A quadratic assignment method for mass distribution in aircraft wing surface dynamics modeling

Abstract

The present invention relates to a method for secondary mass distribution in aircraft wing surface dynamics modeling, belonging to the field of aircraft structural dynamics. A method for secondary mass distribution in aircraft wing surface dynamics modeling includes the following steps: Step 1: Divide the finite element model of the aircraft wing surface into two groups of sub-wing surfaces m and n, and output them from the finite element software respectively; Step 2: Project the mass points in the mass distribution input file and the structural nodes in the sub-wing surfaces m and n onto this two-dimensional plane; Step 3: According to the principle that the mass center of mass remains unchanged before and after mass distribution, calculate the mass mi of the mass point Mi during the process of distributing it to the sub-wing surfaces m and n; Step 4: Calculate the process mass and distribute it to the host unit node masses of the sub-wing surfaces m and n respectively; Step 5: Loop through the processes of Steps 2 to 4 for all the mass points in the mass distribution input file, that is, complete the entire process of secondary mass distribution in aircraft wing surface dynamics modeling. The present invention has the following advantages: 1. The modeling efficiency is greatly improved; 2. It meets the requirements of refined aircraft modeling.

Description

Technical Field

[0001] The present invention relates to a method for quadratic mass distribution in aircraft wing surface dynamics modeling, belonging to the field of aircraft structural dynamics. Background Technique

[0002] The aircraft structural dynamics model is a simulation model related to the study of aircraft dynamic characteristics. Mass property modeling is an important process in aircraft structural dynamics modeling. Taking the aircraft flutter design specialty as an example, mass property modeling is as important as finite element stiffness modeling. To truly reflect the flutter characteristics of aircraft wings, vertical tails, horizontal tails and other wing surfaces, mass property modeling should not only ensure that the overall mass and centroid of the wing surface are consistent with the actual aircraft, but also simulate the mass distribution of the wing surface in detail.

[0003] There are usually two ways to model the mass distribution of aircraft wing surfaces: one is to refer to the wing surface structure digital model and assign appropriate material density attributes to the corresponding structures in the finite element model to simulate the mass distribution of the wing surface. Since the wing surface finite element model is usually simplified based on stiffness equivalence, in actual operation, there may be a certain deviation between the mass distribution of the simplified model and the mass centroid of the actual structure, and multiple trials are required to achieve the consistency of the overall mass centroid of the wing surface and approximately simulate the mass distribution of the wing surface, and a mass model correction process needs to be added in subsequent dynamic corrections; the other is based on the mass distribution input file after the wing surface structure digital model is partitioned and sliced, and the mass distribution modeling process is completed by applying the mass points of the mass distribution input file to the wing surface finite element model. This modeling method decouples the wing surface stiffness model and the mass model, but if the mass distribution is manually applied to each mass point one by one to the wing surface structure, it is time-consuming and laborious, seriously affecting the modeling cycle of the wing surface mass distribution. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems existing in the prior art, and provide a method for quadratic mass distribution in aircraft wing surface dynamics modeling, which is used to automatically and secondarily distribute the wing surface mass distribution to each mass point into the wing surface finite element model to improve the efficiency of aircraft wing surface dynamics mass distribution modeling.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: A method for quadratic mass distribution in aircraft wing surface dynamics modeling, including the following steps:

[0006] Step 1: Divide the aircraft wing surface finite element model into two groups of sub-wing surfaces m and n, and output them separately from the finite element software. For the sub-wing surface m and n finite element models, obtain the two-dimensional plate elements and node coordinate information therein respectively, process all the two-dimensional plate elements into triangular elements, and establish new index data, and its structure is as follows:

[0007] itria_node[]: Store the node numbers of the elements, and the array size is Ntria×3;

[0008] itria_edge[]: The component edge numbers of the storage unit, with the array size Ntria×3;

[0009] iedge_node[]: The node numbers of the storage edge, with the array size Nedge×2;

[0010] iedge_tria[]: The adjacent unit numbers of the storage edge, with the array size Nedge×2;

[0011] inode_xyz[]: The coordinates of the storage node in the body coordinate system, with the array size Nnode×3;

[0012] In the above arrays, Ntria, Nedge, and Nnode are the total numbers of elements, edges, and nodes of the sub-wing surface m or n, respectively;

