Geometric and physical fusion method for predicting assembly deviation of easily-deformed structure

Abstract

The invention provides a geometric and physical fusion method for predicting the assembly deviation of an easy-to-deform structure, and belongs to the technical field of aeronautical engineering manufacturing, and the method comprises the following steps: carrying out the structural decomposition of the easy-to-deform structure, obtaining a space vector point set, and representing and calculating a consistent manufacturing deviation, an assembly positioning deviation and an assembly deformation deviation; constructing an assembly deviation transfer diagram, and mapping the assembly deviation transfer diagram into an assembly state space diagram by using the three multi-source deviations; performing finite element analysis on the assembly precision of the easily-deformed structure by utilizing the assembly state space diagram, and converting the obtained assembly deviation cloud diagram into an assembly deviation contour map; constructing an initial generative adversarial network model, and training and verifying the initial generative adversarial network model by using data in the assembly deviation contour map to obtain a final generative adversarial network model; and inputting the actual deviation image of the easily-deformed structure into the final generative adversarial network model for assembly deviation prediction. The assembly precision and the assembly efficiency of the easy-to-deform structure can be improved.

Description

Technical Field

[0001] The present application relates to the field of aviation engineering manufacturing technology, and in particular to a method for integrating geometry and physics for predicting assembly deviation of deformable structures. Background Art

[0002] With the iterative development of aviation engineering, higher requirements are placed on the size and precision of deformable structures in various aviation products. For example, the size of deformable structures has increased from 1 to 2 meters to 5 to 10 meters, or even larger. At the same time, the assembly precision of deformable structures has changed from millimeter level to sub-millimeter level, or even more microscopic level. In order to meet the requirements of large-size and high-precision assembly, aviation manufacturing companies usually use assembly deviation prediction methods such as repeated adjustment, process compensation, rigidity analysis, and flexibility analysis. However, these traditional assembly deviation prediction methods often have obvious limitations, as follows:

[0003] First, the assembly deviation prediction is carried out by repeated adjustment and process compensation. The prediction reliability is low, the calculation cost is high, and it often relies on repeated adjustment and subsequent process compensation on site. Therefore, it is difficult to meet the requirements of production schedule and quality control.

[0004] Secondly, the rigid analysis method is mainly based on kinematics, and the flexible analysis method mainly relies on statics, but there are incompatibility issues between the two in mathematical thinking and expression systems, which often leads to large deviations between the simulation calculation results and the actual predicted values.

[0005] Thirdly, the easy-to-deform structure has a large size span, weak structural rigidity, and nonlinear factors in the deviation transmission, which makes the assembly deviation field too complex and difficult to accurately predict;

[0006] Finally, traditional assembly deviation prediction methods are often inefficient in processing large-scale data and identifying complex patterns, which limits their application in modern aviation manufacturing.

[0007] Therefore, it is necessary to propose a solution to improve one or more problems existing in the above-mentioned related technical solutions.

[0008] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0009] The embodiment of the present application provides a geometric and physical fusion method for predicting assembly deviation of a deformable structure, the method comprising the following steps:

[0010] Decomposing the deformable structure, extracting all key features of each assembly level, and discretizing all the key features of the part level into a set of space vector points;

[0011] According to the space vector point set, the part warpage deviation and the part fluctuation deviation are used to characterize the part manufacturing deviation, and the part manufacturing deviation that meets the tolerance specification is defined as the consistent manufacturing deviation;

[0012] The deterministic assembly method is used to characterize the assembly positioning deviation, and the finite element analysis is used to calculate the assembly deformation deviation;

[0013] According to the structural decomposition, the assembly information of the deformable structure is processed to construct an assembly deviation transfer diagram; and the assembly deviation transfer diagram is mapped into an assembly state space diagram by using the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation;

[0014] Performing finite element analysis on the assembly accuracy of the deformable structure using the assembly state space diagram to obtain an assembly deviation cloud diagram, and converting the assembly deviation cloud diagram into an assembly deviation contour diagram, wherein the assembly deviation contour diagram includes a training data set and a test data set;

[0015] Constructing an initial generative adversarial network model, training the initial generative adversarial network model using the training data set, and verifying the trained initial generative adversarial network model using the test data set to obtain a final generative adversarial network model;

[0016] The actual deviation image of the deformable structure is input into the final generative adversarial network model to perform assembly deviation prediction and generate an assembly deviation prediction image.

[0017] In an exemplary embodiment of the present application, the types of the assembly levels include component level, assembly level and part level; the space vector point set includes: size information, shape information, position information and waviness information of the part.

[0018] In an exemplary embodiment of the present application, the step of characterizing the part manufacturing deviation by using the part warpage deviation and the part fluctuation deviation according to the space vector point set, and defining the part manufacturing deviation that meets the tolerance specification as the consistent manufacturing deviation includes:

[0019] Superimposing the part warpage deviation and the part fluctuation deviation on the spatial vector point concentration, and performing deviation statistics on all sampling points on the part surface;

[0020] Characterizing the result of the deviation statistics as the part manufacturing deviation using geometric covariance, wherein the part manufacturing deviation includes the part warpage deviation constructed using Legendre polynomials and the part fluctuation deviation constructed using sine polynomials;

[0021] Check whether the manufacturing deviation of the parts meets the tolerance specification;

[0022] If the part manufacturing deviation meets the tolerance specification, the part manufacturing deviation is defined as the consistent manufacturing deviation; if the part manufacturing deviation does not meet the tolerance specification, the deviation statistics process is iterated until the constructed part manufacturing deviation meets the tolerance specification.

