Phased-array antenna assembly process planning method based on digital twinning

Abstract

The invention relates to a phased-array antenna assembly process planning method based on digital twinning, which is used for researching a large-scale phased-array antenna assembly process technology and carrying out assembly process planning based on digital twinning, and comprises the following steps of: extracting a contact surface according to plane geometric characteristics of acquired data, and matching a workpiece contact surface with a panel contact surface; performing assembly feasibility prediction and simulation; carrying out optimal design on an assembly process of the model; and assembling quality online detection and accurate compensation are realized. According to the method, an advanced installation path planning method is adopted for multiple workpieces, so that the problems of low flexibility, low assembly efficiency, difficulty in quality control, high cost and the like of an automatic assembly unit of the large-scale phased-array antenna are solved, and the assembly requirements of high precision, high complexity, high quality and high consistency of the antenna array are met; therefore, the demonstration application of intelligent closed-loop control of the assembly process represented by a large-scale phased-array antenna is formed.

Description

Technical Field

[0001] The present invention relates to computer graphics, data reconstruction, feature extraction and assembly process planning, and is specifically suitable for optimizing the precision assembly process after data collection for a phased array antenna workpiece. Background Art

[0002] The phased array antenna is composed of many radiating units and receiving units (called oscillators), with the number of units ranging from hundreds to tens of thousands. These units are regularly arranged on a plane to form an array antenna. It uses the principle of electromagnetic wave coherence and controls the phase of the current fed to each radiating unit through a computer to change the direction of the beam for scanning, so it is called electronic scanning. The radiating unit sends the received echo signal to the host to complete the radar's search, tracking and measurement of the target.

[0003] Taking a certain type of phased array antenna under development in China as an example, its size range is 1m×1m, and it is mainly composed of four-in-one oscillators, two-in-one oscillators, grid plates, panels, power dividers, back frames and standard parts. Among them, there are 80 four-in-one oscillators, 16 two-in-one oscillators, 23 grid plates, 1 panel, 70 power dividers, 1 back frame, and 181 processed parts of 6 types. The requirements for assembly accuracy are: under the scale of 1m×1m antenna array, the flatness of the antenna array surface is less than 1mm, and the dimensional error of the oscillator spacing is less than 0.5mm.

[0004] The specific assembly method is: number different types of parts separately, measure the key geometric dimensions of all parts (about 180) to accurately verify the correctness of the results, and then conduct traditional inspection and measurement by manual inspection, and generate electronic spreadsheets according to the number records. In order to improve the overall assembly results, this method requires repeated inspection, installation, adjustment and recording, which takes a very long time and cannot guarantee the assembly quality.

[0005] At this stage, the shortcomings of traditional assembly planning are obvious: first, the cycle and quality cannot be effectively guaranteed, and the traditional method is inefficient, time-consuming and labor-intensive; second, the traditional method will waste a lot of manpower, and multiple workers will participate and cooperate in the work, and manual work will lead to the introduction of reading and measurement errors, and will greatly increase the manufacturing cost; if a large phased array antenna is assembled, the number of workpieces that need to be assembled will be several times or even dozens of times that of a small phased array antenna, and the workload and time consumption will be incomparable to that of a small phased array antenna; once the vibrator is replaced, it will affect the entire assembly result, and the universality of the assembly cannot be achieved.

[0006] The precision assembly processing of phased array antennas requires improvement and innovation of the existing assembly process planning methods. The ability to adaptively select assembly solutions and generate the overall optimal assembly process path based on the geometric characteristics of assembly surface contact and the overall optimal assembly tolerance is the core of efficient and high-precision assembly processing. Summary of the invention

[0007] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a phased array antenna assembly process planning method based on digital twins, adopt an application solution of digital blue technology for phased array antenna assembly, and realize a process planning method for the assembly process by constructing an information interaction interface connecting the digital space and the physical space.

[0008] The technical solution adopted by the present invention to achieve the above-mentioned purpose is:

[0009] A phased array antenna assembly process planning method based on digital twins analyzes the spatial positions of phased array antenna assembly parts and obtains subassemblies to provide assemblers with a simulation model with assembly errors and contour lines, and intuitively displays assembly prediction results; the method includes the following steps:

[0010] Step 1: Measure the geometric characteristic data of the phased array antenna assembly components respectively;

[0011] Step 2: Use the plane feature extraction method to segment each plane of the component; remove the non-contact surface data, retain the assembly contact surface data of each component and the corresponding geometric relationship of the assembly contact surface, and generate the subassembly of the current component;

[0012] Step 3: Align the contact surface of the current component with its design model; then obtain the assembly error; divide the assembly errors of all assembly positions in the assembly area into gradient levels, and count the number of assembly errors in each level that belong to the same level;

[0013] Step 4: Abstract the assembly area into an area with the same attribute value on the outer boundary, and obtain the contour trajectory map in the area;

[0014] Step 5: Calculate the contour lines corresponding to the assembly positions and assign the statistical assembly errors to the corresponding positions; finally, a graph with assembly errors and contour lines is obtained.