[0013] Step 2: Take the XZ or XY plane in the body coordinate system parallel to the wing surface, project the mass points in the mass distribution input file and the structural nodes in the sub-wing surfaces m and n onto this two-dimensional plane, and use the host element search algorithm in this projection plane to search for the host elements in the finite element models of the sub-wing surfaces m and n that can enclose each mass point in the mass distribution input file. After determining the host elements, return to the three-dimensional space and calculate the distance from each mass point in the mass distribution input file to its corresponding host element along the projection direction;

[0014] Step 3: According to the principle that the mass center of mass remains unchanged before and after mass distribution, calculate the mass m i of the mass point M i and distribute the process mass to the sub-wing surfaces m and n;

[0015] Step 4: From the arrays itria_node[] and inode_xyz[] in Step 1, in the projection plane in Step 2, according to the principle that the mass center of mass remains unchanged before and after mass distribution, calculate the process mass and distribute it to the node masses of the host elements of the sub-wing surfaces m and n respectively;

[0016] Step 5: Loop through the processes of Step 2 to Step 4 for all mass points in the mass distribution input file, that is, complete the entire process of secondary distribution of the mass distribution for the aircraft wing surface dynamics modeling.

[0017] Preferably, in Step 2, the host element search algorithm is a host element search algorithm based on adjacent elements.

[0018] Preferably, for any mass point M i in the mass input file, the host element search process includes the following steps:

[0019] (1) Enter the projection plane. Assume the projection plane is the UV plane, and the coordinates of any point P in the UV plane are (u P , v P ); Take the unit △A1B1C1 with the current sub-wing surface number 1 as the mass point 's initial search unit. Assume the current search unit is △A j B j C j . Take its centroid to calculate the plane straight line equation of M i O j

[0020]

[0021] (2) According to the array itria_edge[] in step one, take the first edge AB of the unit △A j B j C j to calculate its plane straight line equation j B j and solve the intersection point P

[0022]

[0023] of the straight line AB j B j and M i O j j1

[0024]

[0025] Establish a new search unit advancement criterion: If the inner product of vectors and both hold, then according to the array iedge_tria[] in step one, obtain the other adjacent unit of the edge AB as the new search unit, and restart the search process from step (1); otherwise, take the second edge BC of the unit △A j B j to establish a similar new search unit advancement condition for the intersection point P j B j C j of BC and M j C j O j C j ; For the third edge CA of the unit △A i O j of CA and M j2 establish a similar new search unit advancement condition; for the unit △A j B j C j take the third edge CA j A​​j The intersection point P with M i O j ; j3 ;

[0026] (3) Establish the host unit discrimination criterion: If the current search unit △A of the loop j B j C j For all three sides A j B j 、B j C j 、C j A j do not satisfy the new search unit advancement criterion in step (2), then the current search unit △A j B j C j is the host unit of the mass point M i ;

[0027] (4) Return to the three-dimensional space. Establish a space coordinate system OUVW with the projection plane UV as one of the coordinate planes. The space coordinates of any point P are (u P , v P , z P ). Calculate the space plane equation of the host unit △A i of the mass point M j B j C j in the space coordinate system OUVW

[0028]

[0029] Solve for the distance from the mass point M i to its host unit △A j B j C j along the projection direction ;

[0030]

[0031] Preferably, in step three, the mass m i of the mass point M i is distributed to the sub-wing surfaces m and n. The process mass is

[0032]

[0033] where are the distances from the mass point M i to the host units of the sub-wing surfaces m and n, respectively, and satisfy the condition .

[0034] Preferably, in step four, the allocation to the host unit △A j B j C j Node A j 、B j 、C j The mass is

[0035]

[0036] Wherein, is the process mass allocated to the sub-wing surface m or n.