[0023] In an exemplary embodiment of the present application, the expression of the consistent manufacturing deviation is:

[0024] (1)

[0025] in, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The ideal value of the part manufacturing deviation at each assembly station, represents the geometric covariance of the part surface, Represents the normal direction of a discrete vector point in a space vector point set;

[0026] The expression of the geometric covariance of the part surface is:

[0027] (2)

[0028] in, The weight coefficient representing the part warpage deviation, , The weight coefficient representing the part fluctuation deviation, , Indicates the range of part warpage deviation. Indicates the range of part fluctuation deviation, Indicates the warpage deviation of the part. , represents the first transformation matrix, , Indicates Legendre polynomial of order, represents the Legendre polynomial Step, represents the order of Legendre polynomial, Represents the number of sampling points on the part surface, represents the Legendre polynomial The weight coefficient of the order, The surface of the part The index of the sampling points, The surface of the part The index of the sampling points, The variance matrix representing the deviation variation of the sampling points, Indicates part fluctuation deviation, , represents the second transformation matrix, , , represents the first The weight coefficient of the frequency curve of order, The sine polynomial The weight coefficient of the order, The sine polynomial The frequency of the order, Indicates the column to which the sampling points on the part surface belong, Indicates the row to which the sampling point on the part surface belongs, The sine polynomial Step, Represents the order of the sine polynomial.

[0029] In an exemplary embodiment of the present application, the expression of the assembly positioning deviation is:

[0030] (3)

[0031] in, Indicates that the deformable structure is Assembly positioning deviation of each assembly station, represents the Jacobian matrix, Indicates that the deformable structure The transfer matrix related to the coordinates and control direction of each assembly station, Indicates Deviations caused by each assembly station during the assembly process;

[0032] The expression of the assembly deformation deviation is:

[0033] (4)

[0034] in, Indicates that the deformable structure is The assembly stations are subject to external loads The assembly deformation deviation caused by the action, Indicates external load The mapping function of the deformation of the deformable structure when acting, Indicates that the deformable structure is The overall displacement generated during the positioning of each assembly station, Indicates that the deformable structure is Flexible deformation caused by clamping and riveting at each assembly station, Indicates that the deformable structure is Springback deformation caused by the release of the tooling or fixture at an assembly station.

[0035] In an exemplary embodiment of the present application, the steps of processing the assembly information of the deformable structure according to the structural decomposition to construct an assembly deviation transfer map; and mapping the assembly deviation transfer map into an assembly state space map by using the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation include:

[0036] According to the structural decomposition, the easily deformable structure is sequentially divided into assembly units, the assembly sequence is adjusted, the assembly reference is identified, and the matching constraint is defined, so as to construct the assembly deviation transfer diagram;

[0037] Mapping all part nodes in the assembly deviation transfer graph to corresponding state nodes in the assembly state space graph;

[0038] Mapping all the tooling constraint nodes in the assembly deviation transfer diagram to input variables of the corresponding state nodes in the assembly state space diagram;

[0039] Mapping all the matching constraint nodes in the assembly deviation transfer graph to the transfer relations between the corresponding state nodes in the assembly state space graph;

[0040] Among them, the mapping of various nodes in the assembly deviation transfer graph is carried out in a top-down and left-to-right order, and the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation are respectively integrated into the corresponding nodes to construct the assembly state space graph.

[0041] In an exemplary embodiment of the present application, the expression of the assembly deviation transfer diagram is:

[0042] (5)

[0043] in, represents the assembly deviation transfer diagram, Represents the collection of part nodes in the assembly process, , Represents the set of tooling constraint nodes in the assembly process, , Represents the set of mating constraint nodes in the assembly process. , Represents the set of node transfer directions during the assembly process, ;

[0044] The expression of the assembly state space diagram is:

[0045] (6)

[0046] in, Indicates that the deformable structure is The assembly accuracy after each assembly station is completed, Indicates that the deformable structure is The assembly accuracy of the previous assembly station after the assembly station completes the positioning, represents the first deviation coefficient matrix, represents the second deviation coefficient matrix, represents the third deviation coefficient matrix, represents the fourth deviation coefficient matrix, represents the fifth deviation coefficient matrix, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The assembly deviation obtained by measuring each assembly station is Indicates that the deformable structure is The measurement error caused by the measurement equipment at each assembly station.

[0047] In an exemplary embodiment of the present application, the step of performing finite element analysis on the assembly accuracy of the deformable structure using the assembly state space diagram to obtain an assembly deviation cloud map, and converting the assembly deviation cloud map into an assembly deviation contour map includes:

[0048] In finite element analysis software, using the assembly state space diagram to perform finite element analysis on the assembly accuracy of the deformable structure to obtain the assembly deviation cloud diagram;

[0049] Converting the assembly deviation cloud map into the assembly deviation contour map using a cubic spline interpolation algorithm, a deviation normalization algorithm, and a color mapping algorithm in sequence;

[0050] The data set reflected by the assembly deviation contour map is divided into the training data set and the test data set, and the ratio of the training data set to the test data set is 4:1.

[0051] In an exemplary embodiment of the present application, the steps of constructing an initial generative adversarial network model, training the initial generative adversarial network model using the training data set, and verifying the trained initial generative adversarial network model using the test data set to obtain a final generative adversarial network model include:

[0052] The initial generative adversarial network model includes a finite element module, a generator and a discriminator that are communicatively connected to each other;

[0053] In the process of training and verifying the initial generative adversarial network model to obtain the final generative adversarial network model, a generative adversarial loss function is used. The generative adversarial loss function includes Loss function and feature matching loss function;

[0054] The expression of the generative adversarial loss function is:

[0055] (7)

[0056] in, represents the generative adversarial loss function, represents a generator, represents the discriminator, represents the type of discriminator, express Loss function, , represents the discriminator’s expectation of the difference between the input image and the true image, represents the input image, represents the real image corresponding to the input image, represents a composite image, represents the discriminator’s expectation of the difference between the input image and the synthesized image, Indicates the generated image, represents the probability that the real image is judged as the true value by the discriminator, represents the probability that the generated image is judged as the true value by the discriminator, represents the feature matching loss function, , The discriminator Layer extractor, represents the total number of layers of the discriminator, The discriminator The number of elements in the layer, represents a random variable, represents a given sample generated by a random variable, represents the expected value of the difference between a given sample and the input image, express The weight of the loss function and the feature matching loss function, Represents the norm of 1.

[0057] Beneficial effects:

[0058] The present application provides a geometric and physical fusion method for predicting assembly deviation of deformable structures, which has at least the following beneficial effects:

[0059] (1) This application uses part warpage deviation and part fluctuation deviation to characterize part manufacturing deviation, defines consistent manufacturing deviation, adopts deterministic assembly method to characterize assembly positioning deviation, and uses finite element analysis to calculate assembly deformation deviation. By introducing these three multi-source deviations, aviation manufacturing companies do not need to frequently perform on-site repeated calibration and subsequent process compensation, thereby improving the first-time success rate of product assembly, saving time and resources wasted due to repeated calibration, and thus reducing production costs;

[0060] (2) This application constructs an assembly deviation transfer map and uses multi-source deviations to map the assembly deviation transfer map into an assembly state space map, thereby achieving the simultaneous fusion of geometric information and physical information in the assembly deviation prediction process, thereby improving the assembly accuracy of easily deformable structures;

[0061] (3) This application uses the assembly state space diagram to perform finite element analysis on the assembly accuracy of the deformable structure to obtain an assembly deviation cloud map, and converts the assembly deviation cloud map into an assembly deviation contour map, thereby integrating the advantages of rigid analysis and flexible analysis, overcoming the calculation deviation problem existing in the traditional method, and making the assembly deviation prediction result closer to the actual situation, thereby meeting the aviation manufacturing enterprises' requirements for sub-millimeter or even more microscopic accuracy;

[0062] (4) This application constructs an initial generative adversarial network model, trains and verifies the initial generative adversarial network model, and obtains a final generative adversarial network model as the final model for assembly deviation prediction of actual deviation images of deformable structures. This allows the application to efficiently handle complex assembly deviation prediction tasks. Compared with traditional methods, this method can significantly improve the speed and accuracy of assembly deviation prediction, thereby meeting the needs of efficient production for aviation manufacturing companies. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0064] Figure 1A schematic diagram showing the steps of a geometric and physical fusion method for predicting assembly deviation of a deformable structure in an exemplary embodiment of the present application is shown;

[0065] Figure 2 A schematic flow chart of a geometric and physical fusion method for predicting assembly deviation of a deformable structure in an exemplary embodiment of the present application is shown;

[0066] Figure 3 A schematic diagram showing the structure of an aviation wing box in an exemplary embodiment of the present application is shown;

[0067] Figure 4 A schematic diagram showing structural decomposition of a deformable structure in an exemplary embodiment of the present application;

[0068] Figure 5 A schematic diagram showing a process of constructing a consistent manufacturing deviation in an exemplary embodiment of the present application;

[0069] Figure 6 A schematic diagram showing a process for constructing consistent manufacturing deviations of aviation thin-walled structures in an exemplary embodiment of the present application;

[0070] Figure 7 A schematic diagram showing mapping an assembly deviation transfer diagram into an assembly state space diagram in an exemplary embodiment of the present application is shown;

[0071] Figure 8 A schematic diagram showing an assembly deviation cloud diagram in an exemplary embodiment of the present application;

[0072] Fig. 9 A schematic diagram showing an assembly deviation contour map in an exemplary embodiment of the present application;

[0073] Fig.10 A schematic diagram showing the structure of an initial generative adversarial network model in an exemplary embodiment of the present application is shown;

[0074] Fig.11 A schematic diagram showing the structure of a generator in an exemplary embodiment of the present application is shown;

[0075] Fig.12 A schematic diagram showing the structure of an identifier in an exemplary embodiment of the present application is shown;

[0076] Fig.13 A schematic diagram showing training of an initial generative adversarial network model in an exemplary embodiment of the present application is shown;

[0077] Fig.14 A schematic diagram showing an assembly deviation prediction image generated in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0078] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0079] In addition, the accompanying drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0080] To this end, this example embodiment provides a geometric and physical fusion method for predicting assembly deviation of deformable structures, such as Figure 1 and Figure 2 As shown, the method may include the following steps:

[0081] Step S101: Decompose the deformable structure, extract all key features of each assembly level, and discretize all key features of the part level into a space vector point set;

[0082] Step S102: according to the space vector point set, the part warpage deviation and the part fluctuation deviation are used to characterize the part manufacturing deviation, and the part manufacturing deviation that meets the tolerance specification is defined as the consistent manufacturing deviation;

[0083] Step S103: using a deterministic assembly method to characterize assembly positioning deviation, and using finite element analysis to calculate assembly deformation deviation;

[0084] Step S104: according to the structural decomposition, the assembly information of the easily deformable structure is processed to construct an assembly deviation transfer diagram, and the assembly deviation transfer diagram is mapped into an assembly state space diagram by using the consistent manufacturing deviation, assembly positioning deviation and assembly deformation deviation;

[0085] Step S105: using the assembly state space diagram to perform finite element analysis on the assembly accuracy of the easily deformable structure to obtain an assembly deviation cloud map, and converting the assembly deviation cloud map into an assembly deviation contour map, wherein the assembly deviation contour map includes a training data set and a test data set;

[0086] Step S106: constructing an initial generative adversarial network model, training the initial generative adversarial network model using a training data set, and verifying the trained initial generative adversarial network model using a test data set to obtain a final generative adversarial network model;

[0087] Step S107: inputting the actual deviation image of the deformable structure into the final generative adversarial network model to perform assembly deviation prediction and generate an assembly deviation prediction image.

