CN112307562B - Method for assembling complex parts on large-scale airplane by combining thermal deformation and gravity deformation - Google Patents

Abstract

The invention relates to an assembly method of complex parts on a large airplane with comprehensive thermal deformation and gravity deformation, belonging to the technical field of large airplane manufacturing. The method comprises the following steps of posture adjusting and positioning: (1) dispersing the complex part and key features on the three-dimensional model of the complex part into a key feature point set; (2) obtaining coordinate values of key feature points on the two key feature points; (3) calculating the coordinate deviation of the key characteristic points caused by the temperature and the gravity action by using a finite element analysis model based on the temperature of the assembly environment; (4) correcting the coordinate values of key feature points on the complex part based on the obtained coordinate deviation; (5) and calculating a pose coordination matrix T for adjusting and positioning the complex part to the target pose based on the corrected coordinate values of the key feature points on the complex part and the coordinate values of the key feature points on the three-dimensional model. The method can effectively improve the assembly efficiency and precision of the complex parts on the large-scale airplane, and can be widely applied to the technical field of manufacturing of large-scale airplanes.

Description

Assembly method of complex parts on large aircraft with integrated thermal deformation and gravity deformation

technical field

The invention relates to the technical field of assembly of large aircraft, in particular to a method for assembling complex components on a large aircraft, which integrates thermal deformation and gravitational deformation in the assembly process.

Background technique

In the assembly process of large aircraft, especially the assembly process of complex structural components on it, due to its large size, it needs to be repeatedly adjusted and repaired during the installation process to meet the installation accuracy requirements. Aiming at this technical problem, the patent document with the publication number CN107263044A discloses a design method of a large aircraft outer wing box assembly system considering thermal deformation factors, and the patent document with the publication number CN107052750A discloses a leading edge Attitude adjustment and positioning system for components; in these existing technologies, based on the improvement of the installation tooling structure, in order to solve the problem of thermal deformation coordination between the large-sized components on the aircraft and their positioning tooling during the installation process, such as the wing Positioning and installation of large-scale components such as box leading edge components, trailing edge components, and wing root ribs.

In the above scheme, based on the improvement of the tooling structure, although the problem of thermal deformation coordination between the components and the tooling can be solved, there are still measured coordinate values produced by thermal deformation and gravity deformation during the on-site installation of large aircraft. The problem of deviation, especially the process of attitude adjustment and positioning of a single part or the first part during posture coordination (that is, the reference part that the subsequent part installation process refers to), plus the single part that needs attitude adjustment and positioning on large aircraft Or the reference component is usually a complex structural component, and thermal deformation and gravity deformation will have a serious impact on its installation accuracy and efficiency. The process of adjusting and positioning components usually requires multiple adjustments and repairs.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to provide a method for assembling complex components on large aircraft that integrates thermal deformation and gravitational deformation, so as to improve the installation accuracy and assembly efficiency of complex structural components on large aircraft.

In order to achieve the above-mentioned main purpose, the method for assembling complex components on a large aircraft that integrates thermal deformation and gravitational deformation provided by the present invention includes an attitude adjustment and positioning step and an installation and fixing step, and the attitude adjustment and positioning step includes the following steps:

The discrete processing step is to discretize the complex components and their key features on the 3D model into key feature point sets;

The coordinate measurement step is to obtain the coordinate values of the complex components and the key feature points on the three-dimensional model after the pose adjustment and positioning to the target pose;

In the simulation solution step, based on the 3D model and the current assembly ambient temperature, the finite element analysis model is used to calculate the coordinate deviation of the key feature points caused by the assembly ambient temperature and the action of gravity;

The parameter correction step, based on the coordinate deviation obtained in the simulation solution step, corrects the coordinate value of the key feature point measured from the complex component;

The calculation step is to calculate the pose coordination matrix T for adjusting the pose of the complex part to the target pose based on the corrected coordinate values of the key feature points on the complex part and the coordinate values of the key feature points on the three-dimensional model.