[0037] Principle: According to the aircraft wing surface mass distribution input file provided by the aircraft weight-related specialty, each mass point in the mass distribution input file is re-allocated to the structural nodes in the wing surface finite element model that are close to the mass points, and it is ensured that the mass and the centroid of mass are consistent before and after the re-allocation of each mass in the mass distribution input file.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] 1. The process of the secondary allocation of the mass distribution in the aircraft wing surface dynamics modeling is programmed and automated, and the modeling efficiency of the wing surface mass model is greatly improved;

[0040] 2. The host unit search method based on adjacent units accurately ensures that the centroid of the mass of the wing surface partition is consistent with the original mass distribution input, and can meet the requirements of the aircraft flutter design and other specialties for the refined modeling of the mass distribution of the wing surface dynamics model. Brief Description of the Drawings

[0041] Figure 1 is the implementation process block diagram of the wing surface mass secondary allocation method in the embodiment of the present invention;

[0042] Figure 2 is the illustration of the host unit search process in the embodiment of the present invention (the arrow-connected broken line in the figure is the host unit search process link based on adjacent units);

[0043] Figure 3 is the model illustration in the embodiment of the present invention (wherein, the upper and lower triangular surfaces are the finite element models of the sub-wing surfaces m and n, and the solid dots are the mass points in the mass distribution input to be re-allocated). Specific Embodiments

[0044] The following further describes the present invention with reference to the attached Figures 1 - 3 Refer to the attached Figure 1 、attached Figure 2, according to the aircraft wing surface mass distribution input file provided by the aircraft weight-related specialty, redistribute each mass point in the mass distribution input file to the structural nodes in the wing surface finite element model that are close to the mass points, and ensure that the mass and centroid of each mass in the mass distribution input file remain the same before and after redistribution.

[0045] A method for redistributing the mass distribution in aircraft wing surface dynamics modeling includes the following steps:

[0046] Step 1: Divide the aircraft wing surface finite element model into two groups of sub-wing surfaces m and n and output them from the finite element software respectively. For the finite element models of sub-wing surfaces m and n, obtain the two-dimensional plate elements and node coordinate information therein respectively, process all the two-dimensional plate elements into triangular elements, and establish the following new index data structures:

[0047] itria_node[]: Store the node numbers of the elements, with the array size Ntria×3;

[0048] itria_edge[]: Store the composition edge numbers of the elements, with the array size Ntria×3;

[0049] iedge_node[]: Store the node numbers of the edges, with the array size Nedge×2;

[0050] iedge_tria[]: Store the adjacent element numbers of the edges, with the array size Nedge×2;

[0051] inode_xyz[]: Store the coordinates of the nodes in the body coordinate system, with the array size Nnode×3.

[0052] In the above arrays, Ntria, Nedge, and Nnode are the total numbers of elements, edges, and nodes of sub-wing surface m or n respectively;

[0053] Step 2: Take the XZ or XY plane in the body coordinate system that is approximately parallel to the wing surface, project the mass points in the mass distribution input file and the structural nodes in sub-wing surfaces m and n onto this two-dimensional plane, and search for the host elements in the finite element models of sub-wing surfaces m and n that can enclose each mass point in the mass distribution input file by using the host element search algorithm based on adjacent elements in this projection plane. After determining the host elements, return to the three-dimensional space and calculate the distances from each mass point in the mass distribution input file to its corresponding host element along the projection direction;

[0054] For any mass point M in the mass input file i , taking one of sub-wing surfaces m and n as an example, the host element search process based on adjacent elements is as follows:

[0055] 2.1 Enter the projection plane. Assume the projection plane is the UV plane, and the coordinates of any point P on the UV plane are (u P , v P ). Take the element △A1B1C1 with the current sub-wing surface number being 1 as the initial search element for the mass point . Generally, assume the current search element is △A j B j C j . Take its centroid to calculate the plane straight line equation of M i O j

[0056]

[0057] 2.2 According to the array itria_edge[] in Step 1, take the first edge AB of the element △A j B j C j to calculate its plane straight line equation j B j

[0058]

[0059] and solve the intersection point P j B j of the straight line AB and M i O j j1

[0060]

[0061] Establish a new search element advancement criterion: If the inner product of vectors and both hold, then according to the array iedge_tria[] in Step 1, obtain the other adjacent element of the edge AB as the new search element, and restart the search process from Step 2.1; otherwise, take the second edge BC of the element △A j B j to establish a similar new search element advancement condition for the intersection point P j B j C j of BC and M j C j . Similarly, for the third edge CA of the element △A j C j and M i O j to establish a similar new search element advancement condition for the intersection point P j2 . Similarly, for the third edge CA of the element △A j B j C j , take the third edge CA of the element △A j A​​​j The intersection point P with M i O j ; j3 ;

[0062] 2.3 Establish the host unit discrimination criterion: If the three sides A j B j C j of the current search unit △A j B j 、B j C j 、C j A j do not satisfy the new search unit advancement criterion in step 2.2, then the current search unit △A j B j C j is the host unit of the mass point M i ;

[0063] 2.4 Return to the three-dimensional space, establish a space coordinate system OUVW with the projection plane UV as one of the coordinate planes, and calculate the space plane equation of the host unit △A i of the mass point M j B j C j in the space coordinate system OUVW

[0064]

[0065] Solve the distance from the mass point M i to its host unit △A j B j C j along the projection direction ;

[0066]

[0067] Step 3: Assume that the distances from the mass point M i to the host units of the sub-wing surfaces m and n are respectively Under the condition of satisfying , according to the principle that the mass and centroid remain unchanged before and after mass distribution, the mass m i of the mass point M i is distributed to the sub-wing surfaces m and n, and the process mass is

[0068]

[0069] Step 4: From the arrays itria_node[] and inode_xyz[] in step 1, in the UV projection plane in step 2, according to the principle that the mass and centroid remain unchanged before and after mass distribution, calculate the process mass The mass of the host unit nodes is respectively allocated to the sub-airfoils m and n. Taking one of the sub-airfoils m and n as an example, assume the process mass is Allocated to the host unit △A j B j C j Node A j , B j , C j The mass is

[0070]

[0071] Step 5: Loop the processes of Step 2 to Step 4 for all the mass points in the mass distribution input file, that is, complete the entire process of the secondary distribution of the aircraft wing surface dynamics modeling mass distribution.

[0072] A specific embodiment of the secondary distribution of the mass distribution based on the present invention. The finite element models of the sub-airfoils m and n are as Figure 3 Shown by the upper and lower triangular surfaces above. The solid dots are the mass points for the secondary distribution of the mass distribution input. Tables 1 and 2 respectively give the element and node coordinate data in the finite element models of the sub-airfoils m and n; Table 3 is the mass and centroid coordinate data of each mass point in the mass distribution input; Table 4 is the secondary distribution result of each mass point in the mass distribution input. Compared with the mass and centroid of each mass point in the mass distribution input, the relative error is less than 0.03%. In fact, the above errors are all caused by the rounding of the data decimal places. In principle, the method based on the present invention can ensure that the mass and centroid are exactly the same before and after the secondary distribution of the wing surface mass.

[0073] Table 1 Element and Node Coordinate Data of the Finite Element Model of Sub-airfoil m

[0074]

[0075]

[0076] Table 2 Element and Node Data of the Finite Element Model of Sub-airfoil n

[0077] Unit number Node 1 number Node 2 number Node 3 number Node number x (mm) y (mm) z (mm) 1 10 12 7 1 315.84 20.39 212.94 2 7 11 10 2 426.07 130.03 194.40 3 16 10 11 3 262.90 349.48 216.06 4 11 1 16 4 369.34 349.46 205.97 5 2 16 1 5 149.52 459.23 210.12 6 1 13 2 6 340.96 459.23 210.15 7 15 14 12 7 36.31 20.39 187.00 8 12 10 15 8 155.87 349.50 210.90 9 9 15 10 9 280.55 239.75 215.44 10 10 16 9 10 168.57 130.08 212.32 11 18 9 16 11 174.91 20.39 212.94 12 16 2 18 12 41.06 130.10 188.31 13 8 17 14 13 454.42 20.39 187.00 14 14 15 8 14 45.80 239.81 189.60 15 3 8 15 15 162.22 239.78 211.64 16 15 9 3 16 298.20 130.06 214.40 17 4 3 9 17 50.54 349.52 190.85 18 9 18 4 18 397.71 239.72 200.72 19 5 20 17 19 245.23 459.23 216.27 20 17 8 5 20 55.28 459.23 192.07 21 19 5 8 22 8 3 19 23 6 19 3 24 3 4 6

[0078] Table 3 Mass and Centroid Coordinate Data of the Mass Distribution Input

[0079]

[0080]

[0081] Table 4 Results of the Secondary Distribution of the Mass Distribution

[0082]

[0083] *: The centroid error is calculated by taking the characteristic length of 100 mm.

Claims

1. A quadratic allocation method for mass distribution in aircraft wing surface dynamics modeling, characterized in that: It includes the following steps: Step 1: Divide the finite element model of the aircraft wing surface into two groups of sub-wing surfaces m and n, and output them from the finite element software respectively. For the finite element models of sub-wing surfaces m and n, obtain the two-dimensional plate elements and node coordinate information therein respectively. Process all the two-dimensional plate elements into triangular elements, and establish new index data, the structure of which is as follows: itria_node[]: Store the node numbers of the elements, and the array size is Ntria×3; itria_edge[]: Store the component edge numbers of the elements, and the array size is Ntria×3; iedge_node[]: Store the node numbers of the edges, and the array size is Nedge×2; iedge_tria[]: Store the adjacent element numbers of the edges, and the array size is Nedge×2; inode_xyz[]: Store the coordinates of the nodes in the body coordinate system, and the array size is Nnode×3; In the above arrays, Ntria, Nedge, and Nnode are the total numbers of elements, edges, and nodes of sub-wing surface m or n respectively; Step 2: Take the XZ or XY plane in the body coordinate system parallel to the wing surface, project the mass points in the mass distribution input file and the structural nodes in sub-wing surfaces m and n onto the two-dimensional plane. In the projection plane, use the host element search algorithm based on adjacent elements to search for the host elements in the finite element models of sub-wing surfaces m and n that can enclose each mass point in the mass distribution input file. After determining the host elements, return to the three-dimensional space and calculate the distances from each mass point in the mass distribution input file to its corresponding host element along the projection direction; Step 3: Calculate the mass point M according to the principle that the mass centroid remains unchanged before and after mass distribution i The mass m i Distributed to the process mass of the sub-airfoils m and n; Step 4: From the arrays itria_node[] and inode_xyz[] in Step 1, within the projection plane in Step 2, calculate the process mass according to the principle that the mass centroid remains unchanged before and after mass distribution. Distribute the mass to the host unit node masses of the sub-airfoils m and n respectively; Step 5: Loop through the processes of Step 2 to Step 4 for all mass points in the mass distribution input file, that is, complete the entire process of secondary distribution of the mass distribution in the aircraft wing surface dynamics modeling.

2. The quadratic allocation method for mass distribution in aircraft wing surface dynamics modeling according to claim 1, characterized in that: For any mass point M in the mass input file i , the host cell search process includes the following steps: (1) Enter the projection plane. Assume the projection plane is the UV plane, and the coordinates of any point P in the UV plane are (u P , v P ); Take the unit △A1B1C1 with the current sub-wing surface number 1 as the mass point 's initial search unit. Assume the current search unit is △A j B j C j , and take its centroid to calculate the M i O j plane straight line equation (2) According to the array itria_edge[] in Step 1, take the first edge of element △A j B j C j to calculate its plane straight line equation j B j ​ And solve for line A j B j and M i O j at the intersection point P j1 Establish a new search unit advancement criterion: If the vector inner product If both are true, then according to the array iedge_tria[] in step 1, get edge A j B j The other adjacent unit of is taken as the new search unit and the search process is restarted from step (1); otherwise, take unit △A j B j C j The second side B j C j , for B j C j With M i O j The intersection point P j2 Establish similar new search unit advancement conditions; for unit △A j B j C j The third side C j A j With M i O j The intersection point P j3 , repeat the new search unit advancement condition; (3) Establish the discrimination criterion for the host unit: If the currently searched unit △A in the loop j B j C j All three sides A j B j and B j C j and C j A j do not satisfy the new search unit advancement criterion in step (2), then the currently searched unit △A j B j C j is the host unit of the mass point M i ; (4) Return to the three-dimensional space, establish a space coordinate system OUVW with the projection plane UV as one of the coordinate planes, and the space coordinates of any point P are (u p , v p , w p ). Calculate the mass point M i of the host unit △A j B j C j in the space coordinate system OUVW Solve for the mass point M i to its host element △A j B j C j along the projection direction distance 3. The quadratic allocation method for mass distribution in aircraft wing surface dynamics modeling according to claim 2, characterized in that: In step 3, the mass point M i with mass m i is distributed to the sub-airfoils m and n, and the process mass is Among them, are the distances from the mass point M i to the host units of the sub-wing surfaces m and n, respectively, and satisfy the condition of.

4. The quadratic allocation method for mass distribution in aircraft wing surface dynamics modeling according to claim 2, characterized in that: In step four, allocate to host unit △A j B j C j Node A j , B j , C j with a mass of Among them, is the process quality assigned to sub-airfoil m or n.

Images