A dynamic optimization method for multi-screw tightening with minimum assembly clearance

Abstract

Most existing radar antenna assembly methods rely on theoretical calculations for assembly process surface parameters. These parameters often deviate from the actual product surface, significantly impacting the accuracy of assembly gap prediction. This invention addresses the high-quality electrical signal transmission requirements of complex electromechanical products like radar antennas by proposing a dynamic optimization method for multi-screw tightening aimed at minimizing assembly gaps. Based on measured data, it accurately predicts the impact of the assembly process on interlayer gaps. With the goal of minimizing assembly gaps, it optimizes the screw tightening sequence and loading cycles. By repeatedly acquiring 3D point cloud data of the workpiece to reconstruct the model, searching for adjacent points of holes and calculating gaps, the screw tightening sequence is planned. Finite element analysis is used to compare the average gap with the planned loading cycles, thereby optimizing these process parameters to obtain the minimum assembly gap and ensuring high-quality electrical signal transmission between different layers of the radar antenna.

Description

Technical Field

[0001] This invention relates to the field of fastener assembly, and more specifically to a dynamic optimization method for multi-screw tightening oriented towards minimizing assembly clearance. Background Technology

[0002] Complex electromechanical products, such as radar antennas, are characterized by strong electromechanical coupling. These products typically consist of layered components, with various electrical connectors used between layers for electrical signal transmission. The interlayer gap, a decisive factor in the quality of electrical signal transmission, is primarily influenced by the machining accuracy of the layered components and the product assembly process. Currently, these products are increasingly developing towards high integration and complex shapes. Therefore, some layers are manufactured using polymer materials and composite additive manufacturing processes. While this process can produce products with complex shapes, its manufacturing principles often lead to warping due to residual stress accumulation, resulting in large and uneven interlayer gaps after assembly. Existing assembly process planning methods predict the surface state during product assembly through theoretical calculations and simulations, and then plan the threaded connection tightening process based on the predicted product shape.

[0003] Most existing methods rely on theoretical calculations for assembly process surface parameters. These parameters often deviate from the actual product surface, significantly impacting the accuracy of assembly clearance prediction. Therefore, exploring a simple, efficient, and accurate assembly clearance prediction method for assembly process optimization is of great significance. Summary of the Invention

[0004] To address the high-quality electrical signal transmission requirements of complex electromechanical products such as radar antennas, this invention proposes a dynamic optimization method for multi-screw tightening aimed at minimizing assembly gaps. Based on measured data, it accurately predicts the impact of the assembly process on interlayer gaps, optimizing the screw tightening sequence and loading cycles with the goal of minimizing assembly gaps. According to the research requirements, this method plans the screw tightening sequence by repeatedly acquiring 3D point cloud data of the workpiece to reconstruct a model, searching for adjacent points of holes and calculating gaps, and using finite element analysis to compare the average gap and plan the loading cycles. This optimizes these process parameters to obtain the minimum assembly gap, ensuring high-quality electrical signal transmission between different layers of the radar antenna. Specifically, it includes the following steps:

[0005] Step S1, acquire point cloud data: Use a point cloud camera to repeatedly scan the antenna to acquire surface point cloud data.

[0006] Step S2, Reconstruct the product model: Filter the point cloud data using the PCL library and register the point cloud data, then reconstruct the model based on the point cloud data.

[0007] Step S3, determine the maximum preload and loading times: Determine the maximum preload, preset the loading times scheme to N, simulate the assembly process of a certain tightening sequence under the N loading times scheme using the finite element method, and output the product assembly gap field quantity; determine the assembly loading times based on the average assembly gap at all mesh nodes.

[0008] Step S4: Determine the location of the screws to be tightened: Mesh the reconstructed model obtained in Step S2, calculate the gap between the upper and lower layers without applying a load, and output the position parameters and gap values ​​of all nodes on the contact surface; establish a gap database A = {A1,...A1} after x screws have been tightened. n}, where n is the total number of screws to be installed, A n ={x n ,y n ,z n ,d n} represents the node gap vector, including node coordinate information (x, y, z) and the gap value d of the node position; the number of neighboring points is set to a, and a neighboring point database L = {L1, ..., L...} is established. j}, where j = nx is the number of screws not installed on the product to be assembled; in the clearance database A, find the hole i for which no screw is installed, 1≤i≤j, and the node whose center of the circle intersecting the assembly contact surface has the closest three-dimensional Euclidean distance, and extract it from the clearance database and add it to its neighboring point database L. i In the middle, repeat the search until the nearest point in the database L is reached. i If the number of midpoints reaches 'a', then the neighboring point database L... i Data addition complete; continue adding neighboring point libraries L for uninstalled screw hole i+1. i+1 Add neighboring points. After all j neighboring point libraries have been added, calculate the average gap d between all points in each neighboring point library. i Then the largest value d among the j average gap values max =Max{d1,…,d j The corresponding hole position is the hole position of the threaded connector to be installed; the screw tightening sequence is obtained in this way.

[0009] Step S5: Complete assembly: Repeat steps S1, S2, and S4 until all screws have been loaded for the first time. Then continue loading in the order of the first loading.

