Parametric design system for precision mold structure based on BIM technology

Abstract

The present application relates to the technical field of mold design and BIM modeling, in particular to a precision mold structure parameterization design system based on BIM technology, comprising condition input, data processing, parameter inversion solving, parameterized modeling and simulation analysis feedback modules, receiving mold initial design constraint conditions, constructing cavity geometric features and extracting load and thermal stress features to obtain design reference data, solving key structure parameters that meet performance indicators through state inversion algorithm, generating a three-dimensional skeleton model and assembling into a parameterized assembly. The system carries out interference check and motion simulation on the assembly, and updates the design reference data with interference state and motion trajectory data feedback, realizes the re-inversion solving and optimization of key parameters. The present application optimizes mold structure parameters through inversion solving and closed-loop feedback, improves the precision and adaptability of mold structure design.

Description

Technical Field

[0001] This invention relates to the field of mold design and BIM modeling technology, and in particular to a precision mold structure parametric design system based on BIM technology. Background Technology

[0002] Conventional precision mold structure design relies on BIM technology to achieve parametric modeling. It mainly uses pre-defined structural parameters such as mold base dimensions, insert layout, cooling water channels, and ejection mechanism positions as the core approach. After receiving the product's three-dimensional geometric model, material physical parameters, molding process parameters, and mold performance indicators, it constructs the geometric features of the mold cavity and performs inference calculations on load distribution and thermal stress field characteristics. After forming the mold design benchmark data, it directly generates a parametric assembly. Subsequent steps only involve single-time interference checks and motion simulation analysis, and local structural adjustments are made to address any issues detected.

[0003] In existing design schemes, key parameters of the mold structure cannot be solved using state inversion algorithms combined with mold performance indicators based on mold design baseline data. Instead, parameters can only be preset based on design experience and then verified for performance. There is no direct linkage between the design baseline data and the parameter solution process. The interference state and motion trajectory data obtained from simulation analysis are only used for single structural verification and cannot be fed back to the inference calculation module to update the mold design baseline data. Therefore, it is impossible to use the updated data to perform further inversion solutions and optimizations of key mold structural parameters.

[0004] It is necessary to solve the problem that the key parameters of the mold structure that fit the mold performance index cannot be solved by the state inversion algorithm. At the same time, it is necessary to solve the problem that the simulation analysis data cannot be fed back to update the design baseline data and that the key parameters cannot be inverted and solved and optimized again. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a precision mold structure parametric design system based on BIM technology.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a precision mold structure parametric design system based on BIM technology, comprising: The condition input module receives the initial design constraints of the target precision mold, which include the product's three-dimensional geometric model, material physical parameters, molding process parameters, and mold performance indicators. The data processing module, based on the three-dimensional geometric model of the product, constructs the corresponding mold cavity geometric features, and combines the material physical parameters and the molding process parameters to extract the load distribution and thermal stress field characteristics of the mold structure through inference calculation, thereby obtaining the mold design benchmark data. The parameter inversion solution module, based on the mold design benchmark data, uses a state inversion algorithm to solve for the key parameters of the mold structure that satisfy the mold performance indicators; The parametric modeling module generates a three-dimensional skeleton model of the mold structure, including mold frame size, insert layout, cooling water channel and ejection mechanism position, based on the key parameters of the mold structure obtained by solving. The three-dimensional skeleton model of the mold structure is then assembled with the geometric features of the mold cavity to obtain a parametric assembly of the mold structure. The simulation analysis feedback module performs interference checks and motion simulation analysis on the parameterized assembly of the mold structure, extracts its interference state and motion trajectory data, and feeds it back to the inference calculation module to update the mold design benchmark data, thereby performing inverse solving and optimization of the key parameters of the mold structure again.

[0007] As a further aspect of the present invention, based on the three-dimensional geometric model of the product, the corresponding mold cavity geometric features are constructed, and combined with the material physical parameters and the molding process parameters, the load distribution and thermal stress field characteristics of the mold structure are extracted by inference calculation, including: The parting surface of the product's three-dimensional geometric model is automatically identified and created to obtain the dividing boundary between the moving mold part and the fixed mold part of the mold, and the mold cavity geometric features are generated based on the dividing boundary. Extract the geometric complexity of the surface from the geometric features of the mold cavity, read the material shrinkage rate from the material physical parameters, and read the injection pressure and melt temperature from the molding process parameters; The geometric complexity of the surface, material shrinkage rate, injection pressure and melt temperature are used as conditional variables for inference calculation and input into a trained load distribution prediction network. The load distribution prediction network outputs a dynamic load distribution map of the mold cavity during the injection stage. Meanwhile, the mold cavity geometry, melt temperature, and molding cycle time are input into another trained thermal stress prediction network, which outputs the steady-state and transient thermal stress field characteristics of the mold under continuous working conditions.

[0008] As a further aspect of the present invention, the step of solving the key parameters of the mold structure that satisfy the mold performance indicators through a state inversion algorithm includes: The state inversion algorithm constructs an optimization problem with mold performance indicators as the objective function and mold design benchmark data as the constraint, and iteratively solves it to obtain the key parameters of the mold structure. The quantitative requirements for maximum mold deformation, mold service life, and cooling efficiency are read from the mold performance indicators. Define a parameterized mold structure model, whose model variables are the key parameters of the mold structure to be solved. The mold structure model variables include at least the mold base thickness, the insert fixing method, the diameter and spacing of the cooling water channels, and the diameter and distribution density of the ejector pins. Based on the parameterized mold structure model and the mold design benchmark data, a forward simulation model is constructed. The forward simulation model can calculate the predicted maximum mold deformation, mold service life and cooling efficiency according to the input key mold structure parameters. The quantitative requirements of the mold performance index are set as the target state of the state inversion algorithm, and the key parameters of the mold structure are set as inversion variables. The key parameters of the mold structure that minimize the norm of the difference between the output of the forward simulation model and the target state are solved by the state inversion algorithm.

[0009] As a further aspect of the present invention, the step of generating a three-dimensional skeleton model of the mold structure, including mold base dimensions, insert layout, cooling channels, and ejection mechanism positions, based on the solved key parameters of the mold structure, and assembling the three-dimensional skeleton model of the mold structure with the geometric features of the mold cavity, includes: Call the preset BIM component library, and based on the mold frame dimensions in the key parameters of the mold structure, retrieve or parameterize a standard mold frame 3D model from the BIM component library; Based on the insert layout in the key parameters of the mold structure, parameterized insert models are inserted at predetermined positions in the standard mold frame 3D model, and assembly constraint relationships between the insert models and the mold frame are established. Based on the cooling channel diameter and spacing in the key parameters of the mold structure, three-dimensional models of cooling pipes are arranged along the preset hot spot areas in the standard mold frame three-dimensional model and the insert model. Based on the ejector pin diameter and distribution density in the key parameters of the mold structure, three-dimensional models of ejector pins and reset pins are arranged in the moving mold part of the standard mold frame three-dimensional model to form an ejection mechanism; The entire three-dimensional model, including the mold base, inserts, cooling pipes, and ejection mechanism, is defined as the three-dimensional skeleton model of the mold structure. This three-dimensional skeleton model of the mold structure is then integrated with the geometric features of the mold cavity through coordinate alignment and Boolean operations to obtain the parametric assembly of the mold structure.

[0010] As a further aspect of the present invention, interference checks and motion simulation analyses are performed on the parameterized assembly of the mold structure to extract its interference state and motion trajectory data, including: Static geometric collision detection is performed on all components of the moving mold part, fixed mold part, ejection mechanism and cooling pipe in the parameterized assembly of the mold structure to identify whether there are static interference areas between components, and the static interference position and interference amount are recorded. A driving command for mold opening and closing motion is applied to the parameterized assembly of the mold structure, driving the moving mold part and the ejection mechanism to move according to a preset speed curve; During the movement, the system detects in real time whether there is dynamic interference between the moving components and the fixed components, and records the time, location and component information of the dynamic interference. Simultaneously, the motion trajectory coordinate sequence of the push rod, slider, and inclined push rod moving components is tracked and recorded as motion trajectory data.

[0011] As a further aspect of the present invention, the interference state and motion trajectory data are extracted and fed back to the inference calculation module to update the mold design baseline data, thereby performing a second inversion solution and optimization of the key parameters of the mold structure, including: The location and interference amount of the identified static interference area are converted into boundary constraint modification instructions for the geometric features of the mold cavity in the mold design reference data. The boundary constraint modification instructions are used to shrink or offset the geometric boundaries of the relevant area. The recorded time points of dynamic interference and component information are converted into constraint modification instructions for the load distribution of the mold structure in the mold design reference data. The constraint modification instructions are used to limit the motion envelope space of the relevant components. The recorded motion trajectory data is compared with the preset ideal motion trajectory to generate trajectory deviation data. The trajectory deviation data is used to correct the parameters related to the dynamics of the ejection mechanism in the mold design reference data. The modified and corrected mold design baseline data is re-input into the state inversion algorithm. Using the updated data as constraints, the inversion solution of the key parameters of the mold structure is re-executed to obtain the optimized key parameters of the mold structure.

[0012] As a further aspect of the present invention, the state inversion algorithm constructs an optimization problem with mold performance indicators as the objective function and mold design benchmark data as constraints, and iteratively solves to obtain key parameters of the mold structure, including: Establish an objective function, which is the weighted sum of squares of the differences between the quantified values ​​of the mold performance index and its target value; Establish constraints, which include mathematical expressions for geometric constraints, strength constraints, and thermal constraints derived from mold design baseline data; The gradient descent method is used to initialize the key parameters of the mold structure, and the value of the objective function and the satisfaction of the constraints under the current values ​​are calculated. Based on gradient information and the degree of constraint violation, the key parameters of the mold structure are iteratively updated using a sequential quadratic programming method, so that the objective function value gradually decreases until convergence while satisfying all constraints.

[0013] As a further aspect of the present invention, it also includes: The collaborative version management module stores the key parameters of the mold structure and the corresponding parametric assembly of the mold structure generated in each iteration as design versions in the BIM database, and records the version number, modified content and associated design constraints. When multiple designers work in parallel, the system compares in real time the parametric assemblies submitted by different designers for the same mold area in the BIM database to identify design conflicts. For identified design conflicts, the system automatically marks the conflict areas and provides a parameter comparison view of the conflicting components, prompting designers to coordinate. After receiving confirmation instructions from the designers, the system will merge the final coordinated key parameters of the mold structure with the parametric assembly and update it to the master version in the BIM database.

[0014] As a further aspect of the present invention, after re-performing the inversion solution of the key parameters of the mold structure to obtain optimized key parameters of the mold structure, the method further includes a parameter-driven automatic generation step of mold components: Based on the optimized key parameters of the mold structure, the parametric template components in the BIM component library are automatically updated in size and configuration. The updated parametric template components are automatically reassembled according to the topological relationships defined in the 3D skeleton model of the mold structure to generate a new parametric assembly of the mold structure. The parameterized assembly of the new mold structure is automatically lightweighted, and its three-dimensional geometric information and material property information are extracted to generate mold manufacturing information that can be directly used for CNC machining.

[0015] As a further aspect of the present invention, it also includes: The virtual verification module creates a high-fidelity digital twin model of the mold based on the parameterized assembly of the final mold structure. The digital twin model contains complete geometric, material, constraint and load information. Apply molding process parameters consistent with the initial design constraints to the digital twin model of the mold, and perform finite element simulation of the entire injection molding process; From the finite element simulation results, extract the stress cloud map, deformation cloud map, temperature field distribution and melt filling flow front data of the mold under the simulated working state; The extracted stress cloud map, deformation cloud map, temperature field distribution and melt filling flow front data are compared and verified with the mold performance indicators, and the verification results are used as the final closed-loop feedback for the parameterized design of the mold structure.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: Based on the mold design benchmark data, a state inversion algorithm is used to solve the key parameters of the mold structure. The mold performance index is used as the direct solution guide. It integrates the cavity geometric features, material physical parameters and molding process parameters corresponding to the product's three-dimensional geometric model. It completes the inference calculation based on the load distribution and thermal stress field characteristics. The design benchmark data is transformed into the core basis for parameter solution. The generation of key parameters of the mold structure directly corresponds to the performance index requirements. It abandons the design logic of preset parameters based on experience. The parameter solution process fully integrates the correlation attributes of material, process and structural features. The structural parameters and performance indexes form a direct correspondence.