[0015] The phased array antenna assembly components include: a four-in-one vibrator, a two-in-one vibrator, a grid plate, a panel, and a power divider. The grid plate is used to insert the vibrator, and the panel is used to set the grid plate and the power divider with the inserted vibrator on it to form an assembly area.

[0016] The measurement is to sequentially number the phased array antenna components to be assembled, scan and remove noise.

[0017] The method for extracting plane features uses the feature extraction Feature Detection algorithm of the Computational Geometry Algorithms Library to extract plane features.

[0018] The geometric relationship is a spatial relationship matrix of any two assembly contact surfaces, which is used to obtain another assembly contact surface after a known contact surface is transformed through the spatial relationship matrix.

[0019] The spatial relationship matrix is ​​an adjacency matrix of each axis.

[0020] The alignment operation includes coarse matching and fine matching. The coarse matching is that the user selects the corresponding point pairs in the scan data and the mathematical model in the same order, and then performs an associated matching operation; the fine matching function is to directly call the ICP algorithm function of PCL to perform the operation.

[0021] The isovalue line calculation adopts any one of the isothermal lines in the temperature field, the isovoltage lines in the electromagnetic field or the isodecibel value lines in the sound field.

[0022] The denser the contour lines are, the more accurate the assembly accuracy of the corresponding assembly position is, and the higher the overall assembly accuracy is.

[0023] The present invention has the following beneficial effects and advantages:

[0024] 1. The method of the present invention adopts a phased array antenna assembly process planning method based on digital twins, which can ensure the overall assembly optimization and improve the efficiency and quality of assembly processing.

[0025] 2. The present invention adopts a method of extracting contact planes from the plane features of the geometric features of various workpieces, which can realize the extraction of sub-assemblies and greatly improve the efficiency of calculating assembly alignment.

[0026] 3. In the present invention, the distribution of assembly errors is simulated by using a partial differential equation curve, so that the assembly errors are more in line with natural laws, and parts with the same assembly direction and the same assembly tools are assembled continuously to improve assembly efficiency and reduce assembly costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the method of the present invention;

[0028] Figure 2 This is a diagram showing the vibrator assembly process.

[0029] Figure 3 The effect diagram of the vibrator assembly completion.

[0030] Figure 4This is the finished effect diagram of the power divider assembly.

[0031] Figure 5 Extract the contact surface features of the vibrator.

[0032] Figure 6 Schematic diagram of assembly position and isotherms of assembly area, where: Figure 6 (a) is the graph of the partial differential equation; Figure 6 (b) is an isometric contour diagram.

[0033] Figure 7 Error distribution and optimized assembly effect, among which, Figure 7 (a) Distribution and statistics of assembly errors of original parts; Figure 7 (b) Optimization of assembly error distribution that conforms to the isotherm distribution.

[0034] Figure 8 For the comparison of assembly effects, Figure 8 For (a) isometric contours; Figure 8 (b) is a partial differential equation. DETAILED DESCRIPTION

[0035] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation method of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the invention, so the present invention is not limited by the specific implementation disclosed below.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0037] like Figure 1 , a phased array antenna assembly process planning method based on digital twin, comprising the following steps:

[0038] Step 1: Based on the measurement data, according to the geometric features of the parts including the four-in-one oscillator, the two-in-one oscillator, the grid plate, the panel, and the power divider, each plane is segmented by extracting the plane features;

[0039] Step 2: Remove the data of the non-contact surfaces of each part, retain the corresponding geometric relationship, and re-save the data of each part; this method forms a sub-assembly with some data of complex parts, and then assembles the sub-assembly as a whole with other parts. This method can reduce the scale of problem solving and improve planning speed;

[0040] Step 3: Remove the data of the non-contact surface of the back frame parts, save and retain the corresponding geometric relationships of each assembly station, and re-save the data of the back frame, which is also to reduce the scale of problem solving and improve planning speed;

[0041] Step 4: Align the subassembly data above with the subassembly data of the back frame part to obtain an aligned matching relationship;

[0042] The main issues of assembly process planning are concentrated in two aspects: assembly sequence planning and assembly alignment strategy between the assembly area of ​​the part subassembly and the back frame subassembly:

[0043] (1) The assembly sequence plays a dominant role in the assembly process, guiding the order of the entire assembly process. At present, the assembly sequence planning of phased array antenna parts mainly relies on manual experience and lacks effective adaptive guiding principles. As the volume of phased array products increases, the number of corresponding parts increases, and the structure becomes more complex, relying solely on manual experience is bound to fall into a bottleneck.

[0044] (2) At the same time, the complexity of the assembly structure and the increase in volume will inevitably lead to an increase in assembly data, resulting in extremely low data computing efficiency. Subassemblies treat some parts of an assembly as a whole. The existence of subassemblies can reduce the complexity of assembly sequence planning, simplify assembly processes, and reduce the computational complexity of assembly and analysis.

[0045] Therefore, the core of the assembly process planning problem lies in analyzing the spatial position of parts, obtaining subassemblies, and designing assembly sequence planning algorithms.

[0046] The four-in-one vibrator, two-in-one vibrator, grid plate, and power divider are assembly parts, and the panel and back frame are parts to be assembled for loading the above-mentioned workpieces into a fixed installation position.

[0047] The assembly sequence is to first install the panel and the back frame, position the pins, tighten the screws, and then install the grid plate, vibrator, and power divider on the panel. Figure 2 As shown, position and connect the matched vibrator grid plate through the previously processed pin holes and threaded holes to ensure a reliable connection. Then install the vibrator unit vertically on the plug-in board. Before assembly, apply grease to the slots where the grid plate and vibrator match to ensure smooth installation without deformation. The assembly is completed as shown below. Figure 3 shown.

[0048] like Figure 4 As shown, after all the vibrators are assembled, pre-tighten the screws uniformly from the back of the panel, and then install the guide pins connected to the power divider on the vibrator. Connect the power divider to the vibrator from the back of the panel, and use the guide pins to guide during installation to prevent the pins from bending.

[0049] The assembly sequence planning problem is a typical NP (Non-deterministic Polynomial) problem, which belongs to the discrete space solution problem. NP problems have many possible results, but almost no feasible or optimal solutions. With the development of intelligent algorithms, many scholars have begun to try to apply intelligent optimization algorithms to assembly sequence planning problems, such as cat swarm algorithm, immune particle swarm algorithm and ant colony algorithm.

[0050] The method of using subassemblies to simplify the scale of solution is to remove the non-assembly contact surfaces of parts in a complex assembly, form a subassembly from the assembly surfaces, and then assemble the subassembly as a whole together with other parts. In this way, the scale of problem solution can be reduced and the planning speed can be improved.

[0051] The method of retaining the corresponding geometric relationship is as follows: after extracting the assembly feature information, converting the part assembly information into the adjacency matrix of each axis, and then designing an algorithm to generate the sub-assembly.

[0052] The extraction of assembly feature information is as follows: plane feature extraction of the point cloud model (the contact surfaces of the parts to be assembled of the phased array antenna in the example of this patent are all planes); the conversion of the assembly information into the adjacency matrix of each axis is to obtain the transformation matrix of other planes relative to a certain plane based on the plane, so that the other planes can be expressed in the form of a matrix relationship.

[0053] The feature extraction algorithm uses the Feature Detection function of the Computational Geometry Algorithms Library to extract plane features.

[0054] The alignment operation adopts a method of first coarse matching and then fine matching.

[0055] The assembly sequence planning uses a fitness function including the number of assembly direction changes, the number of assembly tool changes and the assembly tolerance to evaluate each assembly sequence result.

[0056] Embodiment: The present invention is a phased array antenna assembly process planning method based on digital twins. First, the geometric features of the parts are used to extract plane features to segment each plane; the contact surface data and the geometric relationship of each contact surface are retained as a sub-assembly and the point set data of the non-contact surface is deleted to reduce the subsequent calculation amount; coarse alignment and fine alignment are used to match the contact surfaces of the part sub-assembly and the back frame sub-assembly, and an assembly sequence planning algorithm is used to plan the assembly process to achieve overall optimization of assembly accuracy.

[0057] The specific steps of the phased array antenna assembly process planning method based on digital twin are as follows:

[0058] For phased array antennas, manual collection is adopted, and the data of parts including four-in-one oscillator, two-in-one oscillator, grid plate, panel, power divider, and back frame are collected by laser scanner. The scanned data of each part is saved separately as a data file. That is, the phased array antenna components to be assembled are numbered and scanned in turn, and they need to be classified according to the same type of assembly parts, such as two-in-one oscillator, grid plate, four-in-one oscillator, etc. The numbering such as 2-5 represents the fifth assembly part of the two-in-one oscillator; and 3-8 represents the eighth assembly part of the grid plate. Each scanned part is subjected to noise removal and contact surface feature extraction, using existing software functions on the market, commercial software (such as Geomagic Qualify / Geomagic Studio) or commonly used free and open software (such as CloudCompare), or the point cloud processing function of the free and open Computational GeometryAlgorithms Library (CGAL) can be called for operation.