[0010] Furthermore, in step S2, the point cloud data is filtered using the PCL library, including pass-through filtering and statistical filtering. The 3D-NDT algorithm is used to register the partitioned point cloud data, and the Digital Surface Editor (DSE) and the Quick Surface Reconstruction Module (QSR) in CATIA are used as a reverse modeling platform to reconstruct the model.

[0011] Furthermore, step S3 also includes setting a surface-to-surface contact between the upper surface of the antenna and the lower surface of the lower layer, setting the sliding friction coefficient to 0.2, applying a fixed constraint to the lower side, applying a linear force between the centers of the corresponding screw holes on the upper surface of the upper layer and the lower surface of the lower layer, and adding coupling constraints to the centers and the corresponding inner walls of the holes.

[0012] Furthermore, in step S4, the finite element analysis method is used to mesh the model reconstructed in step S2.

[0013] Furthermore, in step S4, the number of neighboring points 'a' is related to the model mesh generation accuracy, satisfying a≥(m / δ). 2 , where m is the distance between adjacent holes and δ is the grid division accuracy.

[0014] The beneficial effects of this invention are as follows:

[0015] 1. By repeatedly collecting 3D point cloud data to reconstruct the product model, the tightening process of threaded connectors can be planned based on the measured surface parameters of the product during the assembly process, thereby improving the accuracy of the assembly gap prediction results.

[0016] 2. By searching for nearby points of the hole and calculating the gap, the tightening sequence of the threaded connector is planned. The average gap is calculated and compared using the finite element method to plan the number of loading times, which can effectively solve the problem of threaded connector tightening process planning for minimum assembly gap. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a threaded fastener for a radar antenna and its structure.

[0018] Figure 2 This is a flowchart of the tightening process for threaded fasteners.

[0019] Figure 3 This is a 3D point cloud camera image of the product surface.

[0020] Figure 4 The result of finite element analysis is an assembly gap contour map.

[0021] The reference numerals in the attached diagram have the following meanings: 1-screw; 2-radiative feed layer; 3-liquid cooling layer. Detailed Implementation

[0022] The following will be combined with the appendix Figure 1-4 The technical solutions in the embodiments of the present invention have been clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] See Figure 1 The threaded connection structure includes a screw 1, a lower liquid cooling layer 3, and an upper radiation feeding layer 2. The liquid cooling layer is a machined metal part, fixed by a fixture, and its spatial position remains unchanged. The radiation feeding layer is composed of multiple layers of heterogeneous parts, which are fixed to the liquid cooling layer 3 by the screw 1.

[0024] See Figure 2 This invention provides a dynamic optimization method for multi-screw tightening with minimum assembly clearance, and the specific implementation steps are combined with... Figure 1 The threaded connection structure shown is described as follows:

[0025] Step S1: Acquire point cloud data

[0026] See Figure 3 The LMI Gocator3210 camera was used to take pictures of the upper 10 partitions to obtain point cloud data. The web page provided by LMI and the point cloud processing tool CsvConverter were used to record the point cloud data of all partitions and export the point cloud data as a txt text file.

[0027] Step S2: Fit the upper surface

[0028] A point cloud processing program was developed using the PCL (Point Cloud Library) tool. Pass-through filtering and statistical filtering were applied to all acquired partitioned point cloud data to obtain high-quality point cloud data. The 3D-NDT algorithm was then used to register the partitioned point cloud data, resulting in the overall point cloud data for the upper surface. The Digital Surface Editor (DSE) and Quick Surface Reconstruction (QSR) modules in the CATIA modeling software were used as reverse modeling platforms to reconstruct the upper-layer model.

[0029] Step S3: Planning of Preload and Loading Cycles

[0030] The maximum preload was determined based on engineering experience. Abaqus was selected as the finite element analysis software. The reconstructed model was imported into the software, and material properties were defined according to the actual working conditions. The upper layer's material properties were defined using engineering constants and solid element layups, while the lower layer's elastic modulus and Poisson's ratio were defined to establish a linear elastic model. The mesh type used was C3D8R. The contact type between the lower surface of the upper layer and the upper surface of the lower layer was set to surface-to-surface contact, with a sliding friction coefficient of 0.2. Fixed constraints were applied to the lower layer's side surfaces. Linear forces were applied between the centers of the corresponding screw holes on the upper surface of the upper layer and the lower surface of the lower layer, and coupling constraints were added to the centers and the corresponding hole inner walls. A tightening sequence is selected, with five different process settings for analysis steps: 1 direct loading; 2 loadings, each applying 50% of the maximum preload; 3 loadings, each applying 30%, 30%, and 40% of the maximum preload; 4 loadings, each applying 25% of the maximum preload; and 5 loadings, each applying 20% ​​of the maximum preload. Five assembly process simulations are performed for each of these five settings, outputting the gap field quantity COPEN between the upper and lower layers. The number of loading operations is determined based on the average gap value of all high-value points, taking into account assembly cost factors.

