Accurate digital twinning method and system for high-end motor iron core mold assembly

Abstract

The invention discloses an accurate digital twinning method and system for high-end motor iron core mold assembly. The method comprises the following steps: carrying out point cloud data acquisition on all template holes and inserts in a mold; and respectively fitting according to the point cloud data of each template hole and the insert to form a curved surface. And according to the curved surfaces corresponding to each template hole and each insert and the ideal position of each insert on the mold, establishing an accurate digital twin assembly model of the template and each insert. And through interference assembly simulation and a multi-target parameter optimization scheme, iterative optimization is carried out on the design position of the template hole. According to the method, the distance between the hole center position, obtained through simulation, of the assembled template hole and the ideal position of the corresponding insert on the mold serves as the fitness function, rapid iteration of the design position of the template hole is achieved in combination with the multi-target genetic algorithm, and the design optimization efficiency of the high-end motor iron core mold is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of precision assembly of high-end motor core molds, and specifically relates to a precise digital twin method and system for assembling high-end motor core molds. Background Art

[0002] In the mold industry, especially in the high-end mold manufacturing industry for new energy vehicle motor cores, the mold assembly process has a crucial impact on production and manufacturing. Due to the large number of holes on the mold and the highly nonlinear problem of multi-axis hole interference assembly, the deformation of the interference assembly of the same size but different structures cannot be directly adjusted by the method of reverse compensation after deformation measurement. At the same time, the structural differences between different molds are huge. These problems make the assembly force analysis caused by interference fit extremely difficult, which in turn makes it difficult to control the position of the template hole, which inevitably leads to low assembly efficiency, extended assembly cycle, and increased labor costs.

[0003] With the continuous improvement of mold processing and manufacturing precision requirements, future mold assembly will tend towards high precision, low damage and high efficiency. This means that the accurate prediction of the position change caused by the stress deformation of the interference assembly of high-end motor core molds becomes increasingly important. The accuracy of this prediction is directly related to the precision and stability of mold assembly, and is the key to achieving high-precision, low-damage and high-efficiency assembly. However, with the improvement of assembly requirements, the difficulty of this prediction also increases. Although in-depth research on the principles of mold assembly can provide theoretical support for high-precision assembly, in the actual factory environment, facing a large number of molds of different shapes and sizes, it is still one of the biggest challenges facing the entire life cycle of high-end motor core molds at home and abroad to achieve accurate prediction of these mold changes. Summary of the invention

[0004] In response to the problems of the prior art, this embodiment provides a method for predicting and optimizing resistance to assembly deformation in the high-end motor core mold industry. The method is based on point cloud data processing, establishing an accurate model containing geometric distribution errors, and simulating and optimizing the model, so as to solve the practical problems of low assembly efficiency and high labor costs in the high-end motor core mold industry.

[0005] In a first aspect, the present invention provides a method for assembling a high-end motor core mold with a precise digital twin, comprising the following steps:

[0006] Point cloud data collection is performed on all template holes in the mold and their corresponding inserts.

[0007] The surface is formed by fitting the point cloud data of each template hole and insert.

[0008] Based on the corresponding surfaces of each template hole and insert, as well as the ideal position of each insert on the mold, an accurate digital twin assembly model of the template and each insert is established.

[0009] The interference assembly between the insert and the template hole is simulated on the precise digital twin assembly model to obtain the hole center position of each template hole after assembly deformation; the design position of each template hole in the precise digital twin assembly model is iteratively updated so that the simulated hole center position of each template hole after assembly deformation approaches the ideal position of each insert on the mold.

[0010] The design position of the template hole is the processing position of the template hole in the mold as indicated in the specific drawing.

[0011] Preferably, the design position of each template hole in the precise digital twin assembly model is iteratively updated by a multi-objective genetic method. The fitness function of the multi-objective genetic method is set to the distance between the center position of the template hole after assembly deformation obtained by simulation and the ideal position of the corresponding insert on the mold.

[0012] Preferably, the multi-target genetic method comprises:

[0013] (1) Initialize population parameters: Take the set of design position coordinates of all template holes on the template as individuals and establish the initial population.