[0017] Interference checks and motion simulation analyses are performed on the parameterized assembly of the mold structure. Interference state and motion trajectory data are extracted and transmitted to the inference calculation module. Based on this data, the mold design benchmark data is updated. The updated design benchmark data drives the re-inversion calculation of key parameters of the mold structure. The simulation analysis results, design benchmark data, and parameter inversion process form a closed-loop linkage. The deviation of interference state and motion trajectory directly affects the parameter optimization process. The configuration of mold base size, insert layout, cooling water channel and ejection mechanism position is continuously corrected with the inversion iteration process. The rationality of structural assembly and motion adaptation gradually conforms to the design requirements with data feedback. Attached Figure Description

[0018] Figure 1 This is a timing diagram of the precision mold structure parametric design system based on BIM technology described in this invention. Figure 2 A flowchart for inverting and solving key parameters of the mold structure; Figure 3 This is a diagram showing the load distribution characteristics of the mold cavity and inserts. Figure 4 The effect diagram of the parametric design and iterative optimization of the mold; Figure 5 A diagram illustrating the effectiveness of BIM mold collaborative version management. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] See Figure 1 This invention provides a parametric design system for precision mold structures based on BIM technology. The overall implementation scheme of the system is as follows: The condition input module receives initial design constraints for the target precision mold provided by the user or upstream system. These constraints include the product's 3D geometric model, material physical parameters, molding process parameters, and mold performance indicators. The data processing module, based on the received product 3D geometric model, constructs the corresponding mold cavity geometric features. Combining the material physical parameters and molding process parameters, it uses specific inference calculation methods to extract the load distribution and thermal stress field characteristics of the mold structure, thereby generating mold design baseline data. The parameter inversion solution module, based on this mold design baseline data, uses a state inversion algorithm to solve for the key parameters of the mold structure that meet the predetermined mold performance indicators. The parametric modeling module, based on the solved key parameters, automatically generates a 3D skeleton model of the mold structure, including information such as mold base dimensions, insert layout, cooling channels, and ejector mechanism positions. This skeleton model is then assembled and integrated with the previously constructed mold cavity geometric features to form a complete parametric assembly of the mold structure. The simulation analysis feedback module performs interference checks and motion simulation analysis on the parameterized assembly, extracts the interference state and motion trajectory data, and feeds this analysis data back to the inference calculation process of the data processing module to update the mold design benchmark data. This, in turn, drives the parameter inversion solution module to solve and optimize the key parameters of the mold structure again, forming a closed-loop design optimization process.

[0022] In one embodiment of the present invention, when constructing the geometric features of the mold cavity and extracting load and thermal stress features, an automatic parting surface identification and creation operation is performed on the three-dimensional geometric model of the product to obtain the dividing boundary between the moving mold part and the fixed mold part. Based on this dividing boundary, a complete mold cavity geometric feature is generated. Subsequently, the geometric complexity of the surface is extracted from the mold cavity geometric feature, the material shrinkage rate is read from the material physical parameters, and the injection pressure and melt temperature are read from the molding process parameters. The geometric complexity of the surface, the material shrinkage rate, the injection pressure, and the melt temperature are used as conditional variables for inference calculation and input into a pre-trained load distribution prediction network. This network outputs a dynamic load distribution map of the mold cavity during the injection molding stage. At the same time, the mold cavity geometric feature, melt temperature, and molding cycle time are input into another pre-trained thermal stress prediction network. This network outputs the steady-state and transient thermal stress field features of the mold under continuous working conditions. The dynamic load distribution map and the thermal stress field features together constitute the mold design reference data required for subsequent calculations.

[0023] In practical implementation, for a specific automotive connector product's 3D geometric model, the automatic parting surface recognition and creation function analyzes the geometric topology and demolding direction of the product's 3D geometric model. It automatically identifies the maximum projected contour line of the product's 3D geometric model as the main parting line and generates parting surfaces along this line. These parting surfaces divide the mold space into a moving mold part and a fixed mold part. Based on the boundaries of the parting surfaces, it generates mold cavity geometric features that perfectly fit the outer surface of the product's 3D geometric model. After the mold cavity geometric features are constructed, the system extracts quantitative indicators describing the surface geometric complexity from these features, such as the average Gaussian curvature and draft angle variation rate. Simultaneously, it reads the material shrinkage rate of the specified plastic for the connector from the material physical parameter database and the preset injection pressure and melt temperature from the molding process parameter settings.

[0024] In some embodiments, the quantitative indicators of surface geometric complexity, material shrinkage rate, injection pressure, and melt temperature are used as conditional variables and input into a trained load distribution prediction network. This load distribution prediction network is a deep neural network model whose input layer receives a vector composed of the four conditional variables. After nonlinear transformation through multiple hidden layers, a dynamic load distribution map covering the entire mold cavity surface is generated at the output layer. The dynamic load distribution map is presented in the form of grid data, with each grid node containing a scalar value representing the pressure experienced at that location during the injection stage. In the same processing flow, the mold cavity geometry, melt temperature, and molding cycle time are input in parallel into another independent thermal stress prediction network. This thermal stress prediction network is also a deep neural network, which outputs the steady-state temperature field at different depths inside the mold core and the transient thermal stress field characteristics during the mold opening and closing cycles under continuous operating conditions.

[0025] It is understandable that the geometric complexity metric of a curved surface is obtained through a calculation formula, as follows: ; in: This represents the geometric complexity coefficient of the surface. This represents the total area of ​​all surfaces in the geometric features of the mold cavity. This indicates the integration domain, which is the entire cavity surface. This represents the Gaussian curvature at a point on the surface. It is a weighting coefficient. This represents the standard deviation of the draft angle at all points on the surface. This coefficient is calculated and used as an input dimension of the load distribution prediction network.

[0026] Optionally, the material shrinkage rate is a physical property read directly from the material physical parameter database in floating-point numerical form, while the injection pressure and melt temperature are process settings read from the molding process parameter configuration file. The dynamic load distribution map output by the load distribution prediction network is a three-dimensional scalar field with a spatial resolution consistent with the discretized mesh of the mold cavity geometry. The value of each cell in the dynamic load distribution map represents the expected pressure borne by that local area during the injection and holding stage. The steady-state and transient thermal stress field characteristics output by the thermal stress prediction network are represented in the form of a three-dimensional tensor of a time series, characterizing the internal thermal stress state of the mold at different working time points. In some embodiments, the training data for the two prediction networks comes from the finite element analysis simulation database of historical successful mold projects. The training process enables the networks to learn the complex mapping relationship from product geometry, material and process parameters to the stress and heat state of the mold. In the application phase, the trained load distribution prediction network and thermal stress prediction network operate in a forward inference manner, and their inference calculation speed is significantly faster than performing a complete finite element analysis. The dynamic load distribution map output by the load distribution prediction network and the thermal stress field characteristics output by the thermal stress prediction network are directly spliced ​​together to form the mold design reference data required for subsequent parameter inversion solution.

[0027] In one embodiment of the present invention, the state inversion algorithm solves for the key parameters of the mold structure by constructing an optimization problem with mold performance indicators as the objective function and mold design baseline data as constraints. See also... Figure 2The quantitative requirements for maximum mold deformation, mold lifespan, and cooling efficiency are obtained from mold performance indicators. A parameterized mold structure model is defined, whose model variables are the key parameters of the mold structure to be solved. These variables include at least the mold base thickness, insert fixing method, cooling channel diameter and spacing, and ejector pin diameter and distribution density. Based on this parameterized mold structure model and mold design baseline data, a forward simulation model is constructed. This model can calculate and predict the corresponding maximum mold deformation, mold lifespan, and cooling efficiency based on a set of input key mold structure parameters. The quantitative requirements of mold performance indicators are set as the target state of the state inversion algorithm, and the key mold structure parameters are set as inversion variables. The specific solution process of the algorithm is as follows: First, an objective function is established, which is the weighted sum of squares of the differences between the predicted values ​​of each mold performance indicator output by the forward simulation model and their corresponding target values. At the same time, constraint conditions are established, which include mathematical expressions of geometric constraints, strength constraints, and thermal constraints derived from the mold design baseline data. The gradient descent method is used to initialize the key parameters of the mold structure, and the objective function value and the satisfaction of each constraint condition under the current values ​​are calculated. Subsequently, based on the gradient information of the objective function with respect to the parameters and the degree of constraint violation, the values ​​of the key parameters of the mold structure are iteratively updated using a sequential quadratic programming method, so that the objective function value gradually decreases while satisfying all constraints, until the algorithm converges. The key parameters of the mold structure obtained at this point are the solution that minimizes the difference between the output of the forward simulation model and the target state.

[0028] In practical implementation, the state inversion algorithm takes mold design benchmark data and mold performance indicators as input to solve for key parameters of the mold structure. The mold performance indicators explicitly require that the maximum mold deformation not exceed 0.05 mm, the mold service life reach more than 1 million cycles, and the cooling efficiency cool the product to the ejection temperature within 15 seconds. The key parameters of the mold structure are defined as mold base thickness, insert fixing method, cooling channel diameter, cooling channel spacing, ejector pin diameter, and ejector pin distribution density. These parameters together constitute a parameterized mold structure model. Based on the parameterized mold structure model and mold design benchmark data, a forward simulation model is constructed. The forward simulation model receives a set of assigned values ​​for the key mold structure parameters and, through built-in simplified mechanical and thermal conduction calculation rules, quickly predicts the corresponding maximum mold deformation, mold service life, and cooling efficiency. The goal of the state inversion algorithm is to make the output of the forward simulation model—the predicted values ​​of maximum mold deformation, mold service life, and cooling efficiency—approach the target values ​​set for the mold performance indicators infinitely.

[0029] In some embodiments, the state inversion algorithm achieves the above objective by solving a constrained optimization problem, establishing an objective function J. The objective function J is expressed as the weighted sum of squares of the differences between the predicted values ​​output by the forward simulation model and the target values ​​of the mold performance indicators, and its mathematical expression is: ; in: Represented by the key parameter vector of the mold structure The objective function value of the independent variable. This represents the predicted maximum deformation of the mold calculated by the forward simulation model. This indicates that the target value for the maximum deformation of the mold, as required by the mold performance indicators, is 0.05 mm. This represents the predicted cooling efficiency calculated by the forward simulation model. This indicates that the target cooling efficiency in the mold performance specifications is 15 seconds. This represents the predicted lifespan of the mold calculated by the forward simulation model. This indicates that the target mold lifespan, as required by the mold performance indicators, is 1 million cycles. , , These are the weighting coefficients for each item. The constraints of the optimization problem include mathematical expressions for geometric constraints, strength constraints, and thermal constraints derived from the mold design baseline data. For example, geometric constraints require that the spacing between cooling channels must be greater than twice the channel diameter, strength constraints require that the calculated maximum stress must be less than the allowable stress of the material, and thermal constraints require that the temperature difference on the mold core surface must be less than a set value.

[0030] It is understandable that the gradient descent method is used to analyze the key parameter vectors of the mold structure. Perform initialization assignment; these initialization values ​​can be randomly generated or sampled from empirical values. Calculate the objective function under the current parameter assignments. Determine the value of , evaluate the satisfaction of all constraints, and calculate the objective function. For parameter vectors gradient Based on gradient information Based on the degree of constraint violation, the key parameter vector of the mold structure is iteratively updated using a sequential quadratic programming method. The sequential quadratic programming method constructs a quadratic programming subproblem in each iteration to solve for the search direction and step size. Optionally, in each iteration, the algorithm determines whether the decrease in the objective function value is less than a preset threshold and whether the constraint violation amount approaches zero. If the convergence condition is met, the iteration stops, and the current key parameter vector of the mold structure is... The assigned values ​​are the solution results, which minimize the norm of the difference between the output of the forward simulation model and the target state of the mold performance index. If the convergence condition is not met, the forward simulation model is run again based on the new parameter assignments to update the predicted values ​​and repeat the gradient calculation and sequential quadratic programming steps. In some embodiments, the forward simulation model is a fast evaluator based on a surrogate model, whose calculation speed is much higher than that of a complete finite element analysis, which allows the state inversion algorithm to complete multiple iterations within an acceptable time. Finally, the state inversion algorithm outputs a set of determined key parameters of the mold structure, including a mold base thickness of 300 mm, insert fixing method of screw locking, cooling water channel diameter of 8 mm, cooling water channel spacing of 25 mm, ejector pin diameter of 4 mm, and ejector pin distribution density of 0.04 pins per square centimeter.

[0031] In one embodiment of the present invention, the parametric modeling module generates a three-dimensional skeleton model of the mold structure based on the solved key parameters of the mold structure and assembles it. It calls a preset BIM component library and, based on the mold frame dimensions in the key parameters, retrieves or parametrically generates a standard mold frame three-dimensional model from the library. Based on the insert layout in the key parameters, parametric insert models are inserted at predetermined positions in the standard mold frame three-dimensional model, and assembly constraints are established between the insert models and the mold frame. Based on the cooling channel diameter and spacing in the key parameters, three-dimensional models of cooling pipes are arranged along preset hotspot areas in the standard mold frame three-dimensional model and the insert models. Based on the ejector rod diameter and distribution density in the key parameters, three-dimensional models of ejector rods and reset rods are arranged in the moving mold part of the standard mold frame three-dimensional model to form an ejection mechanism. The entire three-dimensional model, including the mold frame, inserts, cooling pipes, and ejection mechanism, is defined as the three-dimensional skeleton model of the mold structure. This skeleton model is then integrated with the geometric features of the mold cavity through coordinate alignment and Boolean operations to obtain a parametric assembly of the mold structure. The simulation analysis feedback module performs interference checks on the parametric assembly. It performs static geometric collision detection on all components, including the moving mold, fixed mold, ejector mechanism, and cooling pipes, identifying any static interference areas and recording the location and magnitude of the interference. A mold opening and closing drive command is applied to the parametric assembly, driving the moving mold and ejector mechanism to move according to a preset speed curve. During this movement, the module detects dynamic interference between moving and fixed components in real time, recording the time, location, and component information of any dynamic interference. Simultaneously, it tracks and records the motion trajectory coordinate sequences of the ejector rod, slider, and inclined ejector components as motion trajectory data.