[0088] Figure 2 The FEM (Finite Element Method, FEM) in it stands for finite element.

[0089] The embodiment of the present application proposes a geometric and physical fusion method for predicting assembly deviation of deformable structures, which has at least the following beneficial effects:

[0090] (1) In the embodiment of the present application, part warpage deviation and part fluctuation deviation are used to characterize part manufacturing deviation, consistent manufacturing deviation is defined, a deterministic assembly method is used to characterize assembly positioning deviation, and finite element analysis is used to calculate assembly deformation deviation. By introducing these three multi-source deviations, aviation manufacturing companies do not need to frequently perform on-site repeated calibration and subsequent process compensation, thereby improving the first-time success rate of product assembly, saving time and resources wasted due to repeated calibration, and thus reducing production costs;

[0091] (2) In the embodiment of the present application, an assembly deviation transfer map is constructed, and the assembly deviation transfer map is mapped into an assembly state space map using multi-source deviations, thereby achieving the simultaneous fusion of geometric information and physical information in the assembly deviation prediction process, thereby improving the assembly accuracy of easily deformable structures;

[0092] (3) In the embodiment of the present application, the assembly state space diagram is used to perform finite element analysis on the assembly accuracy of the deformable structure to obtain an assembly deviation cloud map, and the assembly deviation cloud map is converted into an assembly deviation contour map, thereby integrating the advantages of rigid analysis and flexible analysis, overcoming the calculation deviation problem existing in the traditional method, and making the assembly deviation prediction result closer to the actual situation, thereby meeting the sub-millimeter level or even more microscopic level accuracy requirements of aviation manufacturing companies;

[0093] (4) In the embodiments of the present application, an initial generative adversarial network model is constructed, and the initial generative adversarial network model is trained and verified to obtain a final generative adversarial network model, which serves as the final model for predicting assembly deviations of actual deviation images of deformable structures. This makes it possible to efficiently handle complex assembly deviation prediction tasks. Compared with traditional methods, this method can significantly improve the speed and accuracy of assembly deviation prediction, thereby meeting the needs of efficient production of aviation manufacturing companies.

[0094] Next, a geometric and physical fusion method for predicting assembly deviation of a deformable structure proposed in this example embodiment will be described in more detail.

[0095] In step S101 of this embodiment, the deformable structure is structurally decomposed, all key features of each assembly level are extracted respectively, and all key features of the part level are discretized into a space vector point set.

[0096] Furthermore, if Figure 3 As shown, in this embodiment, an aircraft wing box is used as an example for explanation. In this embodiment, the structure is decomposed in a top-down order, and the aircraft wing box is decomposed into components, assemblies, parts, key features and space vector point sets. Figure 3 a shows the decomposition of the aircraft wing box into components and assemblies, that is, into an upper wall panel, a front beam, a rib, a rear beam and a lower wall panel; Figure 3 FIG. 2 b shows that the upper wall panel is further decomposed into parts including a stopper, a suction cup, a positioning device and a tooling frame.

[0097] Furthermore, if Figure 4 As shown in the figure, the types of assembly levels include component level, assembly level and part level. The space vector point set includes the size information, shape information, position information and waviness information of the part, that is, the space vector point set not only includes the size deviation and shape and position deviation of the part on the macro scale, but also includes the waviness information on the meso scale.

[0098] In step S102 of this embodiment, according to the space vector point set, the part warpage deviation and the part fluctuation deviation are used to characterize the part manufacturing deviation, and the part manufacturing deviation that meets the tolerance specification is defined as the consistent manufacturing deviation. Figure 5 and Figure 6 As shown, step S102 in this embodiment may include the following sub-steps:

[0099] Sub-step S1021: superimposing part warpage deviation and part fluctuation deviation on the spatial vector point set, and performing deviation statistics on all sampling points on the part surface.

[0100] Sub-step S1022: using geometric covariance to characterize the result of deviation statistics as part manufacturing deviation, where the part manufacturing deviation includes part warpage deviation constructed using Legendre polynomials and part fluctuation deviation constructed using sine polynomials.

[0101] Furthermore, in this embodiment, according to the length of the component and the deviation wavelength in the space vector point set, the deformable structure is decomposed into three basic modes: surface warping ( ), surface fluctuations ( ) and surface roughness ( ), It represents the ratio of the part length to the deviation wavelength. In the process of predicting the assembly deviation of the deformable structure, surface warpage and surface fluctuation play a major role, while the effect of surface roughness on the prediction of assembly deviation is very small and can be ignored. Therefore, in this embodiment, only the introduction of part warpage deviation and part fluctuation deviation is considered.

[0102] Furthermore, in this embodiment, the upper wall panel of the wing box of an aircraft wing is modeled.

[0103] Sub-step S1023: Check whether the part manufacturing deviation meets the tolerance specification.

[0104] Furthermore, according to the different production stages of the product, the inspection conducted in the design stage needs to see whether the manufacturing deviation of the parts meets the tolerance specifications; the inspection conducted in the product manufacturing stage needs to see whether the manufacturing deviation of the parts meets the measurement data.

[0105] Sub-step S1024: If the part manufacturing deviation meets the tolerance specification, the part manufacturing deviation is defined as a consistent manufacturing deviation; if the part manufacturing deviation does not meet the tolerance specification, the deviation statistics process is iterated until the constructed part manufacturing deviation meets the tolerance specification.