[0014] (2) Perform interference fit simulation for each individual in the current population and calculate the fitness function based on the simulation results.

[0015] (3) Generate a new generation of population through selection, crossover, and mutation.

[0016] (4) Repeat steps (2) to (3) until convergence condition is reached or the maximum number of iterations is reached.

[0017] Preferably, the convergence condition is that the fitness function values ​​corresponding to the coordinates of all template hole processing positions in the same individual are less than or equal to a preset allowable error threshold.

[0018] Preferably, the horizontal and vertical coordinate deviations between the designed positions of the template holes in all generated individuals and the corresponding ideal positions of the inserts are less than a preset displacement threshold, which is preferably 20 μm.

[0019] Preferably, the point cloud data is subjected to elimination of gross errors, systematic errors and random errors.

[0020] As a preferred method, the process of establishing an accurate digital twin assembly model is as follows: using the template hole assembly surface obtained by fitting as the assembly surface on the mold, an accurate digital twin 3D model of the mold is established. Using the insert assembly surface obtained by fitting as the assembly surface on the insert, an accurate digital twin 3D model of the insert is established. The accurate digital twin 3D model of the insert and the accurate digital twin 3D model of the mold form an accurate digital twin assembly model.

[0021] Preferably, the process of establishing an accurate digital twin 3D model of the mold is: first establish a basic 3D model of the mold without template holes; use the assembly surfaces of each template hole to perform Boolean cutting operations on the template hole design positions on the basic 3D model of the mold to obtain an accurate digital twin 3D model of the mold.

[0022] Preferably, the interference fit simulation process is: assigning material properties, contact relationships and boundary conditions to the precise digital twin assembly model divided into grids. Performing assembly simulation on each template hole in the precise digital twin assembly model and the corresponding insert, and outputting the coordinates of the center of each template hole after assembly deformation.

[0023] In a second aspect, the present invention provides a precise digital twin system for assembling a high-end motor core mold, which is used to execute the aforementioned method, and is characterized in that: the precise digital twin system for assembling a high-end motor core mold comprises a point cloud acquisition device, a modeling simulation module, and a position optimization module; the point cloud acquisition device is used to collect point cloud data; the point cloud acquisition device is preferably a three-coordinate measuring machine. The modeling simulation module is used to establish a precise digital twin assembly model. The position optimization module is used to iteratively optimize the design position of the template hole on the mold in the precise digital twin assembly model.

[0024] The beneficial effects of the present invention are:

[0025] 1. The present invention proposes a solution combining digital twin technology to address the common assembly deformation problem in the high-end motor core mold industry. This solution optimizes the design position of the mold template hole by establishing an accurate digital twin model, combining simulation technology and optimization algorithm, so that the position of the template hole after interference assembly deformation reaches the ideal position of the insert, thereby improving the dimensional accuracy of the high-end motor core mold after assembly.

[0026] 2. The present invention uses the distance between the center position of the template hole after assembly obtained by simulation and the ideal position of the corresponding insert on the mold as the fitness function, and combines the multi-objective genetic method to achieve rapid iteration of the template hole design position, thereby improving the design optimization efficiency of high-end motor core molds.

[0027] 3. The present invention uses a high-precision three-dimensional coordinate measuring machine to obtain point cloud data of the assembly surface of the assembly, and uses the point cloud data to fit and generate assembly surfaces. Based on the assembly surfaces, a digital twin model with precise geometric information of the assembly surfaces is constructed in computer-aided engineering (CAE) software. The digital twin model integrates the geometric information, material properties, assembly posture and contact relationship of the mold, and can obtain more accurate assembly deformation prediction results in simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the description of the specific implementation methods. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 It is a schematic diagram of the interference fit between the mold template hole and the insert.

[0030] Figure 2 It is an overall flow chart of an embodiment of the present invention.

[0031] Figure 3 It is a flow chart of the template hole and insert point cloud measurement in step S100 of an embodiment of the present invention.