[0032] In practical implementation, the parametric modeling module generates a three-dimensional skeleton model of the mold structure based on the key parameters of the mold structure output by the state inversion solution module. The key parameters of the mold structure include a mold frame thickness of 300 mm, screw locking method for inserts, cooling water channel diameter of 8 mm, cooling water channel spacing of 25 mm, ejector pin diameter of 4 mm, and ejector pin distribution density of 0.04 pins per square centimeter. The module calls the preset BIM component library and retrieves the corresponding standard mold frame three-dimensional model file based on the mold frame thickness of 300 mm and the standard mold frame model. The standard mold frame three-dimensional model file is a parametric model, and its length, width, plate thickness and other dimensions can be driven by parameters. Based on the insert layout parameters, parametric insert models are inserted at predetermined positions in the standard mold frame three-dimensional model, namely around the cavity and stress concentration areas. After the insertion operation, the system automatically establishes the screw locking assembly constraint relationship between the insert model and the mold frame template. Based on a cooling channel diameter of 8 mm and a cooling channel spacing of 25 mm, 3D models of cooling channels are arranged along the hot spot areas of the mold determined by thermal analysis in the moving mold platen, fixed mold platen, and inserted insert models of the standard mold base 3D model. The 3D models of cooling channels are defined by spatial curve paths and diameter parameters. Based on an ejector pin diameter of 4 mm and an ejector pin distribution density of 0.04 pins per square centimeter, the spatial position coordinates of ejector pins and reset pins are calculated and generated in the moving mold part of the standard mold base 3D model, i.e., the ejector plate area. 3D models of corresponding diameters are then arranged to form the ejection mechanism. The entire 3D model including the mold base, inserts, cooling channels, and ejection mechanism is defined as a 3D skeleton model of the mold structure. This 3D skeleton model of the mold structure is then aligned with the geometric features of the mold cavity generated from the product model in the coordinate system. A Boolean union operation is then performed to integrate the results, resulting in a complete parametric assembly of the mold structure.

[0033] In some embodiments, interference checks are performed on the parameterized assembly of the mold structure. These checks include static geometric collision detection and dynamic motion interference detection. Static geometric collision detection performs pairwise geometric intersection tests on all components of the parameterized assembly, including the moving mold portion, the fixed mold portion, the ejector mechanism, and the cooling pipes, to identify whether static interference regions exist between components. All static interference positions and interference amounts are recorded, with the interference amount typically referring to the volume or penetration depth of the intersection region. After completing the static check, a mold opening and closing motion drive command is applied to the parameterized assembly of the mold structure. This drive command defines the moving mold portion to move relative to the fixed mold portion at a set speed curve, while simultaneously driving the ejector mechanism to eject according to a preset sequence after mold opening. During the motion, the system calculates the minimum distance between the moving and fixed components in real time. When the minimum distance is less than a safety clearance threshold, dynamic interference is determined to exist, and the time point of dynamic interference, the component information involved, and the interference position are recorded. The system also tracks and records the motion trajectory coordinate sequences of the ejector rod, slider, and inclined ejector components as motion trajectory data, which is stored in the form of a time-position coordinate list.

[0034] Optionally, the results of static interferometry can be recorded in tabular form. For details of the recorded content, please refer to Table 1.

[0035] Table 1: Record of Static Interference Inspection Results for Parametric Assembly of Mold Structure The deviation between the motion trajectory data and the preset ideal motion trajectory is evaluated using a deviation calculation formula, which is defined as follows: ; in: Indicates the average trajectory deviation. This represents the total number of sampling points during the motion process. Indicates the first Each sampling time point Indicates in The position coordinate vector of the moving component recorded at any given time. Indicates in The ideal motion trajectory position coordinate vector preset at all times. This represents the Euclidean norm of a vector, which is the straight-line distance between two points.

[0036] It is understandable that the recorded static interference positions and amounts are used for subsequent correction of the mold structure geometry, the time points of dynamic interference and component information are used to analyze motion logic defects, and the recorded motion trajectory coordinate sequences are used to evaluate the motion smoothness and positioning accuracy of the ejector mechanism, slider, and angled ejector. This data extracted from the simulation analysis feedback module provides input for updating the mold design baseline data. In some embodiments, static geometric collision detection uses a bounding box hierarchical tree algorithm to accelerate calculation, while dynamic motion interference detection performs continuous collision detection within discrete time steps. The tracking accuracy of the motion trajectory is directly related to the time step setting of the simulation calculation.

[0037] See Figure 3 This is a load distribution characteristic diagram of the mold cavity and inserts, belonging to the mechanical analysis stage of BIM precision mold structure parametric design. The peak load distribution of the cavity is approximately 140MPa, appearing at coordinates 75–80mm, exhibiting a single-peak Gaussian distribution, with the load concentrating from both ends towards the center of the cavity. The peak load distribution of the inserts is approximately 95MPa, appearing at coordinates 115–120mm, with the peak position lagging behind the cavity load, and the overall load intensity is lower than that of the cavity. The load concentration characteristic shows that the peak load of the cavity is much higher than that of the inserts, indicating that the cavity area is the most stress-bearing part of the mold and a key focus of structural strength design. The positional offset characteristic shows that the peak load of the inserts lags behind that of the cavity, reflecting that as a reinforcing structure, the force transmission of the inserts has a spatial delay, and their layout needs to match the peak load area of ​​the cavity to achieve effective support.

[0038] In one embodiment of the present invention, the interference state and motion trajectory data extracted by the simulation analysis feedback module are fed back to update the mold design reference data. The location and interference amount of the identified static interference area are converted into boundary constraint modification instructions for the geometric features of the mold cavity in the mold design reference data. These instructions are used to shrink or offset the geometric boundaries of the relevant areas. The recorded time points and component information of dynamic interference are converted into constraint modification instructions for the load distribution of the mold structure in the mold design reference data. These instructions are used to limit the motion envelope space of the relevant components. The recorded motion trajectory data are compared with the preset ideal motion trajectory to generate trajectory deviation data. This data is used to correct the parameters related to the dynamics of the ejection mechanism in the mold design reference data. The modified and corrected mold design reference data are re-input into the state inversion algorithm. Using the updated data as constraints, the inversion solution of the key parameters of the mold structure is re-executed to obtain optimized key parameters of the mold structure. After obtaining the optimized key parameters of the mold structure, the parameter-driven automatic generation step of mold components is executed. Based on the optimized key parameters of the mold structure, the parameterized template components in the BIM component library are automatically updated in size and configuration. The updated parametric template components are automatically reassembled according to the topological relationships defined in the 3D skeleton model of the mold structure, generating a new parametric assembly of the mold structure. The new parametric assembly of the mold structure is then automatically lightweighted, and its 3D geometric and material property information is extracted to generate mold manufacturing information that can be directly used for CNC machining.