[0106] Furthermore, the expression of the consistent manufacturing deviation is:

[0107] (1)

[0108] in, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The ideal value of the part manufacturing deviation at each assembly station, represents the geometric covariance of the part surface, Represents the normal direction of a discrete vector point in a space vector point set.

[0109] Furthermore, after the part manufacturing deviation is constructed, it needs to be inspected to ensure that the consistent manufacturing deviation meets the tolerance specification. Here, meeting the tolerance specification means that the part manufacturing deviation that meets this requirement can be defined as the consistent manufacturing deviation within the tolerance range of the key features, that is, , Indicates the upper limit of the tolerance range, Indicates the lower limit of the tolerance range.

[0110] Furthermore, the expression of the geometric covariance of the part surface is:

[0111] (2)

[0112] in, The weight coefficient representing the part warpage deviation, , The weight coefficient representing the part fluctuation deviation, , Indicates the range of part warpage deviation. Indicates the range of part fluctuation deviation, Indicates the warpage deviation of the part. , represents the first transformation matrix, , Indicates Legendre polynomial of order, represents the Legendre polynomial Step, represents the order of Legendre polynomial, Represents the number of sampling points on the part surface, represents the Legendre polynomial The weight coefficient of the order, The surface of the part The index of the sampling points, The surface of the part The index of the sampling points, The variance matrix representing the deviation variation of the sampling points, Indicates part fluctuation deviation, , represents the second transformation matrix, , , represents the first The weight coefficient of the frequency curve of order, The sine polynomial The weight coefficient of the order, The sine polynomial The frequency of the order, Indicates the column to which the sampling points on the part surface belong, Indicates the row to which the sampling point on the part surface belongs, The sine polynomial Step, Represents the order of the sine polynomial.

[0113] Furthermore, in this embodiment, part warpage deviation and part fluctuation deviation are used to characterize part manufacturing deviation, consistent manufacturing deviation is defined, a deterministic assembly method is used to characterize assembly positioning deviation, and finite element analysis is used to calculate assembly deformation deviation. By introducing these three multi-source deviations, aviation manufacturing companies do not need to frequently perform on-site repeated calibration and subsequent process compensation, thereby improving the first-time success rate of product assembly, saving time and resources caused by repeated calibration, and thereby reducing production costs.

[0114] In step S103 of this embodiment, a deterministic assembly method is used to characterize the assembly positioning deviation, and a finite element analysis is used to calculate the assembly deformation deviation.

[0115] Furthermore, the expression of assembly positioning deviation is:

[0116] (3)

[0117] in, Indicates that the deformable structure is Assembly positioning deviation of each assembly station, represents the Jacobian matrix, Indicates that the deformable structure The transfer matrix related to the coordinates and control direction of each assembly station, Indicates The deviations caused by each assembly station during the assembly process.

[0118] Furthermore, the expression of assembly deformation deviation is:

[0119] (4)

[0120] in, Indicates that the deformable structure is The assembly stations are subject to external loads The assembly deformation deviation caused by the action, Indicates external load The mapping function of the deformation of the deformable structure when acting, Indicates that the deformable structure is The overall displacement generated during the positioning of each assembly station, Indicates that the deformable structure is Flexible deformation caused by clamping and riveting at each assembly station, Indicates that the deformable structure is Springback deformation caused by the release of the tooling or fixture at an assembly station.

[0121] In step S104 of this embodiment, Figure 7As shown, according to the structural decomposition, the assembly information of the deformable structure is processed to construct an assembly deviation transfer diagram; and the assembly deviation transfer diagram is mapped into an assembly state space diagram using the consistent manufacturing deviation, assembly positioning deviation and assembly deformation deviation. In this embodiment, step S104 may include the following sub-steps:

[0122] Sub-step S1041: According to the structural decomposition, the easily deformable structure is sequentially divided into assembly units, the assembly sequence is adjusted, the assembly reference is identified, and the matching constraint is defined to construct an assembly deviation transfer diagram.

[0123] Sub-step S1042: Map all part nodes in the assembly deviation transfer graph to corresponding state nodes in the assembly state space graph.

[0124] Sub-step S1043: Mapping all tooling constraint nodes in the assembly deviation transfer graph to input variables of corresponding state nodes in the assembly state space graph.

[0125] Sub-step S1044: Map all the matching constraint nodes in the assembly deviation transfer graph to the transfer relations between the corresponding state nodes in the assembly state space graph.

[0126] Among them, the mapping of various nodes in the assembly deviation transfer diagram is carried out in a top-down and left-to-right order, and the consistent manufacturing deviation, assembly positioning deviation and assembly deformation deviation are respectively integrated into the corresponding nodes to construct the assembly state space diagram.

[0127] Furthermore, the expression of the assembly deviation transfer diagram is:

[0128] (5)

[0129] in, represents the assembly deviation transfer diagram, Represents the collection of part nodes in the assembly process, , Represents the set of tooling constraint nodes in the assembly process, , Represents the set of mating constraint nodes in the assembly process. , Represents the node transfer direction set during the assembly process, .

[0130] Furthermore, based on the above product assembly hierarchical relationship, combined with the assembly process information and the interactive relationship in the assembly process, an assembly deviation transfer diagram is constructed. The assembly deviation transfer diagram is a tree graph structure that reflects the assembly deviation transfer, such as Figure 7As shown, the part nodes in the tree graph structure are mapped to the state nodes in the assembly state space graph, the tooling constraint nodes in the tree graph structure are mapped to the input variables of the corresponding state nodes in the assembly state space graph, and the matching constraint nodes in the tree graph structure are mapped to the transfer relationship between the corresponding state nodes in the assembly state space graph.