[0032] Figure 4 It is a flowchart of point cloud data processing in step S200 of an embodiment of the present invention.

[0033] Figure 5 It is a flowchart of basic three-dimensional model building in step S300 of an embodiment of the present invention.

[0034] Figure 6 It is a flowchart of generating an accurate digital twin assembly model in step S400 of an embodiment of the present invention.

[0035] Figure 7 It is a flowchart of iterative optimization of template hole design position simulation in step S500 of an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of this embodiment clearer, the technical solution in this embodiment will be described more clearly and completely in combination with the drawings in this embodiment. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Note that the ideal surface or ideal model described in this embodiment is a surface or model without geometric distribution errors, and the precise surface or precise model is a surface or model containing geometric distribution errors.

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] See also Figure 1 , Figure 1 This is a schematic diagram of the interference fit of the template hole and insert of the high-end motor core mold of this embodiment. It is particularly important to note that this embodiment can also be applied to high-end mold assembly occasions of different sizes, shapes and high precision requirements. The process and results of the prediction and optimization method for the resistance of high-end motor core mold to assembly deformation provided by this embodiment are not affected by any external conditions of the mold.

[0040] See also Figure 2 , Figure 2 The overall technical solution flow chart of this embodiment is a method for assembling a precise digital twin of a high-end motor core mold, comprising the following steps:

[0041] Step S100: point cloud measurement of template holes and inserts; Figure 3 As shown, the point cloud data of the inner cylindrical surface of all the template holes and the outer cylindrical surface of the corresponding inserts in the mold are measured based on a high-precision three-dimensional coordinate measuring machine. A set of point cloud data is obtained corresponding to the assembly surface of each template hole and insert. This embodiment is applicable to the assembly of any number of the mold template holes and inserts, and is not limited by the shape structure of the mold template holes, such as cylindrical holes, square holes, special-shaped holes and other template holes and corresponding inserts.

[0042] Before point cloud acquisition, it is necessary to conduct a statistical analysis of the number and size of template holes and inserts to obtain the best point cloud data measurement plan; the point cloud data measurement plan includes the layer spacing of the sampled points and the spacing between adjacent points in each layer; the point cloud data measurement plan is determined based on the calculation results of the Pearson correlation coefficient, thereby ensuring the efficiency and rationality of the measurement plan.

[0043] Step S200: point cloud data processing; removing gross errors in the measured point cloud data, systematic errors of the high-precision three-dimensional coordinate measuring machine, and random errors in the measurement process in turn.

[0044] In this embodiment, it should be noted that in step S200, based on the assembly surface point cloud data measured in step S100, gross errors and random errors are eliminated in Matlab software, and the measured point cloud data are reversely corrected for the three coordinates of X, Y and Z based on the system error correction test.

[0045] like Figure 4 As shown, step S200 specifically includes:

[0046] Step S210: gross error elimination: based on the Laida rule, gross errors of the measured point cloud data of the mold template hole and the corresponding assembly surface of the insert are eliminated.

[0047] In this embodiment, it should be noted that in step S210, the gross error is the straight-line distance from the point cloud to the center of the fitting circle. Distance from average The difference is greater than 3 standard deviations ( ) points, the coarse error points are eliminated, and the step of eliminating coarse errors of point cloud data can be completed. The average distance and standard deviation The calculation formula is:

[0048]

[0049]

[0050] Step S220: System error correction: Perform a system error measurement experiment on the measuring machine, and use the experimental results to perform reverse correction of the system error on the point cloud data that has completed the gross error elimination step.

[0051] In this embodiment, it should be noted that in step S220, the system error correction experiment is as follows: using the high-precision surface of the standard rectangular block as the reference surface, making the nominal length direction of the block parallel to the X, Y and Z directions of the three-dimensional coordinate measuring machine, repeating the dot measurement, and performing the difference calculation with the nominal distance to obtain the system error of the high-precision three-dimensional coordinate measuring machine used for measurement. The X, Y and Z coordinates of the point cloud data are compensated and corrected respectively, and the point cloud data after the system error correction is output.