Based on the above technical solution, the coordinate deviation caused by temperature and gravity is calculated by simulation, and the coordinates of key feature points on complex components are corrected by using the coordinate deviation, that is, the influence caused by gravity and temperature is excluded from the existing coordinate measurement values. Deviation, that is, using the coordinate deviation after roughly eliminating the influence of gravity and temperature to solve the pose coordination matrix T, the obtained pose coordination matrix T can better match the target pose in the 3D model, thereby effectively eliminating thermal deformation. The nonlinearity between the coordinate measurements of key feature points and the three-dimensional model caused by gravity deformation is helpful to achieve accurate and efficient attitude adjustment and positioning of complex structures of large aircraft.

A specific solution is that the above step of calculating the pose coordination matrix T used for adjusting the pose of the complex component to the target pose includes the following steps:

(1) Based on the principle of three-dimensional point matching, the least squares method is used to construct the pose coordination optimization model J of the feature point pairs of complex parts and pipe fittings on the three-dimensional model,

Among them, n is the number of discretized key feature point pairs, R and P are the rotation and translation components of the pose coordination matrix T correspondingly, and KCs ^measued is the key feature point on the complex component after correction in the assembly coordinate system The coordinate value of , KCs ^datum is the coordinate value of the key feature point coordinates on the 3D model in the assembly coordinate system;

(2) Integrate the linear SVD and nonlinear L-M algorithms to solve the pose coordination optimization model J, and obtain the optimal pose coordination matrix T.

Based on the principle of three-dimensional point matching, the least squares method is used to construct the pose coordination optimization model J of the feature point pairs of complex parts and pipe fittings on the three-dimensional model, which can effectively eliminate the influence of factors such as assembly deviation and measurement uncertainty existing on complex parts. , to further improve assembly accuracy and efficiency.

A more specific solution is to first use the linear SVD algorithm to solve the pose coordination optimization model J in the function solving step, obtain the optimal pose coordination matrix T, and estimate the estimated value of the pose coordination matrix T; The estimated value is the initial value, and the linear SVD and nonlinear L-M algorithms are combined to solve the pose coordination optimization model J. Using linear SVD to solve the initial value first, it can effectively speed up the convergence in the exact solution process.

The preferred solution is to assign weights to different key feature point pairs according to the preset importance, and obtain the corrected pose coordination optimization model J:

以六元组S与坐标值KCs^measued为参数的函数，ω_i为第i个关键特征点对的权重。Among them, the six-tuple S=[X, Y, Z, A, B, C], X, Y, Z are the parameters of the translation component P, A, B, C are the rotation components R represented by the ZYX Euler angle parameters parameter;

It is a function with the six-tuple S and the coordinate value KCs ^measued as parameters, and ω _i is the weight of the i-th key feature point pair.

By assigning different weights to different key feature points, that is, assigning higher weights to some important positions, such as the hanging position of the transmitter, so that the calculation structure is more in line with the actual assembly situation, and the assembly efficiency and accuracy are further improved.

The preferred solution is that in the parameter correction step, the coordinate value ^KCsmeasured of the key feature point on the revised complex component is calculated based on the following formula;

为仿真获得由装配环境温度所造成的坐标值偏差，

为仿真获得由装重力所造成的坐标值偏差。Among them, KCs ₀ ^measured is the direct measurement value of the coordinates of key feature points on complex parts,

For the simulation to obtain the coordinate value deviation caused by the assembly ambient temperature,

The deviation of the coordinate values caused by the loaded gravity is obtained for the simulation.

The preferred solution is that the complex component is a wall plate, and the adjustment is performed based on a plurality of numerically controlled attitude adjustment and positioning devices arranged at predetermined intervals along the course of the wall plate. The supporting ball head constitutes a ball head locking mechanism of the ball head hinge.

A further solution is that the attitude adjustment and positioning step includes the following steps: based on the formula JPs ^desired =T*JPs ^current , calculate the spherical center coordinates JPs ^desired , JPs of the ball head supported on each numerical control adjustment positioning device when the aircraft structure is adjusted to the target attitude and attitude ^current is the measured value of the spherical center coordinates of the support ball head on each numerical control adjustment positioning device in the assembly coordinate system.

A further solution is that if the number of support balls coupled to the same numerical control attitude adjustment and positioning device is more than two, then the connection between the center of the fixed position of the two or more support balls and the number of the extension lines. Each point is used as a key feature point; if the number of support ball joints coupled to the same numerical control attitude adjustment and positioning device is more than one, at least the center point of the fixed position of the support ball head is taken as a key feature point. On the premise of ensuring the accuracy of the calculation structure, the solution process should be simplified as much as possible.