[0059] According to the measured data, according to the geometric features of each four-in-one oscillator, two-in-one oscillator, grid plate, panel, and power divider, the method of extracting plane features is used to segment each plane, and the corresponding geometric relationships are retained for the planes in the plane point set data of each part, and the planes are saved again as a data file.

[0060] The data of the non-contact assembly surfaces of the back frame parts are manually circled and removed (for example, Geomagic software can be used), the corresponding geometric relationships of each assembly station are saved and retained, and the data of the back frame are saved again.

[0061] As shown in Figure 5, the contact surface feature extraction of the two-in-one vibrator is shown. Each contact surface is indicated by a different color and saved as a point cloud set.

[0062] The assembly scheme is determined according to the assembly position of the phased array antenna components and the isothermal curve of the partial differential equation, that is, the temperature value represented by the assembly position in the temperature field is calculated. Figure 6 (a) The outer boundary is set to the same temperature, and the calculated isotherms are marked in red. The solid points are the assembly positions of the components of the phased array antenna. The temperature represented by each assembly position can be determined by its positional relationship between the outer boundary and the isotherm or between the isotherms. In a similar way, the contours of equidistant contour offsets can also be obtained, such as Figure 6 (b) as shown.

[0063] In order to evaluate the proposed optimization algorithm for the assembly of key components of phased array antennas, the collected assembly data of key components are used as examples, and the optimized assembly scheme is given through the aforementioned algorithm.

[0064] First, measurements are performed according to the original assembly plan to obtain the corresponding measurement results, which are the assembly errors of each key component and the corresponding type of design model. The measurement results are divided into 12 levels to obtain the distribution of the measured assembly errors and the number of assembly errors represented by each level, as shown in the figure. Figure 7 Further, using the aforementioned method for calculating the temperature of the assembly position on the contour line, the distribution of assembly errors of the corresponding assembly positions that conform to the contour line law is obtained, as shown in FIG. Figure 7 b).

[0065] Furthermore, by using the method of equidistant offset from the outer boundary to the central area to obtain the contour line, the distribution of assembly errors corresponding to the offset situation can also be obtained, and the results are displayed in Figure 8 (a), and compared with the results proposed by the present invention (such as Figure 8 (b)) shows that the results proposed in this paper have a more natural electromagnetic wave conduction partitioning law.

[0066] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should be regarded as within the scope of protection of the present invention.

Claims

1. A phased array antenna assembly process planning method based on digital twins, characterized in that: Analyze the spatial position of the phased array antenna assembly parts and obtain the subassembly to provide the assembler with a simulation model with assembly errors and contour lines, and intuitively display the assembly prediction results; including the following steps: Step 1: Measure the geometric characteristic data of the phased array antenna assembly components respectively; Step 2: Use the plane feature extraction method to segment each plane of the component; remove the non-contact surface data, retain the assembly contact surface data of each component and the corresponding geometric relationship of the assembly contact surface, and generate the subassembly of the current component; Step 3: Align the contact surface of the current component with its design model; then obtain the assembly error; divide the assembly errors of all assembly positions in the assembly area into gradient levels, and count the number of assembly errors in each level that belong to the same level; Step 4: Abstract the assembly area into an area with the same attribute value on the outer boundary, and obtain the contour trajectory map in the area; Step 5: Calculate the contour lines corresponding to the assembly positions and assign the statistical assembly errors to the corresponding positions; finally, a graph with assembly errors and contour lines is obtained.

2. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The phased array antenna assembly components include: a four-in-one vibrator, a two-in-one vibrator, a grid plate, a panel, and a power divider. The grid plate is used to insert the vibrator, and the panel is used to set the grid plate and the power divider with the inserted vibrator on it to form an assembly area.

3. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The measurement is to sequentially number the phased array antenna components to be assembled, scan and remove noise.

4. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The method for extracting plane features uses a feature extraction Feature Detection algorithm of a Computational Geometry Algorithms Library to extract plane features.

5. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The geometric relationship is a spatial relationship matrix of any two assembly contact surfaces, which is used to obtain another assembly contact surface after a known contact surface is transformed through the spatial relationship matrix.

6. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The spatial relationship matrix is ​​an adjacency matrix of each axis.

7. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The alignment operation includes coarse matching and fine matching. The coarse matching is that the user selects corresponding point pairs with consistent order in the scan data and the mathematical model, and then performs an associated matching operation; the fine matching function is to directly call the ICP algorithm function of PCL to perform the operation.

8. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The isovalue line calculation adopts any one of the isothermal lines in the temperature field, the isovoltage lines in the electromagnetic field or the isodecibel value lines in the sound field.

9. The method for phased array antenna assembly process planning based on digital twin according to claim 1, characterized in that: The denser the contour lines are, the more accurate the assembly accuracy of the corresponding assembly position is, and the higher the overall assembly accuracy is.

Images