[0052] Step S230: random error filtering: performing random error filtering on the point cloud data that has completed the system error correction based on median filtering to obtain processed point cloud data for subsequent calculations.

[0053] In this embodiment, it should be noted that in step S230, the point cloud data that has completed the system error correction is filtered out by median filtering using Matlab software, thereby obtaining point cloud data of the template holes and insert assembly surfaces that have completed all point cloud data processing and are used for subsequent calculations.

[0054] Step S300: Modeling a basic three-dimensional model of the mold template; based on the CAD drawing of the high-end motor core mold template, establishing the basic three-dimensional model of the mold template in a three-dimensional modeling software.

[0055] In this embodiment, it should be noted that in step S300, the drawing data of the high-end motor core mold CAD two-dimensional drawing (excluding the template hole) is processed in the AutoCAD software, and the three-dimensional ideal model of the template is established in the Solidworks three-dimensional modeling software. It should be particularly noted that this embodiment does not limit the model type and file format, and the obtained basic three-dimensional model does not contain the template holes to be assembled.

[0056] like Figure 5 As shown, step S300 specifically includes:

[0057] Step S310: mold template drawing processing: processing the CAD drawing of the high-end motor core mold template.

[0058] In this embodiment, it should be noted that, in step S310, the specific drawing processing method is to retain only the structural part on the CAD drawing, and remove the template holes to be assembled and the remaining annotations or data.

[0059] Step S320: IGS file output; output the CAD drawing after the template drawing processing in IGS format.

[0060] In this embodiment, it should be noted that, in step S320, the CAD drawing after the drawing processing in step S310 is specifically converted into the IGS format.

[0061] Step S330: establishing a basic three-dimensional model; reading the IGS format drawing based on Solidwords three-dimensional modeling software, performing a stretching operation on it, and realizing the modeling of the basic three-dimensional model without the template hole.

[0062] In this embodiment, it should be noted that in step S330, the basic three-dimensional model is a three-dimensional model without geometric distribution errors. In the Solidworks three-dimensional modeling software, the IGS file output in step S320 is read, so that the two-dimensional drawing is stretched into a basic three-dimensional model without the surface template holes with geometric errors.

[0063] Step S400: Modeling a precise digital twin assembly model; generating a precise three-dimensional model of the assembly with the geometric distribution error based on the basic three-dimensional model without geometric distribution error, NURBS surface modeling theory and Boolean cutting operation; adding physical conditions to the precise assembly three-dimensional model to generate the precise digital twin model.

[0064] like Figure 6 As shown, step S400 specifically includes:

[0065] Step S410: construct NURBS surfaces respectively according to the point cloud data corresponding to each assembly surface; the process of constructing NURBS surfaces includes: obtaining the node parameters and curve node vectors corresponding to the measured data points in the point cloud data by the cumulative chord length method, and thereby generating NURBS surfaces. This specific process belongs to conventional means and will not be described in detail here.

[0066] Step S420: Generate a precise assembly surface with the geometric distribution error based on the NURBS surface, including the precise assembly surface of the mold template hole and the precise assembly surface of the corresponding insert. It should be noted that in the process of generating the precise assembly surface, the Z coordinates of the top and bottom point clouds are changed so that the height of the generated assembly surface with the geometric distribution error is slightly larger than the mold thickness, in order to ensure that when the Boolean operation cutting is performed later, a template hole with two ends through can be formed in the basic three-dimensional model.

[0067] Step S430: Import the basic three-dimensional model without template holes generated in step S300 and the precise assembly surface of template holes with the geometric distribution error generated in step S420 into Comsol software; the import position of the central axis of the precise assembly surface of the template holes is the design position coordinate of the corresponding template holes on the mold; perform Boolean cutting operation on the template three-dimensional model without assembly holes through the precise assembly surface with the geometric distribution error, and then delete the cut and separated parts and the precise assembly surface of the template holes; at this time, template holes with precise geometric distribution error information are generated on the basic three-dimensional model, thereby obtaining a precise digital twin three-dimensional model of the mold.