The preferred solution is to arrange key feature points at installation positions with relatively high precision requirements relative to the surrounding area according to the assembly process specification.

Description of drawings

Fig. 1 is the working flow chart of the attitude adjustment and positioning step in the embodiment of the present invention;

FIG. 2 is a schematic diagram of the positions of key feature points on a complex part of an aircraft according to an embodiment of the present invention.

Detailed ways

The present invention will be further described below with reference to the embodiments and the accompanying drawings.

In the following embodiments, the process of attitude adjustment and positioning during assembly of the wall panel of a large aircraft is used as an example for demonstration. This assembly method is also applicable to the assembly of other reference components that need to be installed on a large aircraft, such as The first fuselage segment in the fuselage docking assembly for attitude adjustment and positioning.

Example

In this embodiment, the assembling method of the present invention is used to assemble the large aircraft panel 01 as shown in FIG. 1 , wherein the device for adjusting the attitude and positioning of the panel 01 during the assembly process adopts the application of the applicant and the The numerical control attitude adjustment and positioning device disclosed in the patent document with the publication number of CN107471171A is used for adjustment, specifically by using a plurality of numerical control attitude adjustment and positioning devices arranged at predetermined intervals along the span direction of the wall panel 01 for adjustment; refer to the appendix of the patent document. As shown in the specific structure disclosed in the figure, the attitude adjustment and positioning device specifically includes a base, a slide column slidably installed on the base through the guide rail slider mechanism, and a set of output ends installed on the slide column in parallel with each other. The above three-coordinate numerical control positioner; the output end is fixed with a ball head locking device for forming a ball head hinge mechanism with a support ball head arranged on the wall plate.

The assembling method of the present invention includes an attitude adjustment and positioning step and an installation and fixing step. As shown in FIG. 2, the attitude adjustment and positioning step includes a discrete processing step S1, a coordinate measurement step S2, a simulation solution step S3, a parameter correction step S4, a calculation step S5 and Step S6 is executed, and the specific process is as follows:

The discretization processing step S1 is to discretize the complex components and the key features on the three-dimensional model into key feature point sets.

In this embodiment, for the process of selecting key feature points on the wall panel 01 of the complex component, the following rules are mainly referred to:

(1) If the number of support ball joints coupled with the same numerical control attitude adjustment positioning device is two or more, the center line of the fixed position of the two or more support ball joints and the extension of the line Multiple points on the line are used as key feature points; for example, in the patent document with the publication number CN107471171A, the three numerically controlled attitude adjustment and positioning devices located at the root end of the wing have two supporting balls fixedly connected to the wall plate 01, then the The central connection line of the fixed connection positions of the two support ball heads and the wall plate, and a plurality of points on the extension line of the connection line are used as key feature points, and are specifically selected in a uniform arrangement.

(2) If the number of support ball joints coupled to the same numerical control attitude adjustment and positioning device is more than one, at least the center point of the fixed position of the support ball head is used as a key feature point. For example, in the patent document with the publication number of CN107471171A, both the two numerically controlled attitude adjustment and positioning devices located at the tip of the wing have only one support ball head and the wall plate 01 fixedly connected, so at least the connection center of the support ball head and the wall plate needs to be The point is used as the key feature point, and the direction of the connection line with the three key feature points at the wing root can also be parallel to the direction of the contour line of the wall plate holding tool, and multiple points on the straight line passing through the connection center point are the key feature points. , which is selected in a uniform arrangement.

(3) According to the assembly process specification, compared with other positions, key feature points should also be set for positions with high precision requirements such as thermal deformation anchor points and engine installation positions.

As shown in FIG. 1 , in this embodiment, a total of 5 columns are selected, with a total of 16 key feature points 1 .

The coordinate measurement step S2 is to obtain the coordinate values of the complex components and the key feature points on the three-dimensional model after the pose adjustment and positioning to the target pose.