[0031] Step S4: Identify the screws to be tightened

[0032] Import the reconstructed model obtained in step S2 into ABAQUS, set the same material properties, boundary conditions, contact methods, and mesh as in step S3, and output the gap field quantity COPEN between the upper and lower layers without applying any load. Set the number of nearest neighbors to 15, and use Visual Studio 2019 to write a nearest neighbor query program to find the 15 nearest mesh nodes in Euclidean space for each hole with an untightened screw in the gap field COPEN obtained in this step, and record them as the nearest neighbors of the corresponding hole. Calculate the average assembly gap of the nearest neighbors for each hole. See the simulation results for [link to simulation results]. Figure 4 Select the hole with the largest average clearance as the screw that needs to be tightened. Install the screw in this hole and tighten it according to the preload set in step S3.

[0033] Repeat steps S1, S2, and S4 until all holes have screws installed; the first loading is now complete. Complete the remaining loading cycles in the same order as the first loading cycle; the product assembly is now complete. Use industrial CT scanning to inspect the assembly gaps and compare them to product quality requirements to determine the effectiveness of the planned assembly process.

[0034] This invention is not limited to the specific embodiments described above, and various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made to the above embodiments based on the technical essence of this invention should be included within the scope of protection of this invention.

Claims

1. A dynamic optimization method for multi-screw tightening oriented towards minimum assembly clearance, characterized in that: The specific steps include: Step S1: Acquire point cloud data The antenna is repeatedly scanned using a point cloud camera to acquire surface point cloud data; Step S2, Reconstruct the product model The point cloud data is filtered and registered using the PCL library, and the model is reconstructed based on the point cloud data. Step S3: Determine the maximum preload and the number of loading cycles. The maximum preload force is determined, and there are N preset loading times. The assembly process of a certain tightening sequence under N loading times is simulated using the finite element method, and the assembly gap field is output. The number of assembly loading times is determined based on the average assembly gap at all mesh nodes. Step S4: Determine the position of the screw to be tightened The reconstructed model obtained in step S2 is meshed, and the gap between the upper and lower layers is calculated without applying a load. The position parameters and gap values ​​of all nodes on the contact surface are output. A gap database A = {A1,...A2} is established after x screws have been tightened. n }, where n is the total number of screws to be installed, A n ={x n ,y n ,z n ,d n } represents the node gap vector, including node coordinate information (x, y, z) and the gap value d of the node position; the number of neighboring points is set to a, and a neighboring point database L = {L1, ..., L...} is established. j }, where j = nx is the number of screws not installed on the product to be assembled; in the clearance database A, find the hole i for which no screw is installed, 1≤i≤j, and the node whose center of the circle intersecting the assembly contact surface has the closest three-dimensional Euclidean distance, and extract it from the clearance database and add it to its neighboring point database L. i In the middle, repeat the search until the nearest point in the database L is reached. i If the number of midpoints reaches 'a', then the neighboring point database L... i Data addition complete; continue adding neighboring point libraries L for uninstalled screw hole i+1. i+1 Add neighboring points. After all j neighboring point libraries have been added, calculate the average gap d between all points in each neighboring point library. i Then the largest value d among the j average gap values max =Max{d1,...,d j The corresponding hole position is the hole position of the threaded connector to be installed; the screw tightening sequence is obtained in this way. Step S5: Complete assembly Repeat steps S1, S2, and S4 until all screws have completed the first loading. Then, continue loading the subsequent screws in the order of the first loading.

2. The multi-screw tightening dynamic optimization method for minimizing assembly clearance according to claim 1, characterized in that: In step S2, the point cloud data is filtered using the PCL library, including pass-through filtering and statistical filtering. The 3D-NDT algorithm is used to register the partitioned point cloud data. The Digital Surface Editor (DSE) and the Quick Surface Reconstruction (QSR) module in CATIA are used as reverse modeling platforms to reconstruct the model.

3. The multi-screw tightening dynamic optimization method for minimizing assembly clearance according to claim 1, characterized in that: Step S3 further includes setting a surface-to-surface contact between the upper surface of the antenna and the lower surface of the lower layer, setting the sliding friction coefficient to 0.2, applying a fixed constraint to the lower side, applying a linear force between the centers of the corresponding screw holes on the upper surface of the upper layer and the lower surface of the lower layer, and adding coupling constraints to the centers and the corresponding inner walls of the holes.

4. The multi-screw tightening dynamic optimization method for minimizing assembly clearance according to claim 1, characterized in that: There are five loading schemes, including: one direct loading; two loadings, each loading 50% of the maximum preload; three loadings, each loading 30%, 30%, and 40% of the maximum preload; four loadings, each loading 25% of the maximum preload; and five loadings, each loading 20% ​​of the maximum preload.

5. The multi-screw tightening dynamic optimization method for minimizing assembly clearance according to claim 1, characterized in that: In step S4, the finite element analysis method is used to mesh the model reconstructed in step S2.

6. The multi-screw tightening dynamic optimization method for minimizing assembly clearance according to claim 1, characterized in that: In step S4, the number of neighboring points 'a' is related to the model mesh generation accuracy, satisfying a≥(m / δ). 2 , where m is the distance between adjacent holes and δ is the grid division accuracy.

Images