[0039] In practical implementation, the interference state and motion trajectory data extracted by the simulation analysis feedback module are fed back to the data processing module to update the mold design baseline data. From the static interference inspection results, the position and amount of the identified static interference area are converted into boundary constraint modification instructions for the geometric features of the mold cavity in the mold design baseline data. For example, if the record shows positional interference between the ejector pin and the cooling channel, the boundary constraint modification instruction calculates and instructs the geometric center line of the cooling channel to be offset by a distance in a specific direction at the interference position, based on the amount of interference and safety clearance requirements. From the dynamic motion interference detection results, the recorded time point of dynamic interference and component information are converted into constraint modification instructions for the load distribution of the mold structure in the mold design baseline data. For example, if the inclined ejector and the moving platen experience dynamic interference in the middle of mold opening, the constraint modification instruction adds a constraint condition to the mold design baseline data that restricts the range of motion of the inclined ejector at that time point, based on the time of interference and the motion envelope of the components. From the recorded motion trajectory data, the actual motion trajectory coordinate sequence of the ejector pin is compared with the preset ideal linear motion trajectory to calculate and generate trajectory deviation data. The trajectory deviation data includes the position offset and angle deflection of the ejector pin at different times. The trajectory deviation data is used to correct the parameters related to the dynamics of the ejection mechanism in the mold design reference data, such as the friction coefficient of the fit clearance between the ejector pin and the template hole.

[0040] In some embodiments, the modified and corrected mold design baseline data is re-input into the state inversion algorithm. The algorithm uses the updated mold design baseline data as new constraints to re-execute the inversion solution optimization process for the key parameters of the mold structure. The optimization process outputs a new set of key mold structure parameters. Compared with the initially solved parameters, the new parameters, while satisfying the original mold performance indicators, ensure that the newly generated geometric layout avoids the identified static and dynamic interferences. A specific example of parameter changes is shown in Table 2.

[0041] Table 2: Comparison of Key Parameters of Mold Structure Before and After Optimization The offset of the geometric boundary can be determined by a function based on the interferometer, and the offset calculation formula is expressed as: ; in: This represents the distance that the calculated geometric boundary needs to be offset. This indicates the specific amount of interference recorded from the static interferometry results. This represents the system's preset minimum safety gap value, the function. This represents a mapping relationship that ensures the final adjusted geometric spacing is at least 1. .

[0042] After obtaining the optimized key parameters of the mold structure, a parameter-driven automatic generation step for mold components is executed. Based on the optimized key parameters, the parametric template components in the BIM component library are automatically updated in size and configuration. For example, the "top rod" family file in the BIM component library automatically updates its cross-sectional dimensions based on the new top rod diameter of 3.8 mm, and the standard template block updates its 3D model based on the new mold frame thickness of 320 mm. The updated parametric template components are then automatically reassembled according to the topology and assembly relationships defined in the 3D skeleton model of the mold structure, generating a new parametric assembly of the mold structure. During the assembly process, new components such as top rods with a diameter of 3.8 mm and cooling water channels with a spacing of 28 mm are automatically placed into the updated standard mold frame according to the optimized layout. Optionally, the parametric assembly of the new mold structure is automatically lightweighted. The lightweighting process includes removing obscured surfaces, simplifying fine features such as fillets and chamfers, and extracting its three-dimensional geometric information and material property information. The extracted information includes the boundary representation model of all parts, mass properties, and the material specified as mold steel. Mold manufacturing information that can be directly used for CNC machining is generated. The mold manufacturing information includes the triangular facet model required for toolpath calculation, the machining coordinate system, and the part process list.

[0043] See Figure 4 This is a diagram illustrating the iterative optimization effect of parametric mold design, visually reflecting the changes in performance, interference, and computational efficiency with each iteration step in BIM-based parametric mold design. Mold performance indicators continuously increase with each iteration step, rising from 75% to 95%, gradually approaching the design target. This demonstrates the effectiveness of the state inversion algorithm and simulation feedback optimization, with core indicators such as maximum mold deformation, service life, and cooling efficiency gradually meeting the standards. The number of interferences continuously decreases with each iteration step, ultimately dropping to zero in the fifth step, achieving interference-free design. This reflects the closed-loop effect of static / dynamic interference checks and structural corrections, completely eliminating the risk of collisions between components and ensuring the reliability of mold opening, closing, and ejection movements. Computation time decreases slowly with each iteration step, from 120 seconds to 95 seconds, a reduction of approximately 20.8%. This indicates that as the design converges, the amount of simulation and inversion calculations in subsequent iterations gradually decreases, and algorithm efficiency improves with the optimization process.

[0044] In one embodiment of the invention, the collaborative version management module stores the key parameters of the mold structure and their corresponding parametric assemblies generated in each iteration as independent design versions in the BIM database, recording the version number, modifications, and associated design constraints. When multiple designers operate in parallel, the system compares the parametric assemblies submitted by different designers for the same mold area in the BIM database in real time, automatically identifying design conflicts. For identified design conflicts, the system automatically marks the conflict area and provides a parameter comparison view of the conflicting components on the interface, prompting relevant designers to coordinate. After receiving confirmation instructions from the designers after coordination, the system merges the finally coordinated key parameters of the mold structure with the parametric assemblies, updating it to the master version in the BIM database. The system also includes a virtual verification module, which creates a high-fidelity digital twin model of the mold based on the finally determined parametric assembly of the mold structure. This digital twin model contains complete geometric, material, constraint, and load information. Molding process parameters consistent with those in the initial design constraints are applied to the digital twin model of the mold, and finite element simulation of the entire injection molding process is performed. From the finite element simulation results, stress cloud map, deformation cloud map, temperature field distribution and melt filling flow front data of the mold under simulated working conditions are extracted. The extracted data are compared and verified with the initially set mold performance indicators. This verification result serves as the final closed-loop feedback of the mold structure parametric design process.

[0045] In practical implementation, the collaborative version management module manages the design iteration process. Each key parameter of the mold structure generated by the state inversion solution module, along with its corresponding parametric assembly generated by the parametric modeling module, is stored as an independent design version in the BIM database. Each version is recorded with a unique version number, a description of the main modifications, and associated initial design constraints. When multiple designers operate in parallel—for example, designer A modifying cooling channel layout parameters while designer B adjusts ejector mechanism position parameters—the system compares the parametric assemblies submitted by different designers for the same mold area in the BIM database in real time. The system identifies design conflicts by comparing changes in component geometry, attributes, and positions. For identified design conflicts, the system automatically highlights the conflict area on the 3D view and 2D drawings and provides a parameter comparison view of the conflicting components. This view displays the modification values ​​of the same set of cooling channel parameters by designers A and B side-by-side in a table format, prompting relevant designers to coordinate. After receiving the coordinated confirmation instruction submitted by the designer through the interactive interface, which includes the final adopted parameter values ​​and modification instructions, the system will merge the final coordinated key parameters of the mold structure with the parametric assembly and update it to the master version design data in the BIM database.

