and the measured datum. Considering the actual measurement process requirements in the assembly process, the nominal coordinate deviation threshold of the ERS point is set to avoid overfitting. The objective function is shown in formula (6), and the constraint function is shown in formula (7):
[0026]
[0027] In the formula, is the conversion parameter between the nominal reference and the measured reference, δ i,j is the nominal coordinate deviation value obtained by the jth optimization of the i-th ERS point, j = 1, 2, ..., n, n is the number of optimizations, δ i,j =(Δx i,j ,Δy i,j , Δz i,j ) T , a and b are the upper and lower limits of the coordinate deviation threshold respectively; the nominal coordinates of the ERS point under the nominal datum Measured coordinates The particle swarm algorithm is used to optimize and solve the coordinate deviation of the ERS point. The fitness function in the algorithm is the objective function. The nominal coordinate value of the ERS point is iteratively corrected, and finally the conversion parameters between the nominal benchmark and the measured benchmark are obtained. And the nominal coordinate deviation value δ of the ERS point obtained by the final nth optimizationi,n , as shown in formula (8);
[0028] δ=[δ 1,n ,δ 2,n ,...,δ i,n ,...,δ m,n ] (8)
[0029] Secondly, based on the conversion parameters, a laser tracker is used to obtain the measured coordinates of a new set of ERS points in the measurement control field. Where i = 1, 2, ..., k, k is the number of ERS points in the measurement control field obtained this time, and then based on the conversion parameters obtained by the optimization function, the measured coordinates of the ERS points under the measured benchmark are converted to the nominal benchmark, as shown in (9),
[0030]
[0031] In the formula, That is, the coordinate value of the i-th ERS point converted to the nominal datum based on the above conversion parameters;
[0032] Then, calculate the measured coordinates of the group of ERS points converted to the nominal datum Nominal coordinates under nominal datum The coordinate deviation E i , as shown in (10);
[0033]
[0034] Beneficial effects of the present invention: The present invention can accurately calculate the coordinate deviation of the ERS point in the measurement control field under the nominal reference, thereby establishing a conversion parameter optimization solution model with the goal of minimizing the sum of the coordinate deviations of each ERS point, thereby improving the conversion parameter solution accuracy of the nominal reference and the measured reference of the measurement control field, and finally based on the above conversion parameters, re-assigning global values to all ERS points in the measurement control field, realizing the calibration of the measurement control field accuracy in the long-time domain assembly process, and ensuring the high-precision transmission of quality data; the method not only has good practicality and versatility, but also has good robustness and high precision, and has broad application prospects; BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 It is a flow chart of a high-precision calibration method for measuring control fields during long-time-domain assembly;
[0036] Figure 2 It is a schematic diagram of the measurement control field during the docking and assembly process of the aircraft nose and the front fuselage;
[0037] In the figure: 1- aircraft nose; 2- aircraft front fuselage; 3- movable additional orientation point; ERS1 ~ ERS 14ERS points are obtained; S1-laser tracker station 1; S2-laser tracker station 2; S3-laser tracker station 3. DETAILED DESCRIPTION
[0038] The specific implementation of the present invention is described in detail below in conjunction with the technical scheme and the accompanying drawings.
[0039] The maximum permissible measurement error MPE of the laser tracker selected in this embodiment is ±(15 μm+6 μm / m); 15 1.5-inch laser tracker target balls are selected as measurement control field ERS points and additional orientation points; and measurement experiments are carried out within a range of 9m×6m.
[0040] like Figure 1 As shown, the method first obtains the measured coordinates of the ERS points named p1 to p7 and the additional orientation points before and after the movement in the measurement control field through a laser tracker at the assembly site of the aviation components; secondly, based on the least squares principle, the measured coordinates and nominal coordinates of a series of ERS points with the same name in the measurement control field are used to calculate the conversion parameters between the measured datum and the nominal datum, and the measured coordinates of the ERS points and the measured coordinates of the orientation points are all aligned to the nominal datum; then, a conversion parameter optimization solution model with the goal of minimizing the sum of the coordinate deviations of each ERS point is established, and the nominal coordinates of the ERS points are iteratively corrected, thereby improving the conversion parameter solution accuracy of the nominal datum and the measured datum of the measurement control field; finally, based on the above conversion parameters, all ERS points in the measurement control field are globally assigned again to realize the calibration of the measurement control field accuracy during the long-time domain assembly process. The specific steps of this method are as follows:
[0041] The first step is to measure the ERS points and directional points in the control field.
[0042] like Figure 2 As shown in the figure, firstly, additional orientation points are placed in the 9m×6m measurement control field, and multiple laser trackers are placed at the same time. The initial measured coordinates q of the additional orientation points are obtained by laser network measurement. l ; Then change the orientation point position and obtain the measured coordinates of the orientation point again At the same time, obtain the measured coordinate values of the ERS points in the measurement control field As shown in Table 1;
[0043] Table 1 ERS point coordinate data set
[0044]
[0045] The second step is to solve the coordinate deviation of the ERS point under the nominal datum.
[0046] Based on the least squares principle, the nominal coordinate values and measured coordinate values of ERS in the measurement control field are substituted into formula (2) to obtain the transfer station parameter rotation matrix RL-G and the translation vector T L-G ;
[0047]
[0048] Then, based on formula (4), the coordinate values of all ERS points and the coordinates of the orientation points measured under the measured benchmark are transformed to the nominal benchmark through the rotation matrix and translation vector, and the coordinate deviation of each ERS point after benchmark unification is calculated according to formula (5), as shown in Table 2;
[0049] Table 2 ERS point coordinate deviation
[0050]
[0051]
[0052] And calculate the coordinate displacement before and after the additional orientation point moves under the nominal benchmark, as shown in Table 3:
[0053] Table 3 Orientation point coordinate deviation
[0054]
[0055] The third step is to solve the transformation parameters based on the iterative correction of the nominal coordinates
[0056] Then, the objective function is to minimize the sum of the residuals of the measured coordinates and the nominal coordinates of the ERS points in the control field. Based on the particle swarm algorithm, the objective function is used as the fitness function, and the parameters such as the maximum number of iterations, coordinate deviation range and number of particles are set. The nominal coordinates of the ERS points under the nominal benchmark are Measured coordinates and the data of the coordinate displacement ε of the orientation point, the nominal coordinate deviation of the ERS point is optimized, as shown in Table 4;
[0057] Table 4 Nominal coordinate deviation of ERS points after optimization
[0058]
[0059] It can be seen from Tables 2 and 4 that the average registration error before correction in the X direction is 0.232 mm, and the average registration error after correction is 0.004 mm; the average registration error before correction in the Y direction is 0.347 mm, and the average registration error after correction is 0.006 mm; the average registration error before correction in the Z direction is 0.247 mm, and the average registration error after correction is 0.006 mm;
[0060] Based on the iteratively corrected nominal value coordinates, the rotation matrix between the nominal datum and the measured datum is solved. and translation vector
[0061]
[0062] The laser tracker is used to obtain the midpoint of the measurement control field named p8–p 14 The measured coordinates of the ERS points are re-assigned globally based on the rotation matrix and translation vector calculated in the above steps, and the coordinate deviation between the measured coordinates and the nominal coordinates under the nominal reference is calculated, as shown in Table 5;
[0063] Table 5 Coordinate deviation of the second group of ERS points
[0064]
[0065] The invention discloses a high-precision calibration method for a measurement control field in a long-time domain assembly process. First, in the long-time domain assembly process of a component, laser networking measurement is regularly used to obtain the measured coordinates of ERS points in the measurement control field and the measured coordinates of additional directional points before and after movement. Secondly, based on the least squares principle, the conversion parameters between the nominal datum and the measured datum of the measurement control field are calculated by measuring the current measured coordinates of a series of ERS points with the same name in the measurement control field and the nominal coordinates of the original calibration, and the coordinates of the ERS points and the directional points under the measured datum are aligned to the nominal datum to obtain the coordinate deviations of each ERS point under the nominal datum. Then, in combination with the actual process requirements of the assembly site, a coordinate deviation threshold is set, and a directional strength constraint based on the additional directional points is constructed, and the constraint is introduced into a solution model with the minimum fitting residual of the ERS points, thereby forming a new conversion parameter optimization solution model with the minimum sum of the coordinate deviations of each ERS point as the goal, thereby improving the conversion parameter solution accuracy of the nominal datum and the measured datum of the measurement control field. Finally, based on the above conversion parameters, all ERS points in the measurement control field are globally assigned again to realize the calibration of the accuracy of the measurement control field in the long-time domain assembly process, thereby ensuring the high-precision transmission of quality data.
