Integrated design method and system for three-dimensional simulation model of plastic mold

Abstract

The invention relates to the technical field of plastic molds, in particular to an integrated design method and system for a three-dimensional simulation model of a plastic mold. The method comprises the following steps: carrying out geometric scanning on a plastic mold cavity, and setting mold foaming simulation configuration data; acquiring processing parameters of the plastic mold; based on the mold foaming simulation configuration data and the plastic mold processing parameters, constructing a plastic mold initial simulation model; performing plastic melt flow-foaming injection simulation by using the initial simulation model of the plastic mold to obtain injection flow field state data; analyzing an abnormal area of the injection flow field according to the state data of the injection flow field; and according to the abnormal area of the injection flow field, abnormal area plastic process parameter optimization is carried out, plastic process optimization parameters are generated, and the integrated design of the injection filling stage of the plastic mold is achieved. Full-time coupling of a mold structure and a flow field is achieved through numerical simulation of the plastic fluid mold, and the quality stability and the yield of a supercritical fluid micro-foaming product are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of plastic molds, and in particular to an integrated design method and system for a three-dimensional simulation model of a plastic mold. Background Art

[0002] Supercritical fluid microfoam injection molding technology is an advanced polymer processing technology. It dissolves a supercritical fluid (usually carbon dioxide CO2 or nitrogen N2) in the molten polymer. Then, during the injection molding process, the rapid changes in pressure and temperature are used to cause the gas dissolved in the melt to precipitate and form a large number of micron-sized bubble nuclei. These bubble nuclei then grow in the mold cavity, eventually forming a plastic product with a microporous structure. However, traditional integrated design methods for three-dimensional simulation models of plastic molds mainly include empirical design methods, two-dimensional flow analysis methods, and simplified three-dimensional simulation methods. The empirical design method relies too much on the designer's subjective experience and lacks scientific basis; the two-dimensional flow analysis method cannot accurately reflect the fluid behavior and temperature distribution in complex three-dimensional structures; and the simplified three-dimensional simulation method, although taking spatial factors into account, ignores the interaction between the supercritical fluid and the plastic melt and the bubble dynamics, resulting in significant differences between the simulation results and the actual situation. It is impossible to achieve precise control of abnormal areas, ultimately leading to unstable product quality and low yield. Summary of the Invention

[0003] Based on this, the present invention provides an integrated design method and system for a three-dimensional simulation model of a plastic mold to solve at least one of the above technical problems.

[0004] To achieve the above objectives, an integrated design method for a three-dimensional simulation model of a plastic mold includes the following steps:

[0005] Step S1: geometrically scanning the plastic mold cavity to obtain mold cavity geometry data; setting mold foaming simulation configuration data based on the mold cavity geometry data; wherein the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology, and candidate foaming injection point locations;

[0006] Step S2: Obtaining plastic mold processing parameters; constructing an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; performing a plastic melt flow-foaming injection simulation using the initial simulation model of the plastic mold to obtain injection flow field state data;

[0007] Step S3: analyzing the injection flow field abnormal area according to the injection flow field state data; identifying the foaming abnormal area based on the injection flow field state data, and performing abnormal space superposition analysis on the injection flow field abnormal area to generate comprehensive flow field abnormal area data;

[0008] Step S4: Optimizing the plastic process parameters in the abnormal area based on the comprehensive flow field abnormal area data, generating plastic process optimization parameters to achieve integrated design of the plastic mold injection filling stage.

[0009] Preferably, the present invention further provides an integrated design system for a three-dimensional simulation model of a plastic mold, which executes the integrated design method for a three-dimensional simulation model of a plastic mold as described above. The integrated design system for a three-dimensional simulation model of a plastic mold includes:

[0010] A mold geometry configuration module is used to perform geometric scanning on the plastic mold cavity to obtain mold cavity geometry data; based on the mold cavity geometry data, mold foaming simulation configuration data is set; wherein, the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology structure, and candidate foaming injection point locations;

[0011] The plastic flow simulation module is used to obtain plastic mold processing parameters; construct an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; and use the initial simulation model of the plastic mold to perform plastic melt flow-foaming injection simulation to obtain injection flow field state data;

[0012] The foaming abnormality diagnosis module is used to analyze the injection flow field abnormal area based on the injection flow field state data; identify the foaming abnormal area based on the injection flow field state data, and perform abnormal space superposition analysis based on the injection flow field abnormal area to generate comprehensive flow field abnormal area data;

[0013] The process parameter optimization module is used to optimize the plastic process parameters in abnormal areas based on the comprehensive flow field abnormal area data, generate plastic process optimization parameters, and realize the integrated design of the plastic mold injection filling stage.

[0014] The present invention establishes a more accurate geometric basic model by accurately capturing the three-dimensional geometric features of the mold cavity, effectively breaking away from the subjective limitations of traditional empirical design. The setting of mold foaming simulation configuration data realizes the accurate description of complex three-dimensional structures. The application of cavity grid data overcomes the shortcomings of two-dimensional analysis in spatial expression. The construction of runner network topology makes the simulation closer to the actual injection molding physical process, and the preset position of candidate foaming injection points provides a scientific basis for subsequent optimization. The introduction of actual processing parameters significantly enhances the practicality and accuracy of the simulation, making the simulation model more in line with actual production conditions. The innovation of plastic melt flow-foaming injection simulation lies in the simultaneous consideration of melt flow characteristics and bubble formation dynamics, which makes up for the key defect of traditional simplified three-dimensional simulation that ignores the interaction between supercritical fluid and melt. The injection flow field state data obtained can fully reflect the pressure, temperature, velocity distribution of the melt in the cavity, as well as the bubble nucleation and growth characteristics. The analysis of abnormal flow field areas can accurately identify potential problem points, especially with high predictive ability in common defects such as uneven bubble distribution, weld line formation, and warping. Identifying abnormal foaming areas solves the problem of inaccurate prediction of microbubble formation by traditional methods, while abnormal spatial superposition analysis enables comprehensive diagnosis of multi-dimensional problems and reveals the interaction mechanism between different types of defects. Process parameter optimization based on comprehensive flow field abnormal area data specifically adjusts key parameters such as injection speed, mold temperature, and holding pressure, no longer relying on empirical judgment, but scientifically regulating based on precise calculation results. This precise optimization strategy for abnormal areas effectively solves common problems in microbubble structures, such as uneven bubble size, uneven distribution, and poor interface bonding. Therefore, the present invention provides an integrated design method for a three-dimensional simulation model of a plastic mold. By acquiring three-dimensional geometric data of the mold cavity, combining processing parameters and supercritical fluid injection technology to construct an initial simulation model, and using high-precision laser scanning to generate point cloud data, the data is processed to form an optimized grid and flow channel topology structure. At the same time, abnormal area identification is achieved through melt flow, temperature, pressure, and shear simulation, and then the injection and holding pressure parameters are automatically adjusted to achieve intelligent optimization of process parameters, thereby accurately controlling the mold injection filling process, improving simulation accuracy and product quality, shortening the design cycle, and reducing production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic flow chart of the steps of an integrated design method for a three-dimensional simulation model of a plastic mold according to the present invention;

[0016] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0019] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0020] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0021] To achieve this, please refer to Figures 1 to 2 The present invention provides an integrated design method for a three-dimensional simulation model of a plastic mold, comprising the following steps:

[0022] Step S1: geometrically scanning the plastic mold cavity to obtain mold cavity geometry data; setting mold foaming simulation configuration data based on the mold cavity geometry data; wherein the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology, and candidate foaming injection point locations;

[0023] Step S2: Obtaining plastic mold processing parameters; constructing an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; performing a plastic melt flow-foaming injection simulation using the initial simulation model of the plastic mold to obtain injection flow field state data;

[0024] Step S3: analyzing the injection flow field abnormal area according to the injection flow field state data; identifying the foaming abnormal area based on the injection flow field state data, and performing abnormal space superposition analysis on the injection flow field abnormal area to generate comprehensive flow field abnormal area data;

[0025] Step S4: Optimizing the plastic process parameters in the abnormal area based on the comprehensive flow field abnormal area data, generating plastic process optimization parameters to achieve integrated design of the plastic mold injection filling stage.

[0026] In an embodiment of the present invention, the integrated design method for a three-dimensional simulation model of a plastic mold includes the following steps:

[0027] Step S1: geometrically scanning the plastic mold cavity to obtain mold cavity geometry data; setting mold foaming simulation configuration data based on the mold cavity geometry data; wherein the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology, and candidate foaming injection point locations;

[0028] In the present invention, for example, this embodiment uses a foam part for an automobile dashboard bracket as an example. First, a constant-speed industrial-grade 3D laser scanner is used to perform a geometric scan of the dashboard bracket mold cavity. The scanning frequency is set to 50Hz and the laser power is set to 2.5W. During the scanning process, the scanning head is kept within a range of 150±5mm from the cavity surface. A spiral layer-by-layer scanning path is used with an overlap rate of 30%, ensuring a point density of 120 points / cm. 2 . The acquired cavity point cloud data is filtered, and the distance threshold between points is set to 3 times the standard deviation. Points with a distance exceeding 0.15mm are deleted, and then the smoothing coefficient is set to 0.8 to smooth the point cloud. The processed point cloud is converted into a three-dimensional surface model, and the cavity size parameters (length 480mm × width 320mm × height 85mm), wall thickness range (1.8-2.5mm) and chamfer radius (2.0mm) are determined. The mold structure is identified in combination with the cavity parameters, including the two inlet positions on both sides of the bottom, the conical gate (diameter 5mm, taper 5°), the "Y"-shaped branch runner layout, and 6 groups of cooling water channels (diameter 8mm) distributed along the bracket structure. The cavity parameters and mold structure information are integrated to generate complete mold cavity geometry data. Surface meshing is performed based on the data, and the mesh size is set to 1.0±0.2mm, the curvature sensitivity is 0.2, and the mesh quality check ensures that the minimum angle is not less than 30 degrees and the maximum angle is not more than 150 degrees. The optimized surface mesh is converted into a volume mesh, and the cavity flow channel network topology is constructed. The distance from each part of the cavity to the inlet is calculated, the flow path is analyzed, and finally the candidate foaming injection point position is extracted 35 mm below the intersection of the Y-shaped flow channel.

[0029] Step S2: Obtaining plastic mold processing parameters; constructing an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; performing a plastic melt flow-foaming injection simulation using the initial simulation model of the plastic mold to obtain injection flow field state data;

[0030] In the embodiment of the present invention, the mold steel used for the instrument panel bracket is NAK80 steel, which has a thermal conductivity of 41W / (m·K), a specific heat capacity of 490J / (kg·K), and a density of 7.92g / cm 3 , hardness is 40HRC. The plastic material is polypropylene (PP) foam, the melting temperature is set to 230℃, and the melt density is 0.73g / cm 3 , in 100s -1 The viscosity under shear rate is 3200 Pa·s, the specific heat capacity is 2.0 kJ / (kg·K), and the crystallization temperature is 110°C. Carbon dioxide (CO2) is selected as the supercritical fluid, with an injection ratio of 1.2wt%, a critical temperature of 31.1°C, a critical pressure of 7.38 MPa, a solubility of 1.0wt% in PP at 230°C and 20 MPa, and a diffusion coefficient of 1.2×10 -9 m 2 / s. Based on the supercritical fluid parameters, the injection point position is processed, and the injection pressure is set to 23MPa and the injection speed is 28cm 3 / s, holding pressure is 18MPa, holding time is 6 seconds, and cooling time is 25 seconds. All parameters are imported into the simulation system, and the boundary conditions are set: injection pressure at the gate is 23MPa, cavity wall temperature is 65℃, and there is no slip wall condition. A three-dimensional finite element model is established, which contains 87652 volume mesh units and 102356 nodes. The four-stage simulation of filling-holding-cooling-foaming is performed, with a time step of 0.01s in the filling stage, 0.05s in the holding stage, an adaptive time step (0.1-1.0s) in the cooling stage, and a time step of 0.005s in the foaming stage. Data is collected every 0.05s at 40 evenly distributed monitoring points to record the temperature field (range 190-280℃), pressure field (range 0-30MPa), flow velocity field (range 0-150cm / s) and shear rate field (range 0-10000s -1 ) data. The data is integrated to generate injection flow field state data, including the filling ratio, cavity pressure distribution, temperature distribution, flow front position and shear stress distribution at each time point.

[0031] Step S3: analyzing the injection flow field abnormal area according to the injection flow field state data; identifying the foaming abnormal area based on the injection flow field state data, and performing abnormal space superposition analysis on the injection flow field abnormal area to generate comprehensive flow field abnormal area data;

[0032] In an embodiment of the present invention, four types of abnormal areas of the instrument panel bracket are analyzed based on the injection flow field state data. Temperature analysis shows that there is an area on the side wall of the bracket with a temperature below 95°C (15°C lower than the crystallization temperature of PP), and the standard deviation of the temperature in the middle of the bracket is 7.2°C, which is marked as a temperature abnormality area. Pressure analysis shows that the pressure at the top of the bracket away from the gate is only 13.5MPa, which is lower than 80% of the set holding pressure (18MPa), and there is an area at the corner with a pressure gradient exceeding 14MPa / mm, which is marked as a pressure abnormality area. Flow velocity analysis shows that the flow velocity direction at the connection between the bracket groove and the reinforcement rib deviates from the mainstream direction by 56 degrees and the flow velocity is as low as 0.3cm / s, which is marked as a flow velocity abnormality area. Shear analysis found that the shear rate near the gate was as high as 2300s -1 The duration is 2.1s, which is marked as the shear abnormal area. The foaming related parameters are calculated in parallel, and the bubble nucleation rate is calculated using the temperature field and pressure field data. The area with the highest nucleation rate is located in the middle of the bracket, reaching 10 25 m -3 s -1 The average initial density of bubbles is calculated to be 8×10 6 cells / cm 3 The initial size is 15-25 μm. According to the calculation of bubble growth dynamics, the final bubble diameter reaches 75-130 μm, with a density of 3×10 6 -9×10 6 cells / cm 3 Compared with the preset ideal foaming parameters (bubble diameter 80μm, density 5×10 6 cells / cm 3 ) comparison revealed areas on the sidewalls and top of the bracket where bubble diameter deviations exceeded 40% and density deviations exceeded 50%, marking them as abnormal foaming areas. Spatial overlay analysis of abnormal areas was performed, and comprehensive anomaly indices were calculated for the overlapping regions. The index for the area where temperature and pressure anomalies overlapped at the top of the bracket reached 0.62, the index for the area where temperature anomalies and foaming anomalies overlapped at the sidewalls reached 0.57, and the index for the area where shear anomalies near the gate overlapped reached 0.53. These areas were all marked as high-risk, generating comprehensive flow field abnormality area data.

[0033] Step S4: Optimizing the plastic process parameters in the abnormal area based on the comprehensive flow field abnormal area data, generating plastic process optimization parameters to achieve integrated design of the plastic mold injection filling stage.

[0034] In an embodiment of the present invention, based on the data of the abnormal area of the comprehensive flow field of the instrument panel bracket, the contribution index of the abnormal factors in each area is calculated. The temperature abnormality index of the top area of the bracket is 0.72, the pressure abnormality index is 0.68, the flow velocity abnormality index is 0.31, and the shear abnormality index is 0.18, which is determined to be a temperature-pressure dual-dominated abnormal area. The temperature abnormality index of the side wall area is 0.65, the pressure abnormality index is 0.42, the flow velocity abnormality index is 0.28, and the shear abnormality index is 0.15, which is determined to be a temperature-dominated abnormal area. The temperature abnormality index of the area near the gate is 0.25, the pressure abnormality index is 0.33, the flow velocity abnormality index is 0.28, and the shear abnormality index is 0.61, which is determined to be a shear-dominated abnormal area. The temperature abnormality index at the connection between the groove and the reinforcement rib is 0.37, the pressure abnormality index is 0.45, the flow velocity abnormality index is 0.58, and the shear abnormality index is 0.22, which is determined to be a flow velocity-dominated abnormal area. Parameters of various abnormal areas were optimized: the melt temperature in the temperature-dominant area was adjusted from 230°C to 245°C, and the mold temperature was increased from 65°C to 75°C; the injection pressure curve in the pressure-dominant area was adjusted, with the initial filling pressure maintained at 23MPa, increased to 25MPa in the middle and 27MPa in the later stages, the holding pressure increased from 18MPa to 21MPa, and the holding time extended from 6 seconds to 8 seconds; the injection speed curve in the flow rate-dominant area was optimized, with the initial filling speed at 25cm / s. 3 / s, increased to 35cm in the middle stage 3 / s, later reduced to 20cm 3 / s; in the shear-dominant region, the gate diameter was increased from 5mm to 6.5mm, and the taper was reduced from 5° to 3°. After checking for parameter conflicts and integrating the optimization results, the final plastic process optimization parameters were formed: melt temperature 245°C, mold temperature 75°C, segmented injection pressure 23-27MPa, and segmented injection speed 20-35cm. 3 / s, holding pressure 21MPa, holding time 8s, gate diameter 6.5mm, taper 3°, supercritical fluid injection ratio reduced from 1.2wt% to 1.0wt%, completing the integrated optimization design of instrument panel bracket foam injection molding process parameters.

[0035] Preferably, performing geometric scanning on the plastic mold cavity in step S1 includes:

[0036] The plastic mold cavity is geometrically scanned by a 3D laser scanner. The scanning frequency is set to 50 Hz, the laser power is 2.5 W, and the point density is not less than 100 points / cm 2 , obtain the point cloud data of the inner surface of the cavity;

[0037] The point cloud data of the inner surface of the cavity is filtered, and the filtering threshold is set to 3 times the standard deviation of the distance between points. When the distance between points is greater than 0.15 mm, it is determined as a noise point and deleted to obtain the filtered cavity point cloud data;

[0038] Set the smoothing coefficient to 0.8, smooth the filtered cavity point cloud data, and obtain accurate cavity point cloud data;

[0039] Convert the precise cavity point cloud data into a 3D surface model, determine the key dimensional parameters of the cavity such as length, width, depth, wall thickness and chamfer radius, and obtain the cavity parameters;

[0040] Identify the mold's inlet position, gate shape and size, runner layout, and cooling channel location based on cavity parameters to generate the mold structure;

[0041] The cavity parameters and mold structure are integrated to obtain the mold cavity geometric structure data.

[0042] In an embodiment of the present invention, a constant-speed industrial-grade three-dimensional laser scanner is used to perform geometric scanning on the plastic mold cavity. The scanner is fixed on a three-axis precision robotic arm, and the scanning frequency is set to 50Hz. The frequency is determined by testing the reflection characteristics of the plastic mold material. The reflection test is performed in the 400-700nm band using a standard spectrometer. The laser power is set to 2.5W. This power value is based on a comprehensive measurement of the heat resistance and reflective properties of the plastic mold to avoid deformation caused by increased mold surface temperature due to excessive power. During the scanning process, the distance between the scanning head and the cavity surface is controlled to be within the range of 150±5mm. The scanning path is designed to go deeper layer by layer in a spiral manner. The scanning overlap rate of each layer is 30%, ensuring that the point density is not less than 100 points / cm 2 . The scanning head movement speed is controlled at 5mm / s, and it takes about 45 minutes to complete a complete cavity scan. Data acquisition adopts real-time transmission mode, and the acquisition signal is transmitted to the data processing workstation through a high-speed data line to form the original cavity inner surface point cloud data containing coordinate information and reflection intensity information. Calculate the Euclidean distance between all adjacent points in the point cloud data set and record it as the distance matrix D. According to the distance matrix D, the mean μ and standard deviation σ of the distance between points are calculated. When it is detected that the distance between any two points is greater than 0.15mm, the point is marked as a noise point. This critical value of 0.15mm is determined according to the mold surface finish and scanning accuracy, and the corresponding mold polishing level is P3. The regional growth algorithm is used to partition the point cloud and perform a deletion operation on the points marked as noise. The deletion operation adopts a secondary confirmation mechanism, that is, deletion is performed only when the distance between a point and at least 5 surrounding points exceeds the threshold. After the noise point deletion is completed, the point cloud data is re-indexed and numbered to generate a filtered cavity point cloud data file, and the data density is maintained at 95-98 points / cm 2The smoothing coefficient is set to 0.8, which is determined by reverse deduction of the mold manufacturing accuracy requirements, and the manufacturing accuracy is ±0.01mm. The smoothing process is performed in three iterations. The first time, a large area Laplace operator (10mm×10mm window) is used for coarse smoothing, the second time, a medium area Laplace operator (5mm×5mm window) is used for moderate smoothing, and the third time, a small area Laplace operator (2mm×2mm window) is used for fine smoothing. The displacement of the point is calculated after each smoothing to ensure that the maximum displacement does not exceed 0.05mm. During the smoothing process, a feature weight protection mechanism is used for edge feature points, and the judgment standard for edge feature points is that the curvature change rate exceeds 0.5 / mm. For feature areas such as chamfers and fillets, a curvature preservation algorithm is applied to retain the original geometric features. Feature recognition is performed on point cloud data to extract the topological features of the cavity, including boundary lines, feature lines, and symmetry axes. Feature line extraction uses a combination of regional growing method and curvature threshold method, and the threshold is set to 0.8 / mm. The entire cavity is divided into multiple surface patches according to the feature lines. The number of control points of each surface patch is automatically determined according to the complexity of the surface, and the complexity is calculated by the variance of Gaussian curvature. NURBS fitting is performed on each surface patch, and the fitting accuracy is controlled within ±0.02mm. During the fitting process, the weight factor is dynamically adjusted according to the density of the point cloud. After the surface fitting is completed, the adjacent surfaces are spliced, and the splicing error is controlled within 0.01mm. The least squares method is used to optimize the key dimensions and determine the key dimensional parameters of the cavity, such as length, width, depth, wall thickness and chamfer radius. For regular geometric bodies such as cylinders and planes, the geometric constraint fitting method is used to improve the accuracy, and finally a complete parametric three-dimensional surface model is formed. The cavity parameters recorded include: cavity size parameters (L×W×H), wall thickness distribution map (thickness range 0.8-2.5mm), and chamfer radius distribution map (radius range 0.5-3.0mm). The entrance position of the mold is identified by curvature analysis and normal vector change detection. The entrance position feature is manifested as a concentrated area of curvature mutation points, and the curvature change rate is greater than 1.5 / mm. The gate shape recognition adopts the cross-section fitting method to extract multiple sections along the entrance position, and the gate shape is determined by roundness and ellipticity analysis. The roundness is defined as the ratio of the radius of the maximum inscribed circle to the minimum circumscribed circle. For the gate size, the diameter and taper at the narrowest point of the section are calculated by differential geometry, and the diameter accuracy is controlled at ±0.05mm. The runner layout is identified by combining the change in the thickness of the inner wall of the cavity and the thermal flow analysis. First, a wall thickness gradient map is constructed, and the area with a wall thickness gradient exceeding 0.2mm / mm is marked as a candidate runner area. The cooling water channel position is determined by hot spot analysis. The hot spot area is defined as a wall thickness exceeding 2.0mm and a surface area greater than 5cm 2Continuous area. The optimal cooling water channel path is calculated for each hot spot area, and the water channel diameter is determined based on heat calculation, with a diameter range of 6-12mm. Based on the above recognition results, a complete mold structure feature model is constructed, including gate position coordinates, gate shape parameters (diameter, taper, length), runner layout schematic and cooling water channel layout diagram. A layered fusion strategy is adopted. A hierarchical data structure is established, with the first layer being the cavity geometry surface data, the second layer being the key dimension parameter data, and the third layer being the mold structure feature data. The fusion process first establishes a common coordinate system, converts all data to the same coordinate system, and sets the origin of the coordinate system at the center point of the cavity bottom surface. A spatial position constraint relationship is established between the mold structure feature and the cavity, the gate position is associated with the cavity inlet coordinates, the runner is associated with the cavity wall thickness distribution, and the cooling water channel is associated with the hot spot area.

[0043] Preferably, setting the mold foaming simulation configuration data according to the mold cavity geometry data in step S1 includes:

[0044] Surface meshing is performed based on the cavity parameters in the mold cavity geometry data. The mesh size is set to 1.0±0.2mm, the curvature sensitivity is set to 0.2, and the surface mesh data is generated.

[0045] Perform quality inspection on the surface mesh data to ensure that the minimum angle of the surface mesh is not less than 30 degrees and the maximum angle is not greater than 150 degrees, and obtain mesh quality inspection data;

[0046] Re-optimize the surface mesh data based on the mesh quality test data and perform volume mesh conversion to obtain the cavity mesh data;

[0047] Constructing the cavity flow channel network topology structure according to the mold structure in the mold cavity geometry data;

[0048] Based on the cavity mesh data and the cavity flow channel network topology structure, the distance from each part of the cavity to the inlet is calculated to obtain the inlet distance distribution data;

[0049] Calculate the cavity gate distance gradient vector field based on the inlet distance distribution data, determine the flow length path in the cavity, and generate flow path data;

[0050] Candidate foaming injection point locations are extracted based on flow path data.

[0051] In an embodiment of the present invention, the NURBS surface data in the cavity parameters is extracted to establish a complete cavity boundary representation. The mesh division adopts the side length control method, and the base mesh size is set to 1.0±0.2mm. The mesh size is determined based on 1 / 10 of the minimum feature size of the cavity. At the same time, the curvature sensitivity parameter 0.2 is introduced. This parameter indicates that in areas with large curvature changes, the mesh size is automatically adjusted according to the curvature multiplied by the sensitivity coefficient. The calculation formula is: local mesh size = base mesh size / (1+0.2×local curvature). The mesh generation process adopts the advancing frontier method, starting from the cavity boundary and gradually generating triangular units inward, and the side length of each triangle is controlled within the range of 0.8-1.2mm. For high curvature areas such as chamfers and fine features, the mesh size is automatically reduced to 0.5mm to ensure accurate capture of geometric features. In flat areas, the mesh size is automatically increased to 1.2mm to improve computational efficiency. The entire meshing process uses an iterative algorithm to control the mesh quality and ultimately generates surface mesh data composed of high-quality triangular units. The typical number of mesh units is in the range of 50,000-100,000, and the specific number depends on the complexity of the cavity. Check the minimum and maximum angles of the mesh units and use the cosine theorem to calculate the three internal angles of each triangular unit to ensure that the minimum angle is not less than 30 degrees and the maximum angle is not greater than 150 degrees. This indicator is calculated by the formula cos(θ)=(a 2 +b 2 -c 2) / (2ab), where a and b are the lengths of the two sides of the triangle, c is the length of the opposite side, and θ is the angle. Secondly, the aspect ratio of the mesh cells is calculated, defined as the ratio of the triangle's circumcircle radius to its incircle radius, with the requirement that the aspect ratio should not exceed 5:1. Thirdly, the area change rate of the mesh cells is checked, and the area ratio of adjacent cells is controlled within the range of 0.5-2.0. A topological verification algorithm is used to check mesh connectivity, ensuring that the mesh is free of dangling points, overlapping cells, and non-manifold edges. Each detected overlapping cell is marked and its coordinates recorded. Mesh density is assessed in areas with drastic curvature changes, with the curvature to mesh density ratio controlled within 5:1. For triangular cells with unacceptable angles (less than 30 degrees or greater than 150 degrees), Laplace smoothing combined with local reconstruction is used for optimization, with a value of 1 / n (n is the number of adjacent nodes). For cells with excessive aspect ratios, edge collapse and node insertion are employed. When the ratio of the longest to shortest side of a triangle exceeds 5, a new node is inserted at the midpoint of the longest side, and the connection is reconnected to form two new triangles. For areas with excessive area change rates, local densification is performed, adding nodes within triangles whose area is twice that of the surrounding elements. Non-manifold areas are processed using a topology repair algorithm, which removes duplicate faces, closes small holes, and reconnects edges. The optimized surface mesh is converted to a volume mesh. A kernel growing method is used to generate tetrahedral elements layer by layer from the surface toward the interior. The interlayer spacing is controlled at 1.0 mm, and the ratio of adjacent layer thicknesses is controlled between 0.8 and 1.2. The tetrahedral quality is assessed by the minimum dihedral angle, which is required to be greater than 20 degrees. During volume mesh generation, the cavity surface boundary remains unchanged, and only the internal nodes are adjusted and optimized. The final cavity mesh data is generated, containing the optimized surface triangular mesh. The gate location, runner layout, and primary and secondary runner data from the mold structure are extracted to determine the physical connectivity of the runner system. The runner network is represented using a directed graph structure G(V, E), where the node set V represents key locations (such as gates, diversion points, and confluence points) and the edge set E represents the runner segments. Each node contains three-dimensional coordinate information and node type identification (entry node, diversion node, end node), and each edge contains runner length, diameter, and shape parameter information. The main runner adopts a decreasing diameter topology description. The starting diameter of the main runner is set to the gate diameter, and the diameter decreases by 10% every 50mm along the flow direction until the minimum diameter of 4mm at the end. The secondary runner adopts a constant diameter design. The diameter value is determined according to the viscosity and flow characteristics of the injection molding material, and the general value range is 3-5mm. For complex cavities, a multi-level runner network is constructed. Each level of branch is controlled by node degree. The degree of the main node does not exceed 4, and the degree of the secondary node does not exceed 3. Rounded corners are used at the intersection of the runners, and the radius of the fillet is 1.5 times the runner diameter. The connection point between the runner and the cavity at the gate is marked as the network entry node, and the entry cross-sectional area and entry coordinates are set. The cavity runner network topology data includes a complete node-edge connection relationship matrix, a runner geometry parameter table, and a network hierarchy diagram.Based on the constructed cavity mesh data and runner network topology, the distance distribution from each cavity location to the gate is calculated. This calculation utilizes a modified version of Dijkstra's shortest path algorithm, treating mesh nodes as nodes in a graph and mesh edges as edges. Distance calculations are performed in two stages: In the first stage, the Euclidean distance from each node on the cavity surface mesh to the nearest gate entry point is calculated using the formula d1(i) = |P. i ―P 入口 |, where P i represents the coordinates of node i, P 入口 Indicates the coordinates of the entry point; in the second stage, considering the influence of the flow channel network, the actual flow distance of the fluid is calculated, and the formula is d2(i)=d1(i)+∑w j L j , where L j is the length of the jth flow channel in the flow channel network, w j is the flow channel resistance weight factor, which is related to the flow channel diameter and shape, and the calculation formula is w j =1+5(D0 / D j -1) 2 , D0 is the standard diameter (take 6mm), D j is the actual flow channel diameter. For multi-inlet cavities, the nearest inlet principle is adopted and the minimum value of the calculated results of each inlet is taken. The cavity wall thickness influence factor is introduced. The thicker the wall, the smaller the flow resistance in the area. The correction formula is d 修正 (i)=d2(i)×(t0 / t i )^0.5, where t0 is the standard wall thickness (take 1.5mm), t i is the actual wall thickness at node i. The distance calculation results are normalized, and the maximum distance is standardized to 1.0 to generate the entrance distance distribution data of each part of the cavity. For node i on the surface grid, the gradient vector calculation formula is: Where d(i) and d(j) are the distances between node i and its adjacent node j, respectively. j represents the position coordinates of the adjacent node j (point coordinates in three-dimensional space), n ij is the unit vector from node i to node j, w ij is the weight coefficient, which is 1 / |P i -P j For internal nodes of the volume mesh, the gradient is calculated using tetrahedral element interpolation, using the formula: where β i is the shape function weight, which is related to the position of the node in the tetrahedron. After the gradient vector field is generated, regularization is performed to ensure that the vector modulus is evenly distributed. The regularization formula is: α is the adjustment coefficient, and its value is 0.7. Special treatment is performed on high curvature areas. When the regional curvature exceeds 1.5 / mm, the gradient vector is weakened by 50% along the normal component of the surface. The gradient vector field is used to indicate the flow direction of the fluid in the cavity. The gradient direction points to the direction where the distance increases fastest, that is, the direction where the fluid resistance is the largest. Combined with the distance information, the flow length path in the cavity is defined as a set of paths starting from the entry point and moving in the opposite direction of the gradient vector field. Each path consists of a path line and a path distance, and the path distance reflects the filling time. The generated flow path data contains multiple path lines from the entry to each end of the cavity. Each path line consists of a sequence of discrete points, and the spacing between points is controlled at 0.5mm. Each flow path is divided into 10 segments according to the cumulative distance, and the potential injection point is marked at the midpoint of each segment to obtain the initial candidate point set. Each candidate point is evaluated for multiple metrics: First, the shortest distance from the injection point to the cavity boundary is calculated. This distance must be at least twice the wall thickness to ensure structural strength. Second, the curvature of the isosurface at the injection point is calculated. A smaller curvature value indicates a flatter geometry. Third, the average distance from the injection point to the endpoints of each flow path is calculated. A smaller standard deviation indicates better filling uniformity. Fourth, the wall thickness of the flow path segment where the injection point is located is calculated. A wall thickness between 1.5 and 2.5 mm is optimal. Thinner walls result in insufficient strength, while thicker walls increase cooling time. A weighted composite score is applied to these four metrics in a 3:2:4:2 weighting ratio. After normalization, a composite score is obtained for each candidate point. The top three scoring points are selected as primary candidate injection points based on descending scores. For each injection point, its 3D coordinates, normal vector, composite score, and individual metric scores are recorded. For complex cavities, the regional segmentation method is used to divide the cavity into multiple relatively independent regions. Each region independently performs the above extraction process, and then global coordination is performed to ensure that the injection points are evenly distributed, and finally a candidate foaming injection point data table is generated.

[0052] Preferably, the cavity flow channel network topology structure is constructed according to the mold structure in the mold cavity geometric structure data as follows:

[0053] Based on the cavity mesh data, the mold structure in the mold cavity geometry data is used to extract the center path of each runner segment and gate to obtain the runner centerline data;

[0054] Mark the runner entrance, bifurcation point, confluence point and gate center point according to the runner centerline data to obtain the runner topology node set;

[0055] Analyze the connection relationship of flow channel nodes based on the flow channel topology node set and flow channel centerline data;

[0056] Based on the connection relationship of the flow channel nodes and the flow channel topology node set, an adjacency table is constructed with nodes as vertices and connection relationships as edges to obtain the initial flow channel topology data;

[0057] Assign attributes to the initial topological data of the runner to generate the cavity runner network topology structure.

[0058] In the embodiment of the present invention, the volume grid subset of the runner system is separated from the mold cavity geometry data, and the density threshold is set to 1.2 g / cm 3 (Typical value of plastic density) Separate the runner from the cavity. Perform a distance field transformation on the separated runner volume mesh, calculate the shortest distance from each point in the mesh to the runner boundary, and form a distance field scalar field D(x, y, z). The distance field calculation uses the fast marching method, advancing layer by layer from the boundary inward. The calculation formula is: Based on the distance field, the local maximum detection algorithm is applied to identify the center area of the flow channel. The judgment criteria are: when point P satisfies and When , point P is the local maximum point. All local maximum points are connected to form the initial skeleton line, which is smoothed by B-spline curve fitting. The control point spacing is 2mm, and the node weight distribution ratio is 1:4:6:4:1 to ensure that the skeleton line is smooth and continuous. The main channel and branch channel are processed separately to obtain the center line of each channel segment. For the gate area, due to the complex geometric shape, a method combining morphological operations and refinement algorithms is used to extract the center path. The corrosion operator diameter is set to 0.5mm and the number of iterations is 5. Finally, the complete channel centerline data is obtained, which includes a discrete point set of the centerline, the spatial coordinates of each point, the corresponding channel diameter and the channel segment number to which it belongs. Four types of key nodes are defined: entry node (the connection point between the channel system and the outside), bifurcation node (the location where a channel is divided into multiple channels), confluence node (the location where multiple channels merge into one channel) and gate node (the connection between the channel and the cavity). Entry node identification uses a boundary detection method to extract the point where the flow channel centerline intersects the mold shell, calculate the intersection angle, and determine it as an entry node when the angle with the normal is less than 15 degrees. Bifurcation and confluence point identification uses a connectivity analysis method to construct a K-nearest neighbor graph (K=5) for the centerline data, calculate the degree of each node (the number of edges connected to it), and when the node degree is greater than 2, further analyze the direction vectors of the connecting edges and calculate the angle θ between each vector. If there are two angles θ>135 degrees, it is determined to be a continuation of the flow direction; if all angles θ<135 degrees and there are at least 3 non-coplanar connecting edges, it is determined to be a bifurcation point or a confluence point. The distinction between bifurcation points and confluence points is based on the direction of fluid flow and is determined by calculating the cross-sectional area change rate ΔA / A of each connecting edge: if the small cross-sectional edge points to the large cross-sectional edge, it is determined to be a confluence point; otherwise, it is a bifurcation point. The gate center point identification combines the mold structure data with the runner end position to calculate the shortest distance from the runner end to the cavity surface. If the distance is less than 0.5mm and the runner diameter mutation area (diameter change rate is greater than 50% / mm), it is determined to be a gate node. Spatial clustering is performed on all marked points, and the clustering radius is set to 1.5mm to ensure that marks in similar positions are merged. Finally, a set of runner topological nodes is obtained, each node contains a unique identifier, three-dimensional coordinates, node type and associated runner segment identifier. Connection relationship analysis is achieved by tracing the runner centerline path, using an improved depth-first search algorithm. First, the runner centerline is discretized into a set of equally spaced points (spacing 0.5mm), each point contains three-dimensional coordinates and the runner segment ID to which it belongs. Starting from each entry node, all points are traversed sequentially along the runner centerline until the next topological node (bifurcation point, confluence point or gate point) is encountered. The cumulative path length L is calculated during the traversal process, and the arc length parameterization formula is used: L = ∑|P i+1 ―P i |, where P iis a point on the center line. At the same time, the average flow channel diameter d_avg along the path is recorded, and the calculation formula is: d_avg=∑(d i / n), d i is the channel diameter at the i-th point on the path, and n is the number of points. For path segments with high curvature (curvature > 0.5 / mm), tracking accuracy is improved by adjusting the step size (reduced to 0.2 mm). When a bifurcation is encountered during the traversal, the current node is recorded, and the other exit paths from the bifurcation are added to the search queue, and the depth-first search continues. Special cases are handled during the traversal: when multiple parallel channels exist between two nodes, their respective paths are calculated and marked as parallel; when a circular channel is detected, an additional marker is added to the topology to indicate a circular dependency. For each pair of connected topological nodes found, a directed connection is established, recording the starting node ID, ending node ID, connection path length, average channel diameter, path type (main channel / branch channel), and flow direction angle (relative to the horizontal plane). After the entire search is completed, a complete set of channel node connections is obtained, containing the physical connection information between all node pairs. The adjacency list is implemented as a two-layer nested array, with the outer array index corresponding to the node ID and the inner array storing information about all nodes directly connected to that node. The adjacency list construction is divided into three stages: the first stage is to initialize the adjacency list structure and create an array of size N (N is the total number of topological nodes), with each array element initialized to an empty linked list. The second stage is to traverse each record in the node connection relationship set, extract the starting node ID (i) and the ending node ID (j), add a reference to node j at the i-th position in the adjacency list, and record the connection attribute (edge weight). The edge weight calculation formula is: w(i, j) = L(i, j) / (d_avg(i, j)) 2, where L(i, j) is the path length from node i to j, and d_avg(i, j) is the average flow channel diameter. This formula reflects the physical property that fluid flow resistance is proportional to the path length and inversely proportional to the square of the diameter. In the third stage, the adjacency list is optimized and verified. Connectivity is checked to ensure there are no isolated nodes. The in-degree and out-degree of each node are calculated. The in-degree is the number of edges pointing to the node, and the out-degree is the number of edges emanating from the node. For entry nodes, the in-degree should be 0; for gate nodes, the out-degree should be 0; for bifurcations, the out-degree should be greater than the in-degree; for junctions, the in-degree should be greater than the out-degree. If topological inconsistencies are detected, dummy edges are inserted for missing connections; for redundant connections, the edge with the smallest weight is retained and the remaining edges are deleted. Finally, the initial flow channel topology data is generated. The data structure consists of a node table (node ID, type, coordinates) and an adjacency table (source node ID, target node ID list, and edge attribute list). The attribute assignment process is divided into two parts: node attribute assignment and edge attribute assignment. Node attribute assignment includes: node fluid mechanics parameters, geometric characteristic parameters and functional parameters. Node fluid mechanics parameters include pressure coefficient Cp (range 0-1.0, entry point is 1.0, end point is 0), flow distribution coefficient Cf (the ratio of each outlet flow at the bifurcation point, determined by the cross-sectional area ratio, Cf = A i / ∑A i , A i Represents the cross-sectional area of a specific outlet branch i at the bifurcation point), fluid residence time Tr (residence time of the fluid at the node, in seconds). Geometric feature parameters include node radius R (radius of the flow channel at the node, in mm), node curvature K (curvature of the centerline of the flow channel at the node, in 1 / mm), and node position coordinates X, Y, and Z (absolute position in the global coordinate system, in mm). Functional parameters include node type Type (entrance = 1, bifurcation = 2, confluence = 3, gate = 4), node status Status (normal = 1, warning = 2, abnormal = 3), and node priority Priority (1-10, the larger the value, the higher the priority). Edge attribute assignment includes: geometric attributes, fluid mechanics attributes, and process attributes. After the assignment is completed, data validation is performed to check the rationality of the attribute value range and physical consistency to ensure that the basic laws of thermodynamics and fluid mechanics are met. The final generated cavity flow channel network topology contains a complete node table, edge table, and attribute mapping table.

[0059] Preferably, the candidate foaming injection point locations are extracted based on the flow path data as follows:

[0060] Calculate the shortest distance from each area of the cavity to the boundary based on the cavity mesh data and generate distance distribution data;

[0061] Perform distance extreme value analysis on the distance distribution data, take the maximum distance point as the flow confluence area, and the minimum distance point as the boundary approach area, and generate flow feature point data;

[0062] In the flow feature point data, points with a wall thickness greater than 1.5 mm and a flow confluence area exceeding 30% of the maximum cavity size are selected as initial candidate injection points.

[0063] Calculate the average flow distance from each point in the initial candidate injection point to all areas of the cavity, and select the five points with the smallest average flow distance as the candidate injection point position sequence;

[0064] Using the flow path data, the flow balance of different foaming injection point locations in the candidate injection point location sequence is evaluated, and the flow length uniformity index is calculated to obtain injection point evaluation data. The index is defined as the ratio of the longest flow path to the shortest flow path.

[0065] According to the injection point evaluation data, the position where the flow length uniformity index is closest to 1 is selected as the candidate foaming injection point position.

[0066] In the embodiment of the present invention, the cavity volume grid data is converted into a three-dimensional grid matrix, and the grid resolution is set to 0.2mm to ensure that subtle geometric features are captured. The grid matrix is binarized, and the voxels inside the cavity are marked as 1 and the voxels on the boundary are marked as 0. A three-dimensional distance transformation algorithm is used to calculate the Euclidean distance from each internal voxel (i, j, k) to the nearest boundary voxel (i′, j′, k′). The calculation process is divided into three stages: In the first stage, the one-dimensional distance is calculated along the X-axis direction, and the formula is Dx(i, j, k) = min{(ii′) 2 |M(i′, j, k)=0}; In the second stage, the two-dimensional distance is calculated based on the Y axis, and the formula is Dxy(i, j, k)=min{Dx(i, j′, k)+(jj′) 2}; In the third stage, the Z axis is integrated to calculate the three-dimensional distance. The formula is In order to improve the computational efficiency, the scanning line technology is combined with the distance propagation algorithm, and the time complexity is controlled at O(n), where n is the total number of voxels. For complex-shaped cavities, the Manhattan distance is applied as the initial estimate in the concave area, and then iterative optimization is converted to the Euclidean distance. The iterative termination condition is that the difference between two adjacent calculation results is less than 0.01mm. The calculation results form the cavity distance field, which contains the three-dimensional coordinates of each voxel point and the shortest distance value corresponding to the boundary. The data is saved as a three-dimensional matrix of floating-point numbers with an accuracy of 0.001mm to obtain complete distance distribution data. The extreme points of the distance field are located by combining three-dimensional gradient analysis with the Laplace operator. First, the distance field gradient vector field is calculated. For each point P(i, j, k), its gradient vector is Calculated using the central difference formula: Then the Laplacian of the distance field is calculated, The second-order derivative uses the three-point formula: The criteria for determining the maximum point are: and Where ε is set to 0.05, indicating that the gradient is close to zero and the Laplace is negative; the minimum point judgment criteria are: and Indicates that the gradient is close to zero and the Laplace is positive. In order to overcome noise interference, Gaussian smoothing preprocessing is applied, the kernel size is set to 3×3×3, and the standard deviation σ=0.8. The detected extreme points are spatially clustered using the density clustering algorithm DBSCAN, and the parameters are set as: neighborhood radius ε=1.5mm, minimum number of points MinPts=5. After clustering, each cluster center is extracted as a representative point, the maximum point cluster is marked as the flow confluence area, and the minimum point cluster is marked as the boundary proximity area. For each feature point P in the flow confluence area, the local wall thickness is calculated using the bidirectional ray casting method: a ray is emitted from point P along multiple directions (26 evenly distributed directions) until it hits the cavity boundary, and the distances d1, d2, ..., d in each direction are recorded. 26 , the wall thickness is defined as the sum of the shortest distances in two relative directions: thickness(P)=min{d i +d j}, where d i represents the distance from the characteristic point P in the flow confluence area to the cavity boundary when the light is emitted in the i-th direction, and d j It represents the distance that the light from the same point P in the direction j opposite to i hits the boundary. Calculate the maximum size of the cavity Dmax, specifically the diagonal length of the rectangular frame surrounding the cavity: Among them, Lx, Ly, and Lz are the dimensions of the cavity in the three main axis directions. The coverage range of each flow confluence area is calculated, and the flow confluence area size S is defined as: the maximum distance from all points P in the area to the area boundary, that is, S = max{D(P)|P∈confluence area}. The size ratio R = S / Dmax is calculated. When R>0.3 (that is, the confluence area exceeds 30% of the maximum size of the cavity), the area meets the screening conditions. The final screening criteria are: wall thickness thickness (P)>1.5mm and size ratio R>0.3. For each flow confluence area that meets the conditions, the center point of the area (the point with the smallest average distance to all points in the area) is selected as the representative point. The cavity is divided into discrete areas of equal volume. The octree space segmentation algorithm is used, and the leaf node volume threshold is set to 1% of the total volume to form about 100 area units. For each initial candidate injection point Pi, the flow distance from it to the center Cj of all regional cells in the cavity is calculated. Using the improved Dijkstra shortest path algorithm, a weighted graph G(V, E) is constructed on the grid. Vertices V are grid nodes, and edges E are connections between adjacent grids. The edge weight w(e) is defined as: w(e) = L(e)·(1+5·exp(-D(e) / 0.5)), where L(e) is the edge length and D(e) is the average distance field value on the edge. This weight function causes the flow to tend to propagate along a wider area. The shortest path length from the injection point Pi to each regional center Cj is recorded as dist(Pi, Cj). The average flow distance is calculated as: Davg(Pi) = ∑dist(Pi, Cj) / N, where N is the total number of regional cells. To evaluate flow uniformity, the standard deviation of the flow distance is calculated: The smaller the standard deviation, the more uniform the flow. Combining the average distance and the standard deviation, calculate the comprehensive score: Score(Pi) = Davg(Pi)·(1+2·σ(Pi) / Davg(Pi)). This formula takes into account both the length and uniformity of the flow path. Sort by score from small to large, extract the 5 points with the lowest score to form a sequence of candidate injection point positions. Each point records its coordinates, average flow distance, standard deviation, comprehensive score, and maximum flow distance to the farthest area. For each candidate point P, extract the flow path set PathSet(P) from P to all directions of the cavity surface. Each path Path_i is represented as a series of discrete point sequences {P, P1, P2, ..., P n}, where P n Located on the cavity surface. Calculate the flow length of each path FL(Path_i) = ∑|P j+1 -P j |,P j It refers to the jth point in this sequence. Considering the change of wall thickness along the path, the flow resistance factor R(x) is introduced. x represents the midpoint position of two adjacent points, that is, (P j +Pj+1 ) / 2, the modified flow length is: MFL(Path_i)=∑|P j+1 -P j |·R((P j +P j+1 ) / 2). Perform statistical analysis on the corrected flow lengths of all paths, and record the longest path length Lmax(P)=max{MFL(Path_i)}, the shortest path length Lmin(P)=min{MFL(Path_i)}, and the average path length Lavg(P)=∑MFL(Path_i) / N, where N is the total number of paths. The flow length uniformity index is defined as the ratio of the longest flow path to the shortest flow path: FUI(P)=Lmax(P) / Lmin(P). The ideal value of this index is 1, indicating that the flow in all directions is completely balanced. In addition, calculate the coefficient of variation of the flow path length: CV(P)=σ(MFL) / Lavg(P). The smaller the coefficient of variation, the better the flow uniformity. Analyze the flow resistance distribution for each path, and calculate the flow resistance concentration index RCI(P)=∑(R max , Path_i) / N, where (R max , Path_i) represents the maximum value of the flow resistance factor R(x) of all position points x on Path_i, that is, R_max, Path_i=max{R(x)|x∈Path_i}, N is the total number of paths, and the lower the resistance concentration, the smoother the flow. Combining the above indicators, the injection point evaluation data is generated, including the three-dimensional coordinates of each candidate point, the flow length uniformity index FUI, the coefficient of variation CV, the resistance concentration index RCI, and the number of flow paths. The weight coefficients of the evaluation indicators are determined. The flow length uniformity index FUI is set to 0.6, the coefficient of variation CV is set to 0.3, and the resistance concentration index RCI is set to 0.1. The weight distribution is based on the high uniformity requirements of the foaming process. The flow length uniformity index FUI is transformed to calculate the proximity index: PI(P)=1 / |FUI(P)-1|. When FUI is close to 1, the larger the PI value, the more balanced the flow. The upper limit of PI is set to 100 to prevent extremely small differences from leading to excessively high scores. Construct a comprehensive scoring function: TotalScore(P)=0.6·PI(P)+0.3·(1-CV(P))+0.1·(1-RCI(P) / RCI max ), where RCI maxis the maximum resistance concentration value among all candidate points. In order to consider the actual processing feasibility, the position constraint factor LC(P) is introduced. When point P is near the mold parting surface (distance <5mm), LC(P) = 0.8, otherwise LC(P) = 1.0, and the final score is: FinalScore(P) = TotalScore(P) · LC(P). The point with the highest score is selected from the candidate injection point position sequence as the final candidate foaming injection point position. A final verification is performed on the selected point. By generating 10 test points within a radius of 2mm around the point, the score of each test point is calculated to ensure that the selected point is not a local optimal solution. The final candidate foaming injection point position records its precise three-dimensional coordinates, scoring indicators, flow characteristic parameters, and recommended injection pressure and temperature parameters.