[0046] In some embodiments, the system identifies design conflicts using a dual geometric and parametric comparison algorithm. For example, a conflict is identified when two design versions modify the diameter or position coordinates of the same top rod beyond the tolerance range. The detection range and severity of version conflicts can be evaluated using a quantification formula, defined as follows: ; in: This indicates the degree of conflict or overlap between two design modifications. This represents the spatial set of geometric regions in the parametric assembly submitted by designer A that have undergone modifications. This represents the spatial set of geometric regions in the parametric assembly submitted by designer B that have undergone modifications. This represents the spatial volume occupied by the intersection of two modified region spatial sets. This represents the spatial volume occupied by the union of two modified region sets, when... When the value is greater than zero, the system determines that there is a design conflict.

[0047] Understandably, the virtual verification module intervenes after the collaborative design is completed. Based on the parametric assembly of the mold structure finally determined in the BIM database, a high-fidelity digital twin model of the mold is created. The mold digital twin model is constructed to achieve integrity by mapping all the geometry of the parametric assembly, assigning realistic mold steel material properties, and applying loads and boundary constraints obtained from the mold design baseline data. Molding process parameters completely consistent with those in the initial design constraints are applied to the mold digital twin model. The injection pressure is set to 80 MPa, and the melt temperature is set to 240 degrees Celsius. Finite element simulation of the entire injection molding process is performed, including four stages: filling, holding pressure, cooling, and warpage. From the finite element simulation results, stress cloud maps, deformation cloud maps, temperature field distribution, and melt flow front data of the mold under simulated working conditions are extracted. The stress cloud map displays the equivalent stress distribution inside the mold in a color-mapped format, and the deformation cloud map shows the displacement amplification effect of the mold under stress. Optionally, the extracted stress cloud map, deformation cloud map, temperature field distribution, and melt filling flow front data are compared and verified with the maximum allowable stress, maximum allowable deformation, temperature uniformity requirements, and flow balance targets specified in the mold performance indicators. The comparison and verification are accomplished by calculating the difference between the simulation results and the target values. In some embodiments, the comparison and verification results generate a verification report. The verification report lists the preset target value and simulation prediction value of each indicator in the form of a structured data list. This verification report serves as the final closed-loop feedback output to the designer in the mold structure parametric design process.

[0048] See Figure 5This is a chart analyzing the effectiveness of BIM mold collaborative version management, visually reflecting the changes in the number of modifications, conflict detection rate, and collaboration efficiency during the design collaboration process from version 1.0 to version 5.0. The number of modifications steadily increased with each version iteration, from 5 to 18 times. This reflects the depth and frequency of design iterations; multiple modifications are an inevitable process for continuous optimization of the mold structure to meet performance indicators, and also reflects the activity level of multiple personnel operating in parallel during collaborative design. The conflict detection rate steadily improved with each version iteration, from 85% to 98%, approaching full detection. This indicates that the conflict detection capability of the collaborative version management module has gradually strengthened, enabling it to more comprehensively identify parameter and assembly conflicts when multiple designers operate in parallel, providing a reliable basis for design coordination. Collaboration efficiency steadily increased with each version iteration, from 75% to 96%. Despite the increase in the number of modifications, the improved conflict detection rate, along with functions such as automatic conflict marking and parameter comparison views, significantly reduced coordination costs, ultimately achieving a steady increase in collaboration efficiency, demonstrating the value of the collaborative version management module.

[0049] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A precision mold structure parametric design system based on BIM technology, characterized in that, include: The condition input module receives the initial design constraints of the target precision mold, which include the product's three-dimensional geometric model, material physical parameters, molding process parameters, and mold performance indicators. The data processing module, based on the three-dimensional geometric model of the product, constructs the corresponding mold cavity geometric features, and combines the material physical parameters and the molding process parameters to extract the load distribution and thermal stress field characteristics of the mold structure through inference calculation, thereby obtaining the mold design benchmark data. The parameter inversion solution module, based on the mold design benchmark data, uses a state inversion algorithm to solve for the key parameters of the mold structure that satisfy the mold performance indicators; The parametric modeling module generates a three-dimensional skeleton model of the mold structure, including mold frame size, insert layout, cooling water channel and ejection mechanism position, based on the key parameters of the mold structure obtained by solving. The three-dimensional skeleton model of the mold structure is then assembled with the geometric features of the mold cavity to obtain a parametric assembly of the mold structure. The simulation analysis feedback module performs interference checks and motion simulation analysis on the parameterized assembly of the mold structure, extracts its interference state and motion trajectory data, and feeds it back to the inference calculation module to update the mold design benchmark data, thereby performing inverse solving and optimization of the key parameters of the mold structure again.

2. The precision mold structure parametric design system based on BIM technology as described in claim 1, characterized in that, Based on the product's three-dimensional geometric model, the corresponding mold cavity geometric features are constructed. Combining the material physical parameters and the molding process parameters, inference calculations are used to extract the load distribution and thermal stress field characteristics of the mold structure, including: The parting surface of the product's three-dimensional geometric model is automatically identified and created to obtain the dividing boundary between the moving mold part and the fixed mold part of the mold, and the mold cavity geometric features are generated based on the dividing boundary. Extract the geometric complexity of the surface from the geometric features of the mold cavity, read the material shrinkage rate from the material physical parameters, and read the injection pressure and melt temperature from the molding process parameters; The geometric complexity of the surface, material shrinkage rate, injection pressure and melt temperature are used as conditional variables for inference calculation and input into a trained load distribution prediction network. The load distribution prediction network outputs a dynamic load distribution map of the mold cavity during the injection stage. Meanwhile, the mold cavity geometry, melt temperature, and molding cycle time are input into another trained thermal stress prediction network, which outputs the steady-state and transient thermal stress field characteristics of the mold under continuous working conditions.

3. The precision mold structure parametric design system based on BIM technology as described in claim 2, characterized in that, The process of solving the key parameters of the mold structure that satisfy the mold performance indicators using the state inversion algorithm includes: The state inversion algorithm constructs an optimization problem with mold performance indicators as the objective function and mold design benchmark data as the constraint, and iteratively solves it to obtain the key parameters of the mold structure. The quantitative requirements for maximum mold deformation, mold service life, and cooling efficiency are read from the mold performance indicators. Define a parameterized mold structure model, whose model variables are the key parameters of the mold structure to be solved. The mold structure model variables include at least the mold base thickness, the insert fixing method, the diameter and spacing of the cooling water channels, and the diameter and distribution density of the ejector pins. Based on the parameterized mold structure model and the mold design benchmark data, a forward simulation model is constructed. The forward simulation model can calculate the predicted maximum mold deformation, mold service life and cooling efficiency according to the input key mold structure parameters. The quantitative requirements of the mold performance index are set as the target state of the state inversion algorithm, and the key parameters of the mold structure are set as inversion variables. The key parameters of the mold structure that minimize the norm of the difference between the output of the forward simulation model and the target state are solved by the state inversion algorithm.

4. The precision mold structure parametric design system based on BIM technology as described in claim 3, characterized in that, Based on the key parameters of the mold structure obtained from the solution, a three-dimensional skeleton model of the mold structure is generated, including the mold base dimensions, insert layout, cooling channels, and ejection mechanism positions. This three-dimensional skeleton model is then assembled with the geometric features of the mold cavity, including: Call the preset BIM component library, and based on the mold frame dimensions in the key parameters of the mold structure, retrieve or parameterize a standard mold frame 3D model from the BIM component library; Based on the insert layout in the key parameters of the mold structure, parameterized insert models are inserted at predetermined positions in the standard mold frame 3D model, and assembly constraint relationships between the insert models and the mold frame are established. Based on the cooling channel diameter and spacing in the key parameters of the mold structure, three-dimensional models of cooling pipes are arranged along the preset hot spot areas in the standard mold frame three-dimensional model and the insert model. Based on the ejector pin diameter and distribution density in the key parameters of the mold structure, three-dimensional models of ejector pins and reset pins are arranged in the moving mold part of the standard mold frame three-dimensional model to form an ejection mechanism; The entire three-dimensional model, including the mold base, inserts, cooling pipes, and ejection mechanism, is defined as the three-dimensional skeleton model of the mold structure. This three-dimensional skeleton model of the mold structure is then integrated with the geometric features of the mold cavity through coordinate alignment and Boolean operations to obtain the parametric assembly of the mold structure.

5. The precision mold structure parametric design system based on BIM technology as described in claim 4, characterized in that, Interference checks and motion simulation analyses are performed on the parameterized assembly of the mold structure to extract its interference state and motion trajectory data, including: Static geometric collision detection is performed on all components of the moving mold part, fixed mold part, ejection mechanism and cooling pipe in the parameterized assembly of the mold structure to identify whether there are static interference areas between components, and the static interference position and interference amount are recorded. A driving command for mold opening and closing motion is applied to the parameterized assembly of the mold structure, driving the moving mold part and the ejection mechanism to move according to a preset speed curve; During the movement, the system detects in real time whether there is dynamic interference between the moving components and the fixed components, and records the time, location and component information of the dynamic interference. Simultaneously, the motion trajectory coordinate sequence of the push rod, slider, and inclined push rod moving components is tracked and recorded as motion trajectory data.

6. The precision mold structure parametric design system based on BIM technology as described in claim 5, characterized in that, The interference state and motion trajectory data are extracted and fed back to the inference calculation module to update the mold design baseline data, thereby enabling a second inversion solution and optimization of the key parameters of the mold structure, including: The location and interference amount of the identified static interference area are converted into boundary constraint modification instructions for the geometric features of the mold cavity in the mold design reference data. The boundary constraint modification instructions are used to shrink or offset the geometric boundaries of the relevant area. The recorded time points of dynamic interference and component information are converted into constraint modification instructions for the load distribution of the mold structure in the mold design reference data. The constraint modification instructions are used to limit the motion envelope space of the relevant components. The recorded motion trajectory data is compared with the preset ideal motion trajectory to generate trajectory deviation data. The trajectory deviation data is used to correct the parameters related to the dynamics of the ejection mechanism in the mold design reference data. The modified and corrected mold design baseline data is re-input into the state inversion algorithm. Using the updated data as constraints, the inversion solution of the key parameters of the mold structure is re-executed to obtain the optimized key parameters of the mold structure.

7. The precision mold structure parametric design system based on BIM technology as described in claim 6, characterized in that, The state inversion algorithm constructs an optimization problem with mold performance indicators as the objective function and mold design baseline data as constraints, and iteratively solves to obtain key parameters of the mold structure, including: Establish an objective function, which is the weighted sum of squares of the differences between the quantified values ​​of the mold performance index and its target value; Establish constraints, which include mathematical expressions for geometric constraints, strength constraints, and thermal constraints derived from mold design baseline data; The gradient descent method is used to initialize the key parameters of the mold structure, and the value of the objective function and the satisfaction of the constraints under the current values ​​are calculated. Based on gradient information and the degree of constraint violation, the key parameters of the mold structure are iteratively updated using a sequential quadratic programming method, so that the objective function value gradually decreases until convergence while satisfying all constraints.

8. The precision mold structure parametric design system based on BIM technology as described in claim 7, characterized in that, Also includes: The collaborative version management module stores the key parameters of the mold structure and the corresponding parametric assembly of the mold structure generated in each iteration as design versions in the BIM database, and records the version number, modified content and associated design constraints. When multiple designers work in parallel, the system compares in real time the parametric assemblies submitted by different designers in the BIM database for the same mold area to identify design conflicts. For identified design conflicts, the system automatically marks the conflict areas and provides a parameter comparison view of the conflicting components, prompting designers to coordinate. After receiving confirmation instructions from the designers, the system will merge the final coordinated key parameters of the mold structure with the parametric assembly and update it to the master version in the BIM database.

9. The precision mold structure parametric design system based on BIM technology as described in claim 8, characterized in that, After re-executing the inversion solution of the key parameters of the mold structure to obtain the optimized key parameters of the mold structure, the process also includes a parameter-driven automatic generation step of mold components: Based on the optimized key parameters of the mold structure, the parametric template components in the BIM component library are automatically updated in size and configuration. The updated parametric template components are automatically reassembled according to the topological relationships defined in the 3D skeleton model of the mold structure to generate a new parametric assembly of the mold structure. The parameterized assembly of the new mold structure is automatically lightweighted, and its three-dimensional geometric information and material property information are extracted to generate mold manufacturing information that can be directly used for CNC machining.

10. The precision mold structure parametric design system based on BIM technology as described in claim 9, characterized in that, Also includes: The virtual verification module creates a high-fidelity digital twin model of the mold based on the parameterized assembly of the final mold structure. The digital twin model contains complete geometric, material, constraint and load information. Apply molding process parameters consistent with the initial design constraints to the digital twin model of the mold, and perform finite element simulation of the entire injection molding process; From the finite element simulation results, extract the stress cloud map, deformation cloud map, temperature field distribution and melt filling flow front data of the mold under the simulated working state; The extracted stress cloud map, deformation cloud map, temperature field distribution and melt filling flow front data are compared and verified with the mold performance indicators, and the verification results are used as the final closed-loop feedback for the parameterized design of the mold structure.

Images