[0068] Use the precise assembly surface of the insert to perform Boolean cutting operations on the rectangular 3D solid model whose thickness is consistent with the mold thickness, and retain the cut independent parts as the precise digital twin 3D model of the insert.

[0069] Step S440: Import the precise digital twin 3D model of the mold and the precise digital twin 3D model of the insert generated in step S430 into CAE simulation software for assembly, thereby generating a precise digital twin 3D assembly model of the mold with the geometric distribution error.

[0070] Step S450: Meshing the precise digital twin assembly model by means of a swept mesh, using C3D8R as the mesh unit.

[0071] Step S460: Assign the corresponding physical conditions of the assembly to the precise digital twin assembly model; assign material properties based on the material properties of the mold template and the insert, set the assembly surface contact relationship and apply boundary conditions to the assembly. The precise digital twin assembly model can accurately characterize the actual mold template hole and insert assembly in terms of physical properties and geometric shape.

[0072] In this embodiment, it should be noted that, in step S450, the corresponding physical conditions given to the assembly include: material properties, boundary conditions, temperature environment, contact relationship between the mold template hole and the insert, assembly posture, etc.

[0073] In some other embodiments, the method of constructing an accurate digital twin assembly model is as follows: first, an ideal three-dimensional model including an ideal mold hole is established according to the drawing, and the ideal three-dimensional model is meshed; the surface of the ideal mold hole is a cylindrical surface; then, a NURBS surface is established according to the point cloud data; then, all surface mesh nodes on the assembly surface of the ideal three-dimensional model are offset to the NURBS surface constructed in step S410; each surface mesh node is offset to the nearest interpolation point or extreme point, so that the assembly surface of the ideal three-dimensional model is transformed from an ideal cylindrical surface to a NURBS surface. In some further embodiments, the interpolation points whose distance from the extreme point is less than the threshold are deleted to ensure that each extreme point can be offset by a surface mesh node, so that the offset surface mesh can better express the morphology of the NURBS surface. Finally, according to the offset of all surface mesh nodes, the internal nodes of the ideal three-dimensional are adjusted according to the smoothness principle to ensure the convergence of the mesh quality, and an accurate digital twin assembly model is obtained. This method can avoid the difficulty of directly meshing on an accurate digital twin assembly model with a complex surface.

[0074] Step S500: iterative optimization of the template hole design position simulation; using the precise digital twin assembly model obtained in step S400, perform interference fit simulation of the mold template hole and the corresponding insert in the CAE simulation software; combining the simulation results, through the interaction of the multi-objective genetic method and the precise digital twin assembly model simulation, continuously iterate the design position of the template hole on the mold, so that the result after interference fit of the optimized mold template hole and the corresponding insert meets the high precision and high efficiency requirements of actual high-end motor core mold manufacturing and assembly.

[0075] like Figure 6 As shown, step S500 specifically includes:

[0076] Step S510: setting the initial position variables of the template holes; the ideal position coordinates of each insert on the mold are identified by the ideal hole center coordinates of each template hole on the mold drawing; the ideal hole center coordinates of each template hole on the mold drawing are used as the initial template hole design position coordinates.

[0077] In this embodiment, it should be noted that in step S510, the initial simulation is performed according to the ideal position of the template hole in the CAD drawing, that is, when the multi-objective optimization algorithm is subsequently combined, the initial population is the X and Y coordinates of each template hole on the template. It should be noted that the variables modify the position of the Boolean cutting operation between the precise surface of the template hole and the ideal three-dimensional entity of the mold template in step S420 and the assembly position in step S430, and then imported into the CAE simulation software through python code.

[0078] Step S520: interference fit simulation of the precise digital twin assembly model; preprocessing and simulating the interference fit of each template hole and its corresponding insert in the precise digital twin assembly model in CAE software.

[0079] In this embodiment, it should be noted that in step S520, simulation preprocessing is specifically performed based on CAE simulation software. The simulation preprocessing includes setting node sets for the grid nodes of all template holes and insert assembly surfaces in the precise digital twin model. After the simulation preprocessing, the simulation calculation of the template hole and insert interference assembly is performed on the precise digital twin assembly model.