For the acquisition of the coordinate values of the key feature points on the wall panel 01 of the complex component, the coordinate value KCs ₀ ^measured is obtained in order to use the laser tracking measurement system for measurement, and the three-dimensional model as the attitude adjustment reference is obtained based on the design software, and the coordinates are obtained. Value KCs ^datum .

In this embodiment, the coordinate value of the key feature point on the wall panel 01 obtained by measurement is the measured value based on the local coordinate system of the component. For the convenience of subsequent calculations, it needs to be converted to the measured value in the assembly coordinates. Specifically, using Formula KCs ^datum = ^W T _L *KCs _L ^datum for conversion, where KCs _L ^datum is the coordinate value of the structural panel of the aircraft to be adjusted in the local coordinate system, KCs ^datum is the coordinate value in the assembly coordinate system, ^W T _L It is the mathematical description of the local coordinate system of the aircraft structure panel in the assembly coordinate system.

In the simulation solution step S3, based on the three-dimensional model and the current assembly environment temperature, the finite element analysis model is used to calculate the coordinate deviation of the key feature points caused by the assembly environment temperature and the action of gravity.

Wherein, the temperature of the assembly environment is acquired by temperature sensors arranged at multiple points in the assembly workshop, for example, a plurality of temperature sensors arranged at intervals along the spanwise direction of the panel are used for measurement.

In the parameter correction step S4, the coordinate values of the key feature points measured from the complex components are corrected based on the coordinate deviation obtained in the simulation solving step.

In this step, the coordinate value ^KCsmeasured of the key feature point on the revised complex component is calculated based on the following formula;

为仿真获得由装配环境温度所造成的坐标值偏差，

为仿真获得由重力所造成的坐标值偏差，即用于计算的坐标值为无重力与温度影响情况的坐标测量值，即消除了相应的影响。Among them, KCs ₀ ^measured is the direct measurement value of the coordinates of key feature points on complex parts,

For the simulation to obtain the coordinate value deviation caused by the assembly ambient temperature,

In order to obtain the coordinate value deviation caused by gravity for simulation, that is, the coordinate value used for calculation is the coordinate measurement value without the influence of gravity and temperature, that is, the corresponding influence is eliminated.

Calculating step S5, based on the coordinate values of the key feature points on the revised complex component and the coordinate values of the key feature points on the three-dimensional model, calculate the pose coordination matrix T for adjusting the pose of the complex component to the target pose . Specifically include the following steps:

(1) Based on the principle of three-dimensional point matching (Arun K S. Least-squares fitting of two 3-Dpoint sets [J]. IEEE Trans. pattern Anal. machine Intell, 1987, 9.), the least squares method is used to construct complex The pose coordination optimization model J of the key feature point pairs on the part and the 3D model:

Among them, n is the number of discretized key feature point pairs, R and P are the rotation and translation components of the pose coordination matrix T correspondingly, and KCs ^measued is the key feature point on the complex component after correction in the assembly coordinate system The coordinate value of , KCs ^datum is the coordinate value of the key feature point coordinates on the 3D model in the assembly coordinate system.

In this step, after considering the factors such as assembly deviation and measurement uncertainty of the complex component panels to be adjusted, the pose coordination matrix is not exact and consistent for all key feature point pairs, and the coordination optimization model is based on the coordination optimization model. J, and try to eliminate factors such as assembly deviation and measurement uncertainty.

(2) Give weights to different key feature point pairs according to the preset importance, and obtain the corrected pose coordination optimization model J:

以六元组S与坐标值KCs^measued为参数的函数，用于表征R*KCs^measued+P，ω_i为第i个关键特征点对的权重。Among them, the six-tuple S=[X, Y, Z, A, B, C], X, Y, Z are the parameters of the translation component P, A, B, C are the rotation components R represented by the ZYX Euler angle parameters parameter;

The function with the hexagram S and the coordinate value KCs ^measued as parameters is used to characterize R*KCs ^measued + P, and ω _i is the weight of the i-th key feature point pair.

The weight of each key feature point pair is manually assigned according to the preset importance level to correct the pose coordination optimization model. For example, according to the assembly process specification, relative to other positions, thermal deformation anchor points, engine installation positions and other positions The accuracy requirements are higher, that is, the accuracy requirements for different regions are different, so that in the assignment process, the weight assignment at the position with the higher accuracy requirement is larger.

(3) Integrate the linear SVD and nonlinear L-M algorithms, solve the pose coordination optimization model J, and obtain the optimal pose coordination matrix T.

The pose coordination matrix T is used to adjust the panel to be pose-adjusted from the current pose to the target pose, that is, the pose designed in the three-dimensional model, that is, KCs ^datum =T*KCs ^measued .

In the function solving step, the linear SVD algorithm is used to solve the pose coordination optimization model J first, the pose coordination matrix T is obtained, and the estimated value of the pose coordination matrix T is estimated; then the estimated value is used as the initial value, Integrate linear SVD and nonlinear L-M algorithm to solve the pose coordination optimization model J. The specific process is as follows:

(1) The pose coordination parameter S ₀ =[X ₀ , Y ₀ , Z ₀ , A ₀ , B ₀ , C ₀ ] calculated based on the non-iterative SVD algorithm.

(2) The coordinates of the key feature points in the attitude adjustment datum and the center of mass of the measured values of the coordinates of the key feature points of the aircraft structure to be adjusted can be expressed as

and order

与

在笛卡尔空间具有相同的质心，即If the obtained pose coordination parameters R and P are the least squares solution, then the coordinates of the key feature points in the pose adjustment benchmark

and

have the same centroid in Cartesian space, i.e.

Based on the above representation, the pose coordination model J can be simplified as:

Solve this formula to get:

Therefore, minimizing the objective function J is equivalent to maximizing the function Q.

in,

SVD decomposition of Q:

Q=UDV ^T

Among them, D is a diagonal matrix, and U and V are standard orthogonal matrices. Based on this formula, the rotation component R can be calculated according to the following formula:

R= ^VUT

计算平移分量：and based on the formula

Compute the translation component:

The rigid body kinematic transformation matrix T composed of the rotation matrix R and the translation vector P can be represented as a six-tuple S.

(3) The pose coordination parameter S ₀ =[X ₀ , Y ₀ , Z ₀ , A ₀ , B ₀ , C ₀ ] calculated by the non-iterative SVD algorithm can be used as the nonlinear least squares Levenberg-Marquard (LM) algorithm The initial value of , solve the optimal pose coordination parameter S=[X, Y, Z, A, B, C] that minimizes the objective function J.

The L-M algorithm is a fusion of the gradient descent method and the Gauss-Newton (G-N) algorithm, which is more robust than the G-N algorithm. In the L-M algorithm, the modified Hessian matrix is used:

H(S,λ)=2J ^T J+λI

where J is the Jacobian matrix, I is the identity matrix, and λ is the damping factor, which is adjusted at each iteration. If λ is small, H approximates a G-N Hessian. Otherwise, H is close to the identity matrix, and the L-M algorithm degenerates into gradient descent.

The steps of the L-M algorithm can be briefly described as follows:

① Let λ be 0.001.

② Calculate δS=-H(S,λ) ^-1 g, where δS is the increment of the estimated pose coordination parameter hexagram S. In the GN algorithm, g=-2J ^T JδS, and in the gradient descent method, g=-λδS.

③ When f(S _n +δS)>f(S _n ), λ=10λ, then return to step ②.

④Otherwise, λ=0.1λ, Sn ₊₁ = _Sn +δS, then go to step ②.

Step S6 is executed, based on the obtained pose coordination matrix T, the pose adjustment and positioning device is controlled to adjust the pose of the wall panel.

In this embodiment, it is specifically calculated based on the formula JPs ^desired =T*JPs ^current , which is used to adjust the aircraft structure to the target posture when the spherical center coordinate JPs ^desired of the ball head supported on each numerical control adjustment positioning device is calculated, and JPs ^current is the assembly coordinate The measured value of the spherical center coordinates of the support ball head on each CNC attitude adjustment and positioning device is determined, and the obtained pose coordination matrix is given to the position and attitude adjustment of the support ball head on each CNC adjustment positioning device, thereby effectively simplifying the calculation process.