[0067] It is particularly important to use the flow path data to evaluate the flow balance of different foaming injection point positions for the candidate injection point position sequence. Specifically:

[0068] Perform spatial mapping processing on the flow path data and the candidate injection point position sequence to generate injection point flow correlation data;

[0069] Based on the injection point flow correlation data, the path length of the melt flowing from the injection point to each area of the cavity is calculated for each candidate injection point position, and the injection point flow distance distribution data is generated;

[0070] The maximum flow distance, minimum flow distance and average flow distance of each candidate injection point are calculated based on the injection point flow distance distribution data to obtain flow distance statistics;

[0071] Calculate the flow length uniformity index for each candidate injection point based on flow distance statistics;

[0072] The flow fluctuation evaluation is performed on the flow uniformity evaluation data, and the flow accessibility index of each injection point position to each area of the cavity is calculated to generate the injection point evaluation data.

[0073] In an embodiment of the present invention, an undirected weighted graph with grid nodes as vertices is constructed, and the edge weights are set to the actual geometric distances between adjacent nodes. For dense areas where the grid unit size is less than 0.8 mm, a weight coefficient of 1.2 is introduced to compensate for the calculation error. Starting from the starting injection point, the Dijkstra algorithm is used to calculate the shortest path length to all grid nodes. Taking into account the flow characteristics of the plastic melt, an additional weight factor of 1.5 is introduced at obstacles or sharp turns to reflect the increase in actual flow resistance. After the calculation is completed, the injection point flow distance matrix is formed, and the matrix element (i, j) represents the shortest flow path length from the i-th injection point to the j-th grid node, generating the injection point flow distance distribution data. By traversing all the grid nodes associated with each injection point, the longest flow path Lmax and the shortest flow path Lmin are found. To calculate the average flow path length Lavg, the formula Lavg=∑L i / n, where n is the total number of associated grid nodes, L i is the flow path length of the i-th node. The standard deviation σ of the flow path length is calculated to reflect the flow uniformity. Abnormal points with path lengths exceeding Lavg+2σ are marked and their influence is corrected and analyzed. According to the Lmax, Lmin, Lavg and σ values of each injection point. The flow length uniformity index is calculated using the formula FUI=Lmin / Lmax. The closer the value is to 1, the more uniform the flow. In order to eliminate the interference of local extreme values on the overall evaluation, 5% of the extreme values of the longest and shortest path lengths are removed before calculating the corrected uniformity index FUI'. At the same time, the standardized flow distance variance SVF=σ is calculated. 2 / (Lavg) 2 , as an auxiliary evaluation indicator. The FUI value is graded into intervals: 0.9-1.0 is excellent, 0.8-0.9 is good, 0.7-0.8 is average, and below 0.7 is poor. The flow fluctuation coefficient WF = (Lmax-Lmin) / Lavg is calculated for each area. The WF threshold for the core area is set to 0.3, the transition area is 0.4, and the edge area is 0.5. The flow accessibility index calculation takes into account the regional importance weight, using the formula AI = ∑(w i ·L i ) / ∑w i Indicates that w i is the regional importance weight, with the core area weight being 0.5, the transition area weight being 0.3, and the edge area weight being 0.2. The flow uniformity index, flow fluctuation coefficient, and flow accessibility index are weighted and fused, with weighting coefficients of 0.4, 0.3, and 0.3, respectively, to obtain the comprehensive score of the injection point.

[0074] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S2 are shown in the flowchart. In this example, step S2 includes:

[0075] Step S21: Plastic mold processing parameters include mold material parameters, plastic material parameters, and supercritical fluid parameters;

[0076] In an embodiment of the present invention, the plastic mold processing parameters are systematically defined. The mold material parameters include physical performance data such as mold steel type (such as P20, H13, etc.), thermal conductivity, specific heat capacity, elastic modulus and thermal expansion coefficient. These parameters are obtained by querying the standard material manual and combining with the actual mold material experimental measurement. The plastic material parameters cover the melting point, density, viscosity temperature change curve and crystallization rate of the matrix resin. The data comes from the technical data sheet provided by the material supplier and the differential scanning calorimeter test results. Supercritical fluid parameters include the critical temperature, critical pressure, density and diffusion coefficient of supercritical carbon dioxide, which are determined by a high-precision fluid parameter database and relevant literature.

[0077] Step S22: performing injection process parameter processing on the candidate foaming injection point position based on the supercritical fluid parameters to obtain injection process parameters;

[0078] In the embodiment of the present invention, a local temperature gradient model is established for the candidate injection point position, and the temperature field distribution T(r)=T0-ΔT·(r / r0) within a range of 10mm around the injection point is calculated. 2 , where T0 is the injection point temperature, ΔT is the temperature gradient, r is the radial distance, and r0 is the reference distance of 10 mm. To calculate the injection pressure, consider the solubility of the supercritical fluid at temperature T and pressure P S(T, P) = S0 exp(αP-βT), where S0 is the solubility under standard conditions, α is the pressure coefficient (0.15 MPa for CO2 -1 , N2 is 0.08MPa -1 ), β is the temperature coefficient (CO2 is 0.007℃ -1 , N2 is 0.005℃ -1 ). The injection pressure range is determined as P_inj=P_crit+ΔP, where P_crit is the critical pressure of the supercritical fluid and ΔP is the additional pressure (10-30MPa). The injection speed V_inj is calculated based on the cavity volume V_cavity and the filling time t_fill: V_inj=V_cavity / (A_gate·t_fill), where A_gate is the injection port area and t_fill is determined based on the thermal sensitivity and viscosity of the material (typical value 1-5s). The holding parameters include holding pressure P_hold=0.8·P_inj, holding time t_hold=k·h 2 , where k is the empirical coefficient (2-3s / mm 2), h is the mold wall thickness. The cooling time t_cool is calculated according to the heat diffusion equation: t_cool=h 2 ·ln(4 / π·(T_melt-T_mold) / (T_eject-T_mold)) / (π 2 ·α), where α is the thermal diffusion coefficient, T_melt is the melt temperature, T_mold is the mold temperature, and T_eject is the ejection temperature. Foaming control parameters include supercritical fluid injection time t_SCF and injection volume m_SCF = ρ_melt·V_cavity·w%, where ρ_melt is the melt density and w% is the weight percentage (0.5-2.0%). Parameter optimization uses the response surface methodology to construct a functional relationship between P_inj, V_inj, and T_melt and foam cell density and dimensional uniformity. The final injection process parameter combination is obtained through iterative calculation (convergence threshold 0.1%).

[0079] Step S23: importing mold material parameters, plastic material parameters, supercritical fluid parameters, and injection process parameters into Moldflow software based on the mold foaming simulation configuration data to construct an initial simulation model of the plastic mold;

[0080] In the embodiment of the present invention, a complete model boundary condition is created: the cavity mesh data is used as the basic geometric boundary; the connection relationship between the runner system and the cavity is defined, the interface between the main channel and the branch channel is treated as a continuous boundary, the unit mesh at the interface is kept consistent, and the size ratio of adjacent units is controlled within 1.2:1; the fluid inlet boundary condition at the gate position is set, the injection pressure is 18-25MPa, or the injection speed is 15-40cm 3 / s; the cavity wall boundary condition is set to a no-slip boundary, and the wall temperature is set to the mold temperature of 35-80°C. Material property data input includes: importing the thermophysical parameters of the mold material, establishing the temperature-dependent thermal conductivity function λ(T) = λ0(1+w t T), where λ0 is the room temperature thermal conductivity, w t is the temperature coefficient (typical value 3×10 -4 K -1 ); Input the rheological data of plastic materials, including the non-Newtonian fluid model η(SR, T) = η0·a T ·(1+(η0·a T SR / τ*)^(1-n))^((n-1) / n), where η0 is the zero shear viscosity, a Tis the temperature translation factor, SR is the shear rate, τ* is the critical stress, and n is the power law exponent; the PVT data of plastic materials are established using the modified Tait equation v(T, P) = v0(T) [1-C ln(1+P / B(T))] + v_t(T, P), where v0(T) is the zero-pressure specific volume, B(T) is the temperature-related pressure sensitivity, C is a constant of 0.0894, and v_t is the phase change correction term. The supercritical fluid parameter settings include: the pressure-volume-temperature relationship of the fluid type (CO2 or N2), the solubility-temperature-pressure relationship curve, and the nucleation and growth kinetic model parameters. In the simulation control parameter settings, the time step is 1 / 100 of the filling time, and the convergence criterion is set to a residual less than 10 -4 , the maximum number of iterations is 20 times / time step, and the SIMPLE algorithm is used to handle pressure-velocity coupling.

[0081] Step S24: using the initial simulation model of the plastic mold to perform plastic melt flow-foaming injection simulation, and periodically collecting temperature field data, pressure field data, flow velocity field data, and shear rate field data in the cavity;

[0082] In the embodiment of the present invention, the simulation process is divided into four stages: filling stage, pressure holding stage, cooling stage and foaming stage. 40 evenly distributed cavity monitoring points are set for data collection, and data is collected every 0.05s. A total of 30-60 sets of data are collected during the filling stage. The temperature field data collection range is 190-280℃, including the instantaneous temperature value and its gradient of each cavity monitoring point; the pressure field data collection range is 0-30MPa, including the pressure distribution and pressure drop rate dp / dx in the cavity; the velocity field data collection range is 0-150cm / s, including the three-dimensional velocity components vx, vy, vz and velocity gradient; the shear rate field data collection range is 0-10000s -1 , including surface shear rate and volume average shear rate. A larger time step of 0.05s is used in the pressure holding stage, and the data acquisition frequency is reduced to once every 0.2s, focusing on monitoring changes in the pressure field and temperature field. An adaptive time step is used in the cooling stage, with a step range of 0.1-1.0s and an acquisition frequency of once every 0.5s, mainly monitoring the temperature field attenuation process. The foaming stage is the key monitoring stage, with a refined time step of 0.005s, and the acquisition frequency is increased to once every 0.02s. The supercritical fluid concentration field c (x, y, z, t) is added for monitoring, with a concentration range of 0-2wt%, and the local supersaturation Δc = c-c_eq (T, P) is calculated, where c_eq is the equilibrium solubility. All collected data are stored in a structured format, containing spatial coordinates, timestamps, and physical quantities to form a complete four-dimensional field data set. in Indicates the physical quantity being monitored.

[0083] Step S25: Injection filling state integration is performed based on the temperature field data, the pressure field data, the flow velocity field data, and the shear rate field data to obtain injection flow field state data.

[0084] In one embodiment of the present invention, field data from different acquisition frequencies are unified onto grid nodes via cubic spline interpolation, forming a regular data set with a time interval of 0.01s. The temperature field T(x, y, z, t) is dimensionlessly processed to calculate the temperature coefficient θT = (T-Tmold) / (Tmelt-Tmold), where Tmold is the mold temperature and Tmelt is the material melting temperature. The pressure field P(x, y, z, t) is normalized to calculate the pressure coefficient θP = P / Pmax, where Pmax is the maximum injection pressure. The flow number Fr = |v|2 / (g·L) is calculated for the velocity field v(x, y, z, t), where g is the acceleration of gravity and L is the characteristic length (the maximum cavity dimension). The Weisenberg number We = λ·SR is calculated for the shear rate field SR(x, y, z, t), where λ is the characteristic relaxation time of the material (typically 0.01-0.1s) and SR is the shear rate. Further calculation of composite indicators: flow front position function F(x, y, z, t), when the unit filling rate is greater than 0.5, F = 1, otherwise F = 0; temperature-pressure coupling factor Characterizes the consistency of temperature gradient and pressure gradient; viscoelastic coefficient De = τ / t*, where τ is the stress relaxation time and t* is the characteristic flow time; flow uniformity index UI = 1-σv / v - , where σv is the standard deviation of flow velocity, v - is the average flow velocity. Principal component analysis was used for dimensionality reduction, extracting the first three principal components (PC1, PC2, and PC3), with a cumulative variance contribution rate exceeding 85%. A flow field state classifier was constructed using the PC1-PC2-PC3 three-dimensional space. Ultimately, injection flow field state data was generated, including the filling ratio (0-100%, step size 1%), cavity pressure distribution, flow front position, temperature distribution, fluid flow vector field, shear stress distribution, and predicted bubble formation area.

[0085] Preferably, the injection flow field abnormal area includes a temperature abnormal area, a pressure abnormal area, a velocity abnormal area, and a shear abnormal area. The analysis of the injection flow field abnormal area according to the injection flow field state data in step S3 includes:

[0086] Based on the temperature field data in the injection flow field state data, the average temperature and temperature standard deviation of each area in the cavity are calculated to generate temperature distribution statistics;

[0087] Based on the temperature distribution statistics, identify areas where the temperature is 15°C below the glass transition temperature of the plastic material, or where the temperature standard deviation is greater than 5°C, and mark them as temperature anomaly areas;

[0088] Based on the pressure field data in the injection flow field state data, the average pressure and pressure gradient of each area in the cavity are calculated to generate pressure distribution statistics;

[0089] Based on the pressure distribution statistics, the areas where the pressure is lower than 80% of the set holding pressure, or the areas where the pressure gradient is greater than 12 MPa / mm are identified as abnormal pressure areas;

[0090] The velocity vector field in the cavity is analyzed based on the velocity field data in the injection flow field state data. When the velocity direction deviates from the main flow direction by more than 45 degrees and the velocity is lower than 0.5 cm / s, the area is marked as an abnormal velocity area.

[0091] The shear rate abnormality threshold is set to 1500s based on the shear rate field data in the injection flow field state data. -1 ,When the shear rate exceeds the threshold and the duration exceeds 1.5s, the area is marked as a shear abnormality area.

[0092] In an embodiment of the present invention, the temperature field data calculation process adopts a grid partitioning statistical method to divide the cavity space into 10×10×10 cubic cells of equal volume, and the side length of each cell is 1 / 10 of the maximum size of the cavity. Read the temperature value T(i, j, k) of each grid node in the temperature field data matrix, and perform statistics on the temperature value in each cubic cell. Calculate the average temperature of each cell Tavg = ∑T(i, j, k) / n, where n is the number of grid nodes in the cell, usually around 125. Calculate the temperature standard deviation at the same time Reflects the degree of temperature fluctuation. Nodes at the cell boundaries are processed using a weighted average method, with the weight coefficient set inversely proportional to the distance from the node to the cell center. The generated temperature distribution statistics include the center coordinates, average temperature, temperature standard deviation, maximum and minimum temperature values of each cell, forming a temperature distribution matrix of 1000 data points. Identifying abnormal temperature areas is based on two core indicators: absolute temperature value and the degree of temperature fluctuation. First, the glass transition temperature (Tg) of the plastic material used is obtained. For the commonly used ABS material, this value is set to 105°C. The lower temperature threshold (Tthreshold) is calculated as Tg - 15°C, which is 90°C. Each cell in the temperature distribution statistics is iterated over. If the cell average temperature (Tavg) is less than 90°C, the cell is marked as a "low temperature abnormal area." The temperature standard deviation (σT) is also checked. If σT is greater than 5°C, the cell is marked as a "temperature fluctuation abnormal area." Cells that meet both abnormality criteria are marked as "severe temperature abnormal areas." Adopting the region growing algorithm, adjacent abnormal cells are merged into a continuous region. When the volume of the continuous region exceeds 2% of the total volume of the cavity, it is recorded as a key monitoring temperature abnormality region, and the abnormality severity index SSI = (90-Tavg) × (1 + σT / 5) is calculated. The larger the value, the more serious the abnormality. Read the pressure field matrix P (i, j, k) in the injection flow field state data. Divide the cavity space into 6 × 6 × 6 cubic pressure analysis units of equal volume, and the side length of each unit is approximately 1 / 6 of the maximum size of the cavity. Calculate the average pressure Pavg = ∑P (i, j, k) / m in each analysis unit, where m is the number of pressure sampling points in the unit. For pressure gradient calculation, the central difference method is used, and the calculation formula is: Where Δx, Δy, and Δz are the grid spacings. Calculate the pressure gradient modulus The unit is MPa / mm. For the area near the cavity boundary, a one-sided difference formula is used to avoid boundary processing errors. The generated pressure distribution statistics include the average pressure, pressure gradient vector and gradient modulus of 216 analysis units. Get the mold setting holding pressure Phold, the typical value is 60MPa. Calculate the pressure lower limit threshold Plow = 0.8 × Phold = 48MPa, and mark the analysis unit with an average pressure Pavg < 48MPa in the pressure distribution statistics as a "low pressure abnormal area". At the same time, set the pressure gradient threshold The pressure gradient modulus The analysis unit marked as "abnormal pressure gradient area" was marked as abnormal. To improve the accuracy of abnormal area identification, the analysis unit marked as abnormal was internally subdivided into 8 subunits (2×2×2). Pressure analysis was repeated on the subunits to eliminate false positives. Adjacent abnormal units were merged using a three-dimensional agglomerative hierarchical clustering algorithm to form a continuous abnormal area. The pressure anomaly severity index of each abnormal area was calculated. Abnormal regions are sorted by PSI value to generate a spatial distribution map of abnormal pressure regions. The velocity vector field V(i, j, k) is extracted from the injection flow field state data. Each vector contains three components (Vx, Vy, Vz). The ideal mainstream direction within the cavity is determined: for the area near the injection point, the mainstream direction is defined as the radial direction radiating outward from the injection point; for the flow channel area, the mainstream direction is defined as the tangent direction of the flow channel centerline; and for the general cavity area, the mainstream direction is defined as the direction opposite to the normal of the nearest wall to the point. The angle θ = arccos[(V·M) / (|V|·|M|)] between the actual flow velocity direction and the ideal mainstream direction at each sampling point is calculated, where M is the unit vector of the mainstream direction. A direction deviation threshold θthreshold = 45 degrees and a lower velocity threshold Vthreshold = 0.5 cm / s are set. When θ > 45 degrees and |V| < 0.5 cm / s, the point is marked as a velocity anomaly. To reduce the influence of discrete noise, kernel density estimation is applied to velocity anomalies. When the density of anomalies within a radius of 2 mm exceeds 60%, it is confirmed as a velocity anomaly area. Shear rate field data SR(i, j, k, t) is extracted from the injection flow field state data, where t represents the time dimension. The shear rate threshold SR_threshold is set to 1500s. -1 , time threshold t_threshold = 1.5s. For each spatial point, calculate the duration of the shear rate exceeding the threshold Δt = ∑δt·I[SR(i, j, k, t)>1500s -1 ], where δt is the time step, I[·] is the indicator function, which is 1 when the condition is met and 0 otherwise. When Δt>1.5s at a point, the point is marked as a shear anomaly. The spatial clustering algorithm DBSCAN is applied to all shear anomaly points, with the minimum distance parameter ε=1.2mm and the minimum number of points parameter MinPts=8. The clustering results are used as shear anomaly areas. The average shear rate SR_avg, maximum shear rate SR_max and high shear duration SR_tavg of each shear anomaly area are calculated, and the shear anomaly severity index SCI=(SR_avg / 1500)×(SR_tavg / 1.5) is constructed. The anomaly areas are sorted by SCI value to generate shear anomaly area marking data.

[0093] It is particularly important that the velocity vector field analysis in the cavity is as follows:

[0094] The velocity field data in the injection flow field state data is spatially discretized, and vector sampling points are established based on the cavity grid nodes. The sampling density is set to 1 / 20 of the cavity characteristic size to obtain the velocity vector distribution data.

[0095] Based on the velocity vector distribution data, the angle between the velocity vector at each sampling point and the main flow direction of the cavity is calculated. The main flow direction is defined as the straight line from the injection point to the cavity boundary. When the fluid flows along the cavity wall, the main flow direction is defined as the tangent direction of the wall. The flow deviation angle data is generated.

[0096] The flow deflection intensity index is calculated based on the flow deviation angle data. The index is defined as the weighted product of the flow velocity deflection angle and the flow velocity, with the weight coefficients set to 0.7 and 0.3.

[0097] In an embodiment of the present invention, the three-dimensional space of the cavity is divided into a regular grid, and the side length of each grid unit is set to 1 / 20 of the maximum characteristic size of the cavity, ensuring that the grid density is sufficient to capture the flow details but does not cause excessive computational burden. For curved or variable cross-section areas, an adaptive grid encryption technique is used to reduce the side length of the grid unit to 1 / 30 to improve local accuracy. The velocity field data is mapped to each grid node through an interpolation algorithm, and each node stores the three-dimensional velocity vector (Vx, Vy, Vz) and the velocity magnitude |V|. In addition, for the area near the wall, a boundary layer grid is set, the first layer grid height is 0.2mm, and a total of 5 layers are set to ensure that the wall flow characteristics are accurately captured. For nodes inside the cavity, the main flow direction is defined as the direction vector of the line from the current node to the nearest injection point; for nodes less than 1.5mm from the wall, the main flow direction is defined as the tangent direction vector of the wall where the current node is located, and the tangent direction is calculated by the vertical component of the normal vector of the wall grid unit. Calculate the angle θ between the velocity vector V of each node and the main flow direction vector M using the formula: θ = arccos[(V·M) / (|V|·|M|)]×180 / π, in degrees. When a numerical singular point appears in the calculation result, the average value of the 8 surrounding neighboring nodes is used instead. Establish an angle distribution matrix for all calculation results. The flow deflection intensity index = 0.7×(θ / 90)+0.3×(1-|V| / Vmax), where θ is the flow deviation angle, |V| is the flow velocity of the current node, and Vmax is the global maximum flow velocity. The index ranges from 0 to 1, and the larger the value, the more severe the flow deflection. The flow deflection intensity threshold is set to 0.65, and areas exceeding this threshold are marked as high-risk flow anomaly areas. Statistical analysis of the flow deflection intensity index at all nodes was performed to generate distribution maps and heat maps. The heat maps use red to indicate high-risk areas (flow deflection intensity index > 0.65), yellow to indicate medium-risk areas (0.45-0.65), and green to indicate normal areas (< 0.45). Regional cluster analysis was used to group adjacent high-risk points into blocks of abnormal flow areas.

[0098] Preferably, in step S3, identifying the abnormal foaming area based on the injection flow field state data, and performing abnormal space superposition analysis based on the abnormal injection flow field area includes:

[0099] The temperature field data and pressure field data in the injection flow field state data are used to calculate the bubble nucleation rate in each area, thereby generating the bubble nucleation area distribution data;

[0100] Calculate the initial density and initial size of bubbles based on the bubble nucleation area distribution data;

[0101] Analyze the simulation holding time parameters based on the injection flow field state data;

[0102] Based on the simulation holding time parameters, the bubble diameter increment is calculated by the bubble initial density and initial size to evaluate the plastic foaming diameter and bubble distribution density in each area to obtain the plastic simulation foaming data;

[0103] Compare the plastic simulation foaming data with the preset ideal foaming parameters. When the plastic foam diameter deviates from the preset ideal foaming parameters by more than 30% or the bubble distribution density deviates by more than 40%, it is marked as a foaming abnormal area.

[0104] The foaming abnormal area and the injection flow field abnormal area are spatially superimposed and analyzed, and the comprehensive abnormality index is calculated based on the overlap weight coefficient. When the comprehensive abnormality index exceeds 0.5, it is marked as a high-risk area, and the comprehensive flow field abnormality area data is generated; among them, the overlap weight coefficient sets the temperature abnormality weight to 0.3, the pressure abnormality weight to 0.4, the flow velocity abnormality weight to 0.2, and the shear abnormality weight to 0.1.

[0105] In the embodiment of the present invention, the calculation of the bubble nucleation rate based on the injection flow field state data requires the comprehensive temperature field T (x, y, z, t) and pressure field P (x, y, z, t) data to construct a classical nucleation theory model. The bubble nucleation rate N (T, P) is calculated using the improved Blander-Katz equation: N (T, P) = N0 · exp [-16πγ 3 f(θ) / (3k e T(ΔP) 2 )], where N0 is the pre-exponential factor (value 10 28 m -3 s -1 ), γ is the gas-liquid interfacial tension (about 0.015 N / m for the CO2-PP system at 170°C), and f(θ) is the contact angle function (f(θ) = 1 for homogeneous nucleation and f(θ) = (2 + cosθ)(1 - cosθ) for heterogeneous nucleation) 2 / 4), k e is the Boltzmann constant (1.38×10 -23J / K), ΔP is the supersaturation pressure, calculated as ΔP = Psat(T) - P, where Psat(T) is the saturation pressure of the supercritical fluid at temperature T, calculated using the improved Peng-Robinson equation of state. Adaptive grids are used to divide the cavity area, and the grid is refined in areas where the temperature gradient is greater than 10°C / mm or the pressure gradient is greater than 5MPa / mm, with a minimum unit size of 0.5mm. Nucleation rate calculations are performed for each calculation unit with a discrete time step of 0.02s. The thermodynamic compatibility coefficient C(T) is introduced to characterize the temperature-dependent compatibility of the supercritical fluid and the polymer: C(T) = C0·exp[-E_c / (RT)], where C0 is a constant, E_c is the compatibility activation energy (approximately 15kJ / mol for the CO2-PP system), R is the gas constant, and T is the temperature. The nucleation rate calculation results are corrected by the density function to form the bubble nucleation area distribution data. The initial bubble density n0(x, y, z) is solved by time integration: Where t1 is the end time of pressure holding, t2 is the time for the material to cool to the crystallization temperature or glass transition temperature, and the integral interval (t1, t2) is the nucleation window time, which is the period from the end of pressure holding to the cooling of the material to the crystallization temperature or glass transition temperature, with a typical range of 1-5s. The calculation process uses the Simpson integration method with a time step of 0.05s to ensure that the integral accuracy error is less than 3%. The initial bubble size R0(x, y, z) is calculated based on the Laplace equation: R0(x, y, z) = 2γ / ΔP(x, y, z), where γ is the gas-liquid interfacial tension and ΔP is the local supersaturation pressure. For complex geometric cavities, considering the influence of wall effects on the initial bubble size, the wall correction factor F_w = 1-exp(-d / 5R0) is introduced, where d is the distance to the nearest wall, and the corrected initial size is R'0 = R0·F_w. Considering the influence of shear stress on bubble morphology, the local Weber number We = ρv is calculated. 2R0 / γ, where ρ is the melt density and v is the local flow velocity. When We > 0.5, the initial bubble shape is an ellipsoid with a major-minor axis ratio a / b = 1 + 0.2We. For high-molecular-weight plastics (such as high-density PE), the influence of viscoelastic effects on bubble stability is considered, and the Deborah number De = λ_r / t_g is introduced, where λ_r is the material relaxation time and t_g is the characteristic time for bubble growth. When De > 1, the initial bubble density correction factor is 0.7-0.9. Critical time points are identified from the pressure field data P(x, y, z, t): the injection switching point t_switch (defined as the time when the cavity pressure reaches 90% of its maximum value), the packing start point t_pack_start (defined as the time when the pressure begins to stabilize, satisfying |dP / dt| < 0.5 MPa / s), and the packing end point t_pack_end (defined as the time when the cavity pressure begins to drop significantly, satisfying dP / dt < -2 MPa / s and lasting for more than 0.2 seconds). The actual packing time, t_pack_actual, is calculated as t_pack_end - t_pack_start. A gate solidification criterion is introduced, analyzing the temperature changes in the gate area based on the temperature field data T(x, y, z, t). When the gate temperature is lower than the material no-flow temperature, T_no_flow (typically the crystallization temperature Tc + 30°C or the glass transition temperature Tg + 60°C), it is recorded as the gate solidification time, t_gate_freeze. The effective packing time, t_pack_effective, is calculated as min(t_pack_actual, t_gate_freeze - t_pack_start). The cavity pressure transfer efficiency is analyzed, defining the pressure transfer coefficient, PTC(x, y, z) = P(x, y, z, t_mid) / P_inlet(t_mid), where t_mid is the mid-packing moment and P_inlet is the inlet pressure. The PTC distribution determines the effective packing area (PTC > 0.8) and the insufficient pressure area (PTC < 0.5). Calculate the cavity pressure uniformity time t_pressure_uniform, which is defined as the ratio of the standard deviation of the cavity pressure to the average pressure σ p / P - The earliest moment less than 0.1, where σ p Indicates the standard deviation of the pressure in the cavity, P -Represents the arithmetic mean of the pressure values at all spatial points in the cavity. Based on the cooling characteristics of the material, the critical holding time t_critical is calculated to ensure that the temperature of 90% of the cavity volume is lower than the material crystallization onset temperature Tc-onset or the glass transition temperature Tg. The bubble diameter growth rate in the diffusion-controlled stage is: dR / dt = D_eff·(c_∞-c_s) / (R·ρ_g), where D_eff is the effective diffusion coefficient of the supercritical fluid in the polymer (the temperature dependence is expressed as D_eff = D0·exp(-E_d / RT), D0 is the pre-exponential factor, and E_d is the diffusion activation energy), c_∞ is the far-field fluid concentration, c_s is the equilibrium concentration on the bubble surface (calculated according to Henry's law c_s = k_H·P_bubble, k_H is Henry's constant), and ρ_g is the gas density. The bubble diameter growth rate in the viscosity-limited stage is: dR / dt=(P_bubble-P_∞-2γ / R) / (4η), where P_bubble is the pressure inside the bubble, P_∞ is the surrounding melt pressure, and η is the melt viscosity. For common foaming plastics such as PP foaming materials, the ideal bubble diameter R_ideal is set to 50-100μm, and the ideal bubble density n_ideal is set to 10 6 -10 7 cells / cm 3 , the ideal bubble distribution uniformity CV_ideal (coefficient of variation) is set to less than 0.25. For each grid point (x, y, z), the bubble diameter deviation rate DR(x, y, z) = |R_final(x, y, z)-R_ideal| / R_ideal is calculated. When DR>0.3 (i.e. the deviation exceeds 30%), it is marked as a diameter anomaly point. The bubble density deviation rate DN(x, y, z) = |n_final(x, y, z)-n_ideal| / n_ideal is calculated. When DN>0.4 (i.e. the deviation exceeds 40%), it is marked as a density anomaly point. To calculate the bubble distribution uniformity index, local standard deviation analysis is used: within a spherical area with a radius of 5 mm, the bubble size standard deviation σ is calculated. r (x, y, z) and density standard deviation σ n (x, y, z), when the coefficient of variation CV r =σ r / R - >0.3 or CV n =σ n / n - When it is >0.4, it is marked as an unevenly distributed point, where R - represents the average bubble size in the local area, n -Represents the average value of bubble density in a local area. The bubble morphology analysis uses the ellipticity index EI = a / b, where a and b are the major and minor axes of the bubble respectively. When EI>1.5, it is marked as a morphological abnormal point. Combining the above four types of abnormal indicators, the foaming quality index FQI (x, y, z) = w1 (1-DR (x, y, z)) + w2 (1-DN (x, y, z)) + w3 (1-CV r (x, y, z)) + w4·(1-EI(x, y, z) / 1.5), where the weight coefficients w1=0.4, w2=0.3, w3=0.2, and w4=0.1. When the foaming quality index FQI<0.6, the area is marked as a foaming abnormality area. All abnormal area data are mapped to a regular three-dimensional grid with a resolution of 1 mm. For each grid cell (i, j, k), an abnormality type marking matrix M[i, j, k] is established with a dimension of 5×3. The rows represent the abnormality type (temperature, pressure, flow rate, shear, foaming), and the columns represent the degree of abnormality (1=mild, 2=moderate, 3=severe). A weight coefficient is set for each type of anomaly: temperature anomaly weight w_T = 0.3, pressure anomaly weight w_P = 0.4, velocity anomaly weight w_V = 0.2, shear anomaly weight w_S = 0.1, and foaming anomaly weight w_F = 1.0. First, the weighted severity of each type of anomaly is calculated, then the sum of the flow field anomalies is calculated, and finally the comprehensive anomaly index is calculated. When the comprehensive anomaly index is > 0.5, the unit is marked as a high-risk area.

[0106] Preferably, step S4 includes the following steps:

[0107] Step S41: evaluating the abnormal factor contribution index based on the comprehensive flow field abnormal area data, and then dividing it into temperature-dominated abnormal area, pressure-dominated abnormal area, velocity-dominated abnormal area, and shear-dominated abnormal area according to the abnormal factor contribution index;

[0108] Step S42: Optimizing the abnormal region plastic process parameters for the temperature-dominated abnormal region, the pressure-dominated abnormal region, the velocity-dominated abnormal region, and the shear-dominated abnormal region based on the abnormal factor contribution index to generate plastic process optimization parameters.

[0109] In an embodiment of the present invention, an abnormal region feature matrix M is constructed, and the matrix dimension is n×4, where n is the number of abnormal regions, and the four columns correspond to the temperature anomaly index It, the pressure anomaly index Ip, the velocity anomaly index Iv, and the shear anomaly index Is, respectively. For abnormal region i, the calculation formulas of each index are as follows: temperature anomaly index It(i) = (Tmax(i) - Tref) / ΔTcrit, where Tmax(i) is the maximum temperature of the region, Tref is the ideal processing temperature, and ΔTcrit is the critical temperature deviation (set to 30°C); pressure anomaly index Ip(i) = |Pavg(i) - Pref| / Pref, where Pavg(i) is the average pressure of the region and Pref is the ideal holding pressure; flow velocity anomaly index Iv(i) = (θavg(i) / 45°) × (0.5 / |v|avg(i)), where θavg(i) is the average flow direction deviation angle and |v|avg(i) is the average flow velocity (unit: cm / s); shear anomaly index Is(i) = (SR_avg(i) / 1500) × (τs(i) / 1.5), where SR_avg(i) is the average shear rate and τs(i) is the duration of the over-threshold condition. Each index is normalized to unify the numerical range to the interval of 0-1. The anomaly proportion coefficient β(i, j) is introduced to represent the severity of type j anomaly in region i. The calculation formula is β(i, j) = Aj(i) / A(i)×Sj(i), where Aj(i) is the area of type j anomaly in region i, A(i) is the total area of the region, and Sj(i) is the severity of type j anomaly (1 = mild, 2 = moderate, 3 = severe). The abnormal factor contribution index CFI(i, j) = Sj(i)×β(i, j) is calculated, and the dominant anomaly type is determined according to the maximum value of CFI. Based on the abnormal factor contribution index, the plastic process parameters of the four dominant abnormal areas are optimized. For the temperature-dominated abnormal area, the response surface method is used to optimize the melting temperature parameters, and a quadratic response surface between the temperature anomaly index and the melt temperature and mold temperature is established: It = a0+a1T_melt+a2T_mold+a3T_melt 2 +a4T_mold 2 +a5T_melt·T_mold, where the coefficients are obtained through fitting multiple simulation tests. The optimization goal is to minimize the temperature anomaly index It, and the constraints are: the lower limit melting temperature of the plastic material +15°C ≤ T_melt ≤ the upper limit melting temperature of the plastic material -10°C; the lower limit of the recommended mold temperature of the material +5°C ≤ T_mold ≤ the upper limit of the recommended mold temperature of the material -5°C. For pressure-dominated anomaly areas, the gradient descent method is used to optimize the injection pressure and holding pressure parameters. The design variables include injection pressure P_inj, holding pressure P_hold, and holding time t_hold. The pressure anomaly index optimization function is established: Ip_opt = w1(P_max - P_target) 2+w2(P_min-0.8P_target) 2 +w3(dP / dx-dP_limit) 2 , where w1=0.4, w2=0.4, w3=0.2 are weight coefficients, P_target is the target pressure, and dP_limit is the pressure gradient limit (12MPa / mm). For velocity-dominated abnormal areas, a genetic algorithm is used to optimize the injection velocity curve, and the injection process is divided into n segments (typical value n=5), with the velocity of each segment being v1, v2, ..., v n . The fitness function is designed as: f_v = 1 / (α1Flow_index+α2Air_trap+α3Fill_balance), where Flow_index is the flow uniformity index, Air_trap is the number of gas traps, Fill_balance is the filling balance, α1 = 0.5, α2 = 0.3, α3 = 0.2 are weight coefficients. For shear-dominated abnormal areas, the particle swarm optimization algorithm is used to adjust the runner geometric parameters, including the runner diameter d_runner, the gate size d_gate, and the gate shape factor S_gate. The objective function is: min(Is) = min(SRmax / SRcrit), where Is is the shear strength index, SRmax is the maximum shear rate in the runner system, and SRcrit is the critical shear rate that the plastic material can withstand (unit: s -1 ), the constraints are: 2mm≤d_runner≤8mm, 0.5mm≤d_gate≤3mm, 1≤S_gate≤3. The optimized parameters of various abnormal areas are checked for conflicts. When there are contradictions in the parameters, they are processed by weighted average according to the severity of the abnormality. Finally, the plastic process optimization parameters are generated, including melt temperature (adjustment range ±15℃), mold temperature (adjustment range ±10℃), injection pressure curve (pressure values at 5-7 time points), injection speed curve (speed values at 5-7 time points), holding parameters (pressure, time, switching point) and supercritical fluid injection parameters (proportion, injection time, mixing temperature). The robustness of the optimized parameters is evaluated through sensitivity analysis to ensure good performance within the range of process fluctuations of ±5%.

[0110] Preferably, the present invention further provides an integrated design system for a three-dimensional simulation model of a plastic mold, which executes the integrated design method for a three-dimensional simulation model of a plastic mold as described above. The integrated design system for a three-dimensional simulation model of a plastic mold includes:

[0111] A mold geometry configuration module is used to perform geometric scanning on the plastic mold cavity to obtain mold cavity geometry data; based on the mold cavity geometry data, mold foaming simulation configuration data is set; wherein, the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology structure, and candidate foaming injection point locations;

[0112] The plastic flow simulation module is used to obtain plastic mold processing parameters; construct an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; and use the initial simulation model of the plastic mold to perform plastic melt flow-foaming injection simulation to obtain injection flow field state data;

[0113] The foaming abnormality diagnosis module is used to analyze the injection flow field abnormal area based on the injection flow field state data; identify the foaming abnormal area based on the injection flow field state data, and perform abnormal space superposition analysis based on the injection flow field abnormal area to generate comprehensive flow field abnormal area data;

[0114] The process parameter optimization module is used to optimize the plastic process parameters in abnormal areas based on the comprehensive flow field abnormal area data, generate plastic process optimization parameters, and realize the integrated design of the plastic mold injection filling stage.

[0115] The present invention realizes high-precision reconstruction of the three-dimensional simulation model of the plastic mold and automatic optimization of the simulation process parameters by adopting an integrated design method that deeply integrates the three-dimensional geometric scanning of the plastic mold cavity with the numerical simulation data. It can obtain multi-field coupling data such as plastic melt flow, temperature, pressure, shear rate, etc. in real time during the simulation of plastic mold processing parameters, supercritical fluid injection and foaming injection, and monitor the injection flow field state at all times, thereby realizing accurate diagnosis of abnormal areas and optimization of regional process parameters. Through the spatial superposition analysis of abnormal areas of the injection flow field, the present application can not only quickly identify local areas where key parameters such as temperature, pressure, flow rate and shear rate do not meet the preset standards, but also quantitatively evaluate the foaming defects caused by each abnormal area, thereby realizing targeted adjustment of plastic process parameters in the abnormal area. It has strong universality and scalability, is suitable for the design requirements of plastic molds with various complex structures, and can shorten the design cycle, reduce production costs, and effectively improve product quality while ensuring high-precision simulation. By closely combining shape, size, flow characteristics and processing parameters, it can not only effectively reduce the risk of defects caused by parameter mismatch in mold design, but also realize adaptive optimization of process parameters according to simulation results, providing a scientific, systematic and automated design method for mold manufacturing and plastic foam molding.

[0116] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0117] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. An integrated design method for a three-dimensional simulation model of a plastic mold, characterized in that: The following steps are involved: Step S1: performing geometric scanning on the plastic mold cavity to obtain mold cavity geometric structure data; Setting mold foaming simulation configuration data according to mold cavity geometry data; wherein the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology structure, and candidate foaming injection point locations; Step S2: Obtaining plastic mold processing parameters; constructing an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; performing a plastic melt flow-foaming injection simulation using the initial simulation model of the plastic mold to obtain injection flow field state data; Step S3: analyzing the injection flow field abnormal area according to the injection flow field state data; identifying the foaming abnormal area based on the injection flow field state data, and performing abnormal space superposition analysis on the injection flow field abnormal area to generate comprehensive flow field abnormal area data; Step S4: Optimizing the plastic process parameters in the abnormal area based on the comprehensive flow field abnormal area data, generating plastic process optimization parameters to achieve integrated design of the plastic mold injection filling stage.

2. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: The geometric scanning of the plastic mold cavity in step S1 includes: The plastic mold cavity is geometrically scanned by a 3D laser scanner. The scanning frequency is set to 50 Hz, the laser power is 2.5 W, and the point density is not less than 100 points / cm 2 , obtain the point cloud data of the inner surface of the cavity; The point cloud data of the inner surface of the cavity is filtered, and the filtering threshold is set to 3 times the standard deviation of the distance between points. When the distance between points is greater than 0.15 mm, it is determined as a noise point and deleted to obtain the filtered cavity point cloud data; Set the smoothing coefficient to 0.8, smooth the filtered cavity point cloud data, and obtain accurate cavity point cloud data; Convert the precise cavity point cloud data into a 3D surface model, determine the key dimensional parameters of the cavity such as length, width, depth, wall thickness and chamfer radius, and obtain the cavity parameters; Identify the mold's inlet position, gate shape and size, runner layout, and cooling channel location based on cavity parameters, and generate the mold structure. The cavity parameters and mold structure are integrated to obtain the mold cavity geometric structure data.

3. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: In step S1, setting mold foaming simulation configuration data according to the mold cavity geometry data includes: Surface meshing is performed based on the cavity parameters in the mold cavity geometry data. The mesh size is set to 1.0±0.2mm, the curvature sensitivity is set to 0.2, and the surface mesh data is generated. Perform quality inspection on the surface mesh data to ensure that the minimum angle of the surface mesh is not less than 30 degrees and the maximum angle is not greater than 150 degrees, and obtain mesh quality inspection data; Re-optimize the surface mesh data based on the mesh quality test data, and perform volume mesh conversion to obtain the cavity mesh data; Constructing the cavity flow channel network topology structure according to the mold structure in the mold cavity geometry data; Based on the cavity mesh data and the cavity flow channel network topology structure, the distance from each part of the cavity to the inlet is calculated to obtain the inlet distance distribution data; Calculate the cavity gate distance gradient vector field based on the inlet distance distribution data, determine the flow length path in the cavity, and generate flow path data; Candidate foaming injection point locations are extracted based on flow path data.

4. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 3, characterized in that: The cavity flow channel network topology structure is constructed based on the mold structure in the mold cavity geometry data as follows: Based on the cavity mesh data, the mold structure in the mold cavity geometry data is used to extract the center path of each runner segment and gate to obtain the runner centerline data; Mark the runner entrance, bifurcation point, confluence point and gate center point according to the runner centerline data to obtain the runner topology node set; Analyze the connection relationship of flow channel nodes based on the flow channel topology node set and flow channel centerline data; Based on the connection relationship of the flow channel nodes and the flow channel topology node set, an adjacency table is constructed with nodes as vertices and connection relationships as edges to obtain the initial flow channel topology data; Assign attributes to the initial topological data of the runner to generate the cavity runner network topology structure.

5. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 3, characterized in that: The candidate foaming injection point locations extracted based on the flow path data are as follows: Calculate the shortest distance from each area of the cavity to the boundary based on the cavity mesh data and generate distance distribution data; Perform distance extreme value analysis on the distance distribution data, take the maximum distance point as the flow confluence area, and the minimum distance point as the boundary approach area, and generate flow feature point data; In the flow feature point data, points with a wall thickness greater than 1.5 mm and a flow confluence area exceeding 30% of the maximum cavity size are selected as initial candidate injection points. Calculate the average flow distance from each point in the initial candidate injection point to all areas of the cavity, and select the five points with the smallest average flow distance as the candidate injection point position sequence; Using the flow path data, the flow balance of different foaming injection point locations in the candidate injection point location sequence is evaluated, and the flow length uniformity index is calculated to obtain injection point evaluation data. The index is defined as the ratio of the longest flow path to the shortest flow path. According to the injection point evaluation data, the position where the flow length uniformity index is closest to 1 is selected as the candidate foaming injection point position.

6. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: Plastic mold processing parameters include mold material parameters, plastic material parameters, and supercritical fluid parameters; Step S22: performing injection process parameter processing on the candidate foaming injection point position based on the supercritical fluid parameters to obtain injection process parameters; Step S23: importing mold material parameters, plastic material parameters, supercritical fluid parameters, and injection process parameters into Moldflow software based on the mold foaming simulation configuration data to construct an initial simulation model of the plastic mold; Step S24: using the initial simulation model of the plastic mold to perform plastic melt flow-foaming injection simulation, and periodically collecting temperature field data, pressure field data, flow velocity field data, and shear rate field data in the cavity; Step S25: Injection filling state integration is performed based on the temperature field data, the pressure field data, the flow velocity field data, and the shear rate field data to obtain injection flow field state data.

7. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: The abnormal injection flow field area includes the abnormal temperature area, the abnormal pressure area, the abnormal velocity area and the abnormal shear area. The abnormal injection flow field area analyzed according to the injection flow field state data in step S3 includes: Based on the temperature field data in the injection flow field state data, the average temperature and temperature standard deviation of each area in the cavity are calculated to generate temperature distribution statistics; Based on the temperature distribution statistics, identify areas where the temperature is 15°C below the glass transition temperature of the plastic material, or where the temperature standard deviation is greater than 5°C, and mark them as temperature anomaly areas; Based on the pressure field data in the injection flow field state data, the average pressure and pressure gradient of each area in the cavity are calculated to generate pressure distribution statistics; Based on the pressure distribution statistics, the areas where the pressure is lower than 80% of the set holding pressure, or the areas where the pressure gradient is greater than 12 MPa / mm are identified as abnormal pressure areas; The velocity vector field in the cavity is analyzed based on the velocity field data in the injection flow field state data. When the velocity direction deviates from the main flow direction by more than 45 degrees and the velocity is lower than 0.5 cm / s, the area is marked as an abnormal velocity area. Based on the shear rate field data in the injection flow field state data, the shear rate abnormality threshold is set to 1500s-1. When the shear rate exceeds the threshold and the duration exceeds 1.5s, the area is marked as a shear abnormality area.

8. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: In step S3, the abnormal foaming area is identified based on the injection flow field state data, and abnormal space superposition analysis is performed based on the abnormal injection flow field area, including: The temperature field data and pressure field data in the injection flow field state data are used to calculate the bubble nucleation rate in each area, thereby generating the bubble nucleation area distribution data; Calculate the initial density and initial size of bubbles based on the bubble nucleation area distribution data; Analyze the simulation holding time parameters based on the injection flow field state data; Based on the simulation holding time parameters, the bubble diameter increment is calculated by the bubble initial density and initial size to evaluate the plastic foaming diameter and bubble distribution density in each area to obtain the plastic simulation foaming data; Compare the plastic simulation foaming data with the preset ideal foaming parameters. When the plastic foam diameter deviates from the preset ideal foaming parameters by more than 30% or the bubble distribution density deviates by more than 40%, it is marked as a foaming abnormal area. The foaming abnormal area and the injection flow field abnormal area are spatially superimposed and analyzed, and the comprehensive abnormality index is calculated based on the overlap weight coefficient. When the comprehensive abnormality index exceeds 0.5, it is marked as a high-risk area, and the comprehensive flow field abnormality area data is generated; among them, the overlap weight coefficient sets the temperature abnormality weight to 0.3, the pressure abnormality weight to 0.4, the flow velocity abnormality weight to 0.2, and the shear abnormality weight to 0.

1.

9. The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: evaluating the abnormal factor contribution index based on the comprehensive flow field abnormal area data, and then dividing it into temperature-dominated abnormal area, pressure-dominated abnormal area, velocity-dominated abnormal area, and shear-dominated abnormal area according to the abnormal factor contribution index; Step S42: Optimizing the abnormal region plastic process parameters for the temperature-dominated abnormal region, the pressure-dominated abnormal region, the velocity-dominated abnormal region, and the shear-dominated abnormal region based on the abnormal factor contribution index to generate plastic process optimization parameters.

10. An integrated design system for a three-dimensional simulation model of a plastic mold, characterized in that: The integrated design method for a three-dimensional simulation model of a plastic mold according to claim 1 is used to implement the integrated design system for a three-dimensional simulation model of a plastic mold, comprising: A mold geometry configuration module is used to perform geometric scanning on the plastic mold cavity to obtain mold cavity geometry data; based on the mold cavity geometry data, mold foaming simulation configuration data is set; wherein, the mold foaming simulation configuration data includes cavity mesh data, cavity flow channel network topology structure, and candidate foaming injection point locations; The plastic flow simulation module is used to obtain plastic mold processing parameters; construct an initial simulation model of the plastic mold based on the mold foaming simulation configuration data and the plastic mold processing parameters; and use the initial simulation model of the plastic mold to perform plastic melt flow-foaming injection simulation to obtain injection flow field state data; The foaming abnormality diagnosis module is used to analyze the injection flow field abnormal area based on the injection flow field state data; identify the foaming abnormal area based on the injection flow field state data, and perform abnormal space superposition analysis based on the injection flow field abnormal area to generate comprehensive flow field abnormal area data; The process parameter optimization module is used to optimize the plastic process parameters in abnormal areas based on the comprehensive flow field abnormal area data, generate plastic process optimization parameters, and realize the integrated design of the plastic mold injection filling stage.

Images