[0131] Furthermore, the expression of the assembly state space diagram is:

[0132] (6)

[0133] in, Indicates that the deformable structure is The assembly accuracy after each assembly station completes positioning, Indicates that the deformable structure is The assembly accuracy of the previous assembly station after the assembly station completes the positioning, represents the first deviation coefficient matrix, represents the second deviation coefficient matrix, represents the third deviation coefficient matrix, represents the fourth deviation coefficient matrix, represents the fifth deviation coefficient matrix, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The assembly deviation obtained by measuring each assembly station is Indicates that the deformable structure is The measurement error caused by the measurement equipment at each assembly station.

[0134] Furthermore, the assembly accuracy here refers to the The accumulated assembly deviation after each assembly station completes positioning.

[0135] Furthermore, in this embodiment, by constructing an assembly deviation transfer graph and mapping the assembly deviation transfer graph into an assembly state space graph using multi-source deviations, it is possible to simultaneously integrate geometric information and physical information in the assembly deviation prediction process, thereby improving the assembly accuracy of deformable structures. The multi-source deviations here include part manufacturing deviations, assembly positioning deviations, and assembly deformation deviations.

[0136] In step S105 of this embodiment, finite element analysis is performed on the assembly accuracy of the deformable structure using the assembly state space diagram to obtain an assembly deviation cloud map, and the assembly deviation cloud map is converted into an assembly deviation contour map, and the assembly deviation contour map includes a training data set and a test data set. Step S105 of this embodiment may include the following steps:

[0137] Sub-step S1051: Figure 8 As shown, in the finite element analysis software, the assembly state space diagram is used to perform finite element analysis on the assembly accuracy of the deformable structure to obtain the assembly deviation cloud diagram.

[0138] Sub-step S1052: Fig. 9 As shown, the assembly deviation cloud map is converted into an assembly deviation contour map by using the cubic spline interpolation algorithm, the deviation normalization algorithm and the color mapping algorithm in sequence;

[0139] In this embodiment, the resolution of the generated assembly deviation contour map is .

[0140] Sub-step S1053: Divide the data set reflected by the assembly deviation contour map into a training data set and a test data set, with the ratio of the training data set to the test data set being 4:1.

[0141] Furthermore, in this embodiment, the data set reflected by the assembly deviation contour map is divided into 2000 sets of training data sets and 500 sets of testing data sets.

[0142] Furthermore, in this embodiment, the assembly state space diagram is used to perform finite element analysis on the assembly accuracy of the easily deformable structure to obtain an assembly deviation cloud map, and the assembly deviation cloud map is converted into an assembly deviation contour map, thereby integrating the advantages of rigid analysis and flexible analysis, overcoming the calculation deviation problem existing in the traditional method, and making the results of assembly deviation prediction closer to the actual situation, thereby meeting the aviation manufacturing companies' requirements for sub-millimeter or even more microscopic levels of accuracy.

[0143] In step S106 of this embodiment, an initial generative adversarial network model is constructed, the initial generative adversarial network model is trained using a training data set, and the trained initial generative adversarial network model is verified using a test data set to obtain a final generative adversarial network model. Step S106 of this embodiment may include the following sub-steps:

[0144] Sub-step S1061: The constructed initial generative adversarial network model includes a finite element module, a generator and a discriminator that are communicatively connected to each other.

[0145] Furthermore, if Fig.10 As shown, Fig.10 The overall structure of the initial generative adversarial network model is shown. It can be seen that the finite element module is responsible for performing finite element analysis on the assembly accuracy of the deformable structure and generating an assembly deviation cloud map.

[0146] Furthermore, if Fig.11As shown in the figure, the generator adopts a coarse-to-fine design strategy. The coarse generator generates a low-resolution image to ensure the continuity of the image while capturing the overall structure and key features of the image. Then, the fine generator is used to focus on enhancing the subtle local key features of the high-resolution image to improve the detail quality and realism of the image. This process is connected through multiple residual blocks to enhance the transmission of key features, improve training stability, and maintain the information integrity of key features. The network architecture includes convolutional layers for extracting initial key features, residual blocks for processing deep key features, and transposed convolutional layers for image upsampling. In terms of working mechanism, first, the image is downsampled by 2 times in the fine generator, then a low-resolution image is generated based on the downsampled image on the coarse generator, and finally, the low-resolution image is sent back to the fine generator and fused with the 2-times downsampled image at the pixel level to generate a high-quality image, thereby achieving efficient generation from low-resolution to high-resolution images.

[0147] Furthermore, if Fig.12 As shown in the figure, the designed discriminator adopts a multi-scale architecture to evaluate the authenticity of high-resolution images. It contains two-scale discriminators, which process images of original resolution and downsampled (reduced resolution by 2 times) respectively, so that the model can analyze details at different levels, focusing on the authenticity of image details while evaluating global consistency. The input includes "fake" images synthesized by the generator and "real" images generated by the finite element module, which are fed into discriminators of different scales to extract key features and train the model by calculating element-by-element loss. This design effectively avoids the problems of overfitting and increased storage space caused by deepening the network or increasing the convolution kernel.

[0148] Sub-step S1062: Fig.13 As shown in Figure 2, the generation adversarial loss function is used in the process of training and verifying the initial generation adversarial network model to obtain the final generation adversarial network model. The generation adversarial loss function includes Loss function and feature matching loss function.