Claims (7)

1. The method for assembling the complex parts on the large-scale airplane integrating thermal deformation and gravity deformation comprises a posture adjusting and positioning step and an installation and fixing step, and is characterized in that the posture adjusting and positioning step comprises the following steps: a discrete processing step, namely, discretizing the complex part and key features on the three-dimensional model of the complex part into a key feature point set; coordinate measurement, namely acquiring coordinate values of the complex component and key feature points on the three-dimensional model to be adjusted and positioned to reach a target pose state; a simulation solving step, namely calculating the coordinate deviation of the key characteristic points caused by the assembly environment temperature and the gravity action by utilizing a finite element analysis model based on the three-dimensional model and the current assembly environment temperature; a parameter correction step of correcting the coordinate values of the key feature points measured from the complex component based on the coordinate deviation obtained in the simulation solving step; calculating a pose adjusting matrix T for adjusting and positioning the complex component to the target pose based on the coordinate values of the key feature points on the complex component and the coordinate values of the key feature points on the three-dimensional model after correction;

the step of calculating a pose adjustment matrix T for pose positioning of the complex component to the target pose comprises the steps of: (1) constructing a pose adjustment optimization model J of the complex component and key feature point pairs on the three-dimensional model by adopting a least square method based on a three-dimensional point matching principle,

wherein n is the number of the dispersed key feature point pairs, R and P are respectively the rotation component and the translation component of the pose adjustment matrix T, and KCs^measuedFor the coordinate values, KCs, of key feature points on the complex part under the assembly coordinate system and after correction^datumThe coordinate value of the key characteristic point coordinate on the three-dimensional model under the assembly coordinate system; (2) and fusing a linear SVD algorithm and a nonlinear L-M algorithm, solving the pose adjustment optimization model J, and acquiring an optimal pose adjustment matrix T.

2. The assembly method of claim 1, wherein: in the simulation solving step, firstly, solving the pose adjustment optimization model J by using a linear SVD algorithm to obtain an optimal pose adjustment matrix T, and estimating a pre-estimated value of the pose adjustment matrix T; and then, the estimated value is used as an initial value, a linear SVD algorithm and a nonlinear L-M algorithm are fused, and the pose adjustment optimization model J is solved.

3. The assembly method of claim 1, wherein: weighting different key feature point pairs according to preset importance, and acquiring a pose adjustment optimization model J1 after correction:

wherein, six-membered group S ═ X, Y, Z, A, B, C]X, Y, Z are parameters of the translational component P, A, B, C are characterized by ZYX Euler angle parametersThe parameter of the rotational component R of (a);

by six-membered group S and coordinate value KCs^measuedAs a function of the parameter, ω_iIs the weight of the ith key feature point pair.

4. An assembly method according to any one of claims 1 to 3, characterized in that: in the parameter correction step, the coordinate values KCs of the key feature points on the complex part after correction are calculated based on the following formula^measured；

Wherein, KCs₀ ^measuredFor direct measurement of the coordinates of key feature points on the complex part,

the coordinate value deviation caused by the temperature of the assembly environment is obtained for the simulation,

the coordinate value deviation caused by gravity is obtained for simulation.

5. An assembly method according to any one of claims 1 to 3, characterized in that: the complex part is a wallboard and is adjusted based on a plurality of numerical control posture adjusting positioning devices arranged at preset intervals along the spanwise direction of the wallboard, and each numerical control posture adjusting positioning device comprises a ball head locking mechanism which is used for forming ball head hinge joint with a supporting ball head arranged on the outer board surface of the wallboard.

6. The assembly method of claim 5, wherein the step of adjusting the position comprises the steps of: based on formula JPs^desired＝T*JPs^currentEach numerical control adjusting and positioning device for calculating and adjusting airplane structure to the target poseCenter of sphere coordinates JPs for supporting ball head^desired，JPs^currentThe measured value of the spherical center coordinate of the supporting ball head on each numerical control adjusting and positioning device under the assembly coordinate system is obtained.

7. The assembly method of claim 6, wherein: if the number of the supporting ball heads for coupling the wall plate with the same numerical control posture adjusting positioning device is more than two, taking a connecting line of fixed connection positions of the more than two supporting ball heads and a plurality of points on an extension line thereof as the key characteristic points; if the number of the supporting ball heads for coupling the wall plate with the same numerical control posture adjusting positioning device is more than one, at least the central point of the fixed connection position of the supporting ball head is used as the key characteristic point.

Images