[0080] Step S530: Read and output the coordinates of the center of each template hole after deformation obtained by simulation.

[0081] In this embodiment, it should be noted that in step S530, the coordinates of all nodes in the node set after the interference fit simulation calculation are read and output, and the least squares fitting is performed based on the node coordinates to obtain the hole center position of the template hole on the mold after simulation.

[0082] Step S540: Multi-objective genetic method; based on the multi-objective optimization algorithm, the optimization iteration of the template hole center position is performed, and steps S520 and S530 are re-executed after each iteration to obtain the updated hole center coordinates after the interference fit, so as to realize the combination of simulation and optimization algorithms; during the iteration process, the hole center coordinates after the interference fit deformation gradually approach the ideal hole center coordinates. In this embodiment, the fitness function of the multi-objective genetic method is set to the distance between the hole center coordinates after the interference fit deformation obtained by simulation and the ideal hole center coordinates. When the fitness functions corresponding to all template holes reach the set allowable error threshold, the convergence condition of the optimization algorithm can be reached in advance, and the multi-objective genetic algorithm process is shown in the figure.

[0083] The specific process of the multi-objective genetic method is as follows:

[0084] (1) Initialize population parameters: A set of values ​​of the design position coordinates of all template holes on the template is taken as an individual, and multiple individuals are set to establish an initial population. In this embodiment, the number of populations is set to 20, and the maximum number of iterations is 50; the range of the horizontal and vertical coordinates of the center of each template hole is set to plus or minus 20 μm.

[0085] (2) Simulate each individual in the current population according to the process of steps S520 and S530, and calculate the fitness function based on the simulation results; if there is no individual in the current population that can meet the convergence conditions, then perform non-dominated sorting on all individuals according to the fitness function of the current population; stratify the individuals in the current population according to the dominance relationship between different individuals in the current population, and find the non-dominated solution in the current population.

[0086] In this embodiment, the convergence condition is that the fitness function values ​​corresponding to the design position coordinates of all template holes in the same individual are less than or equal to a preset allowable error threshold; the value of the allowable error threshold is set according to the accuracy requirement, and is set to 5μm in this embodiment; in other embodiments, the allowable error threshold can also be set to 2μm, 6μm, 8μm, 10μm or other values.

[0087] (3) Select parent individuals based on the results of the non-dominated sorting in step (2), and generate a new generation of sub-populations based on the parent individuals through selection, crossover, and mutation.

[0088] (4) For the new generation population, re-execute steps (2) to (3) and continue iterating until the convergence condition is reached or the maximum number of iterations is reached.

[0089] This embodiment combines the interference fit simulation of the precise digital twin model with the multi-objective optimization algorithm. The design position of the template hole is modified iteratively through the multi-objective genetic method. The simulated position is read and output through the node set to calculate the fitness function, thereby realizing the combination of simulation and optimization algorithms. This greatly improves the optimization efficiency of the template hole design position coordinates and scientifically guides the process parameters of high-end motor core mold companies.

[0090] In view of the problem that assembly deformation in high-end motor core molds is difficult to analyze, this embodiment proposes a prediction and optimization method based on a precise digital twin model, which can accurately predict the position deviation of the extrusion deformation after the interference fit of the template hole, and optimize the initial design position of the template hole through a multi-objective genetic optimization algorithm, so that the position deviation of each template hole after the interference fit is within the allowable error range. The present invention can scientifically guide the process parameters of the high-end motor core mold industry, and well solves the current biggest assembly deformation problem in the mold industry.

[0091] To sum up, this embodiment provides a prediction and optimization method for anti-assembly deformation for the high-end motor core mold industry. It uses a high-precision accurate digital twin model and combines CAE simulation software and multi-objective optimization algorithm to predict the interference assembly deformation of the mold, and provides scientific guidance for the initial design position of the template hole. It can not only greatly reduce the scrap rate of high-end motor core mold production and manufacturing, but also improve the assembly efficiency, thereby greatly reducing the labor cost and time cost of mold companies. At the same time, it can also provide accuracy guarantee for the production of core products for high-end motor core molds, greatly promoting the high-precision, high-efficiency and low-cost development of the mold manufacturing industry.