Claims
1. A high-precision calibration method for measuring control fields in a long-time domain assembly process, characterized in that: The specific steps are as follows: The first step is to measure the ERS points and directional points in the control field. First, a movable orientation point is added to the measurement control field at the assembly site, and multiple laser trackers are placed in the measurement control field. The measured coordinate values of the ERS points in the measurement control field are obtained through laser networking measurement, and the data set Data is constructed, as shown in formula (1): In the formula, is the measured coordinate of the i-th ERS point under the measured benchmark, i = 1, 2, … m, m is the number of ERS points in the measurement control field; Secondly, obtain the initial measured coordinates q of the additional movable orientation point l =(x l ,y l ,z l ), then control the additional movable orientation point to move in a directional manner, and obtain the measured coordinates of the additional orientation point The second step is to measure the coordinate deviation of the ERS point under the nominal reference of the control field. set up is the nominal coordinate of the i-th ERS point under the known nominal datum. Based on the least squares principle, the nominal coordinates of the ERS point in the nominal datum and the measured datum are Measured coordinates Solve the transformation parameters between the above two benchmarks, namely the rotation matrix and the translation vector, as shown in equation (2); In the formula, R L-G and T L-G is the rotation matrix and translation vector for converting the measured datum to the nominal datum, as shown in (3); Where γ, β, and α represent the rotation angle between the measured reference and the nominal reference respectively; c and s represent the cosine function and the sine function respectively; The measured coordinates of the ERS point in the measured benchmark Additional movable orientation point starting point measured coordinates q l And the measured coordinates of the endpoint of the additional movable orientation point Convert to nominal basis, as shown in formula (4); In the formula, is the measured coordinate of the ith ERS point in the control field under the nominal datum, The starting point coordinates of the additional movable orientation point under the nominal reference. is the end point coordinate of the additional movable orientation point under the nominal datum; then, the deviation between the measured coordinates and the nominal coordinates of the series of ERS points with the same name under the nominal datum and the displacement before and after the orientation point is moved are calculated by formula (5); In the formula, δ i,0 is the deviation between the measured coordinates and the nominal coordinates of the ith ERS point under the nominal reference, so δ i,0 =(Δx i,0 ,Δy i,0 , Δz i,0 ) T , ε is the displacement of the orientation point before and after the movement under the nominal datum; The third step is to solve the transformation parameters based on the iterative correction of the nominal coordinates Firstly, a transformation parameter optimization solution model with the goal of minimizing the sum of the coordinate deviations of each ERS point is established. The parameters to be optimized are the nominal coordinate deviation value of the ERS point and the transformation parameter between the nominal datum and the measured datum. Considering the actual measurement process requirements in the assembly process, the nominal coordinate deviation threshold of the ERS point is set to avoid overfitting. The objective function is shown in formula (6), and the constraint function is shown in formula (7): s.t. a≤δ i,j ≤b,i=1,2,...,m (7) In the formula, is the conversion parameter between the nominal reference and the measured reference, δ i,j is the nominal coordinate deviation value obtained by the jth optimization of the i-th ERS point, j = 1, 2, ..., n, n is the number of optimizations, δ i,j =(Δx i,j ,Δy ij , Δz ij ,) T , a and b are the upper and lower limits of the coordinate deviation threshold respectively; the nominal coordinates of the ERS point under the nominal datum Measured coordinates The particle swarm algorithm is used to optimize and solve the coordinate deviation of the ERS point. The fitness function in the algorithm is the objective function. The nominal coordinate value of the ERS point is iteratively corrected, and finally the conversion parameters between the nominal benchmark and the measured benchmark are obtained. And the nominal coordinate deviation value δ of the ERS point obtained by the final nth optimization i,n , as shown in formula (8); δ=[δ 1,n ,d 2,n ,...,d i,n ,...,d m,n ] (8) Secondly, based on the conversion parameters, a laser tracker is used to obtain the measured coordinates of a new set of ERS points in the measurement control field. Where i = 1, 2, ..., k, k is the number of ERS points in the measurement control field obtained this time, and then based on the conversion parameters obtained by the optimization function, the measured coordinates of the ERS points under the measured benchmark are converted to the nominal benchmark, as shown in (9), In the formula, That is, the coordinate value of the i-th ERS point converted to the nominal datum based on the above conversion parameters; Then, calculate the measured coordinates of the group of ERS points converted to the nominal datum Nominal coordinates under nominal datum The coordinate deviation E i , as shown in (10);
Citation Information
Patent Citations
Automatic calibration method, device and equipment for measuring reference network and medium
CN117989978A
Method for increasing coordinate measurement field accuracy through space multi-length constraint
CN104315983A
Oppositely-closing attitude-adjusting method based on multi-laser tracker measuring field
CN109613519A
Gradient evaluation method based on maximum entity requirement (MMR) hole
CN110348118A
Coordinate conversion method based on multi-dimensional dynamic etalon
CN110487182A