[0149] Furthermore, the initial generative adversarial network model mentioned in this embodiment is a High Resolution-Precision Generative Adversarial Networks (HiRes-PreciGAN) model, which consists of Fig.13You can see the changes in the loss of the generator and discriminator during the training of the HiRes-PreciGAN model. As the number of training rounds increases, the loss of the generator decreases rapidly and stabilizes, indicating that the generator is constantly learning and improving the quality of the generated images. At the same time, the loss of the discriminator remains at a low level, showing its stable high discrimination ability. Fig.13 The arrows and example images in the figure also highlight how the quality of the generated images gradually improves as the training progresses, from the initial blurry to the final clear and structurally accurate. This demonstrates the ability of the HiRes-PreciGAN model to deal with warping and fluctuation deviations, as well as its effectiveness in image generation tasks.

[0150] Furthermore, the expression of the generated adversarial loss function is:

[0151] (7)

[0152] in, represents the generative adversarial loss function, represents a generator, represents the discriminator, represents the type of discriminator, express Loss function, , represents the discriminator’s expectation of the difference between the input image and the true image, represents the input image, represents the real image corresponding to the input image, represents a composite image, represents the discriminator’s expectation of the difference between the input image and the synthesized image, Indicates the generated image, represents the probability that the real image is judged as the true value by the discriminator, represents the probability that the generated image is judged as the true value by the discriminator, represents the feature matching loss function, , The discriminator Layer extractor, represents the total number of layers of the discriminator, The discriminator The number of elements in the layer, represents a random variable, represents a given sample generated by a random variable, represents the expected value of the difference between a given sample and the input image, express The weight of the loss function and the feature matching loss function, Represents the norm of 1.

[0153] Further, The loss function is beneficial to recover the low-frequency part of the image; the feature matching loss function is used to reconstruct the high-frequency details of the image.

[0154] Further, The weight of loss function and feature matching loss function It can be fine-tuned to achieve the best image quality according to the specific requirements of the composite image.

[0155] Furthermore, in this embodiment, by constructing an initial generative adversarial network model, training and verifying the initial generative adversarial network model, a final generative adversarial network model is obtained, which serves as the final model for predicting assembly deviations of actual deviation images of deformable structures. This makes it possible to efficiently handle complex assembly deviation prediction tasks. Compared with traditional methods, the speed and accuracy of assembly deviation prediction are significantly improved, meeting the needs of efficient production of aviation manufacturing companies.

[0156] In step S107 of this embodiment, if Fig.14 As shown, the actual deviation image of the deformable structure is input into the final generative adversarial network model for assembly deviation prediction to generate an assembly deviation prediction image.

[0157] Further, by Fig.14 It can be seen that after obtaining the final generative adversarial network model, the actual assembly deviation contour map of any deformable structure is input into the final generative adversarial network model for assembly deviation prediction, and the following can be obtained: Fig.14 The assembly deviation prediction image is shown in .

[0158] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0159] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification.

[0160] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and these modifications or substitutions should be included in the protection scope of the present application.

[0161] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application.

Claims

1. A geometric and physical fusion method for predicting assembly deviation of deformable structures, characterized in that: The method comprises the following steps: Decomposing the deformable structure, extracting all key features of each assembly level, and discretizing all the key features of the part level into a set of space vector points; According to the space vector point set, the part warpage deviation and the part fluctuation deviation are used to characterize the part manufacturing deviation, and the part manufacturing deviation that meets the tolerance specification is defined as the consistent manufacturing deviation; The deterministic assembly method is used to characterize the assembly positioning deviation, and the finite element analysis is used to calculate the assembly deformation deviation; According to the structural decomposition, the assembly information of the deformable structure is processed to construct an assembly deviation transfer map; and the assembly deviation transfer map is mapped into an assembly state space map by using the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation; Performing finite element analysis on the assembly accuracy of the deformable structure using the assembly state space diagram to obtain an assembly deviation cloud diagram, and converting the assembly deviation cloud diagram into an assembly deviation contour diagram, wherein the assembly deviation contour diagram includes a training data set and a test data set; Constructing an initial generative adversarial network model, training the initial generative adversarial network model using the training data set, and verifying the trained initial generative adversarial network model using the test data set to obtain a final generative adversarial network model; The actual deviation image of the deformable structure is input into the final generative adversarial network model to perform assembly deviation prediction and generate an assembly deviation prediction image.

2. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 1 is characterized in that: The types of the assembly level include component level, assembly level and part level; The space vector point set includes: size information, shape information, position information and waviness information of the part.

3. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 1 is characterized in that: The step of characterizing the part manufacturing deviation by using the part warpage deviation and the part fluctuation deviation according to the space vector point set, and defining the part manufacturing deviation that meets the tolerance specification as the consistent manufacturing deviation comprises: Superimposing the part warpage deviation and the part fluctuation deviation on the spatial vector point concentration, and performing deviation statistics on all sampling points on the part surface; Characterizing the result of the deviation statistics as the part manufacturing deviation using geometric covariance, wherein the part manufacturing deviation includes the part warpage deviation constructed using Legendre polynomials and the part fluctuation deviation constructed using sine polynomials; Check whether the manufacturing deviation of the parts meets the tolerance specification; If the part manufacturing deviation meets the tolerance specification, the part manufacturing deviation is defined as the consistent manufacturing deviation; if the part manufacturing deviation does not meet the tolerance specification, the deviation statistics process is iterated until the constructed part manufacturing deviation meets the tolerance specification.

4. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 3 is characterized in that: The expression of the consistent manufacturing deviation is: (1) in, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The ideal value of the part manufacturing deviation at each assembly station, represents the geometric covariance of the part surface, Represents the normal direction of a discrete vector point in a space vector point set; The expression of the geometric covariance of the part surface is: (2) in, The weight coefficient representing the part warpage deviation, , The weight coefficient representing the part fluctuation deviation, , Indicates the range of part warpage deviation. Indicates the range of part fluctuation deviation, Indicates the warpage deviation of the part. , represents the first transformation matrix, , Indicates Legendre polynomial of order, represents the Legendre polynomial Step, represents the order of Legendre polynomial, Represents the number of sampling points on the part surface, represents the Legendre polynomial The weight coefficient of the order, The surface of the part The index of the sampling points, The surface of the part The index of the sampling points, The variance matrix representing the deviation variation of the sampling points, Indicates part fluctuation deviation, , represents the second transformation matrix, , , represents the first The weight coefficient of the frequency curve of order, The sine polynomial The weight coefficient of the order, The sine polynomial The frequency of the order, Indicates the column to which the sampling points on the part surface belong, Indicates the row to which the sampling point on the part surface belongs, The sine polynomial Step, Represents the order of the sine polynomial.

5. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 1 is characterized in that: The expression of the assembly positioning deviation is: (3) in, Indicates that the deformable structure is Assembly positioning deviation of each assembly station, represents the Jacobian matrix, Denotes the deformable structure in The transfer matrix related to the coordinates and control direction of each assembly station, Indicates Deviations caused by each assembly station during the assembly process; The expression of the assembly deformation deviation is: (4) in, Indicates that the deformable structure is The assembly stations are subject to external loads The assembly deformation deviation caused by the action, Indicates external load The mapping function of the deformation of the deformable structure when acting, Indicates that the deformable structure is The overall displacement generated during the positioning of each assembly station, Indicates that the deformable structure is Flexible deformation caused by clamping and riveting at each assembly station, Indicates that the deformable structure is Springback deformation caused by the release of the tooling or fixture at an assembly station.

6. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 5 is characterized in that: The steps of processing the assembly information of the deformable structure according to the structural decomposition to construct an assembly deviation transfer map; and mapping the assembly deviation transfer map into an assembly state space map by using the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation include: According to the structural decomposition, the easily deformable structure is sequentially divided into assembly units, the assembly sequence is adjusted, the assembly reference is identified, and the matching constraint is defined, so as to construct the assembly deviation transfer diagram; Mapping all part nodes in the assembly deviation transfer graph to corresponding state nodes in the assembly state space graph; Mapping all the tooling constraint nodes in the assembly deviation transfer diagram to input variables of the corresponding state nodes in the assembly state space diagram; Mapping all the matching constraint nodes in the assembly deviation transfer graph to the transfer relations between the corresponding state nodes in the assembly state space graph; Among them, the mapping of various nodes in the assembly deviation transfer graph is carried out in a top-down and left-to-right order, and the consistent manufacturing deviation, the assembly positioning deviation and the assembly deformation deviation are respectively integrated into the corresponding nodes to construct the assembly state space graph.

7. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 6 is characterized in that: The expression of the assembly deviation transfer diagram is: (5) in, represents the assembly deviation transfer diagram, Represents the collection of part nodes in the assembly process, , Represents the set of tooling constraint nodes in the assembly process, , Represents the set of mating constraint nodes in the assembly process. , Represents the set of node transfer directions during the assembly process, ; The expression of the assembly state space diagram is: （6） in, Indicates that the deformable structure is The assembly accuracy after each assembly station completes positioning, Indicates that the deformable structure is The assembly accuracy of the previous assembly station after the assembly station completes the positioning, represents the first deviation coefficient matrix, represents the second deviation coefficient matrix, represents the third deviation coefficient matrix, represents the fourth deviation coefficient matrix, represents the fifth deviation coefficient matrix, Indicates that the deformable structure is Consistent manufacturing deviations at each assembly station, Indicates that the deformable structure is The assembly deviation obtained by measuring each assembly station is Indicates that the deformable structure is The measurement error caused by the measurement equipment at each assembly station.

8. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 1 is characterized in that: The step of performing finite element analysis on the assembly accuracy of the deformable structure by using the assembly state space diagram to obtain an assembly deviation cloud diagram, and converting the assembly deviation cloud diagram into an assembly deviation contour diagram comprises: In finite element analysis software, using the assembly state space diagram to perform finite element analysis on the assembly accuracy of the deformable structure to obtain the assembly deviation cloud diagram; Converting the assembly deviation cloud map into the assembly deviation contour map using a cubic spline interpolation algorithm, a deviation normalization algorithm, and a color mapping algorithm in sequence; The data set reflected by the assembly deviation contour map is divided into the training data set and the test data set, and the ratio of the training data set to the test data set is 4:

1.

9. The geometric and physical fusion method for predicting assembly deviation of deformable structures according to claim 1 is characterized in that: The steps of constructing an initial generative adversarial network model, training the initial generative adversarial network model using the training data set, and verifying the trained initial generative adversarial network model using the test data set to obtain a final generative adversarial network model include: The initial generative adversarial network model includes a finite element module, a generator and a discriminator that are communicatively connected to each other; In the process of training and verifying the initial generative adversarial network model to obtain the final generative adversarial network model, a generative adversarial loss function is used. The generative adversarial loss function includes Loss function and feature matching loss function; The expression of the generative adversarial loss function is: (7) in, represents the generative adversarial loss function, represents a generator, represents the discriminator, represents the type of discriminator, express Loss function, , represents the discriminator’s expectation of the difference between the input image and the true image, represents the input image, represents the real image corresponding to the input image, represents a composite image, represents the discriminator’s expectation of the difference between the input image and the synthesized image, Indicates the generated image, represents the probability that the real image is judged as the true value by the discriminator, represents the probability that the generated image is judged as the true value by the discriminator, represents the feature matching loss function, , The discriminator Layer extractor, represents the total number of layers of the discriminator, The discriminator The number of elements in the layer, represents a random variable, represents a given sample generated by a random variable, represents the expected value of the difference between a given sample and the input image, express The weight of the loss function and the feature matching loss function, Represents the norm of 1.

Images