Claims

1. A precise digital twin method for assembling high-end motor core molds, characterized by: The method includes: Collect point cloud data of all template holes in the mold and their corresponding inserts; The assembly surface is formed by fitting the point cloud data of each template hole and insert respectively; According to the assembly surfaces corresponding to each template hole and insert, as well as the ideal position of each insert on the mold, an accurate digital twin assembly model of the template and each insert is established; The interference assembly between the insert and the template hole is simulated on the precise digital twin assembly model to obtain the hole center position of each template hole after assembly deformation; the design position of each template hole in the precise digital twin assembly model is iteratively updated so that the simulated hole center position of each template hole after assembly deformation approaches the ideal position of each insert on the mold.

2. According to claim 1, a high-end motor core mold assembly precise digital twin method is characterized by: The design position of each template hole in the digital twin assembly model is iteratively updated through a multi-objective genetic method; the fitness function of the multi-objective genetic method is set as the distance between the center position of the template hole after assembly deformation obtained by simulation and the ideal position of the corresponding insert on the mold.

3. According to claim 2, a high-end motor core mold assembly precise digital twin method is characterized by: The multi-target genetic method comprises: (1) Initialize population parameters: Take the set of coordinates of all the template hole design positions on the template as individuals and establish the initial population; (2) Perform interference assembly simulation for each individual in the current population and calculate the fitness function based on the simulation results; (3) Generate a new generation of population through selection, crossover, and mutation; (4) Repeat steps (2) to (3) until convergence condition is reached or the maximum number of iterations is reached.

4. According to claim 3, a high-end motor core mold assembly precise digital twin method is characterized by: The convergence condition is that the fitness function values ​​corresponding to the coordinates of all template hole processing positions in the same individual are less than or equal to a preset allowable error threshold.

5. According to claim 3, a high-end motor core mold assembly precise digital twin method is characterized by: The horizontal and vertical coordinate deviations between the design positions of the template holes and the corresponding ideal positions of the inserts in all generated individuals are less than the preset displacement threshold.

6. The high-end motor core mold assembly precise digital twin method according to claim 1, characterized in that: The point cloud data is subjected to elimination of gross errors, systematic errors, and random errors.

7. According to claim 1, a high-end motor core mold assembly precise digital twin method is characterized by: The process of establishing a precise digital twin assembly model is as follows: using the fitted template hole assembly surface as the assembly surface on the mold to establish a precise digital twin 3D model of the mold; using the fitted insert assembly surface as the assembly surface on the insert to establish a precise digital twin 3D model of the insert; and combining the precise digital twin 3D model of the insert and the precise digital twin 3D model of the mold to form a precise digital twin assembly model.

8. The high-end motor core mold assembly precise digital twin method according to claim 7, characterized in that: The process of establishing an accurate digital twin 3D model of the mold is as follows: first establish a basic 3D model of the mold without template holes; use the assembly surfaces of each template hole to perform Boolean cutting operations on the template hole design positions on the basic 3D model of the mold to obtain an accurate digital twin 3D model of the mold.

9. The high-end motor core mold assembly precise digital twin method according to claim 1, characterized in that: The interference fit assembly simulation process is as follows: assigning material properties, contact relationships and boundary conditions to the precise digital twin assembly model divided into grids; assembling each template hole in the precise digital twin assembly model with the corresponding inserts, and outputting the coordinates of the center of each template hole after assembly deformation.

10. A high-end motor core mold assembly precision digital twin system, characterized by: Used to execute the method as claimed in claim 1, characterized in that: the high-end motor core mold assembly precise digital twin system includes a point cloud acquisition device, a modeling simulation module and a position optimization module; the point cloud acquisition device is used to collect point cloud data; the modeling simulation module is used to establish a precise digital twin assembly model; the position optimization module is used to iteratively optimize the design position of the template hole on the mold in the precise digital twin assembly model.

Images