CN111444619B - Online analysis method and equipment for injection mold cooling system - Google Patents

Online analysis method and equipment for injection mold cooling system Download PDF

Info

Publication number
CN111444619B
CN111444619B CN202010238565.9A CN202010238565A CN111444619B CN 111444619 B CN111444619 B CN 111444619B CN 202010238565 A CN202010238565 A CN 202010238565A CN 111444619 B CN111444619 B CN 111444619B
Authority
CN
China
Prior art keywords
cooling
cooling system
model
heat source
plastic part
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010238565.9A
Other languages
Chinese (zh)
Other versions
CN111444619A (en
Inventor
张云
王明中
黄志高
侯斌魁
周华民
李德群
吕帅
周晓伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CN202010238565.9A priority Critical patent/CN111444619B/en
Publication of CN111444619A publication Critical patent/CN111444619A/en
Application granted granted Critical
Publication of CN111444619B publication Critical patent/CN111444619B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C33/00—Moulds or cores; Details thereof or accessories therefor
    • B29C33/38—Moulds or cores; Details thereof or accessories therefor characterised by the material or the manufacturing process
    • B29C33/3835—Designing moulds, e.g. using CAD-CAM
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
    • B29C45/17—Component parts, details or accessories; Auxiliary operations
    • B29C45/72—Heating or cooling
    • B29C45/73—Heating or cooling of the mould
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • Manufacturing & Machinery (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Injection Moulding Of Plastics Or The Like (AREA)
  • Moulds For Moulding Plastics Or The Like (AREA)

Abstract

The invention belongs to the field of mold design and evaluation, and discloses an online analysis method and equipment for an injection mold cooling system. The method comprises the following steps: performing model cooling analysis on the mold and the plastic part without considering a cooling system to obtain internal factors representing the influence of the internal factors on the cooling effect of the plastic part; introducing a finite-length continuous-line heat source model of a constant-temperature boundary, and calculating an external factor representing the influence of the external factor on the cooling effect of the plastic part; simulating the whole system model to obtain the cooling time of the plastic part, namely the comprehensive heat influence effect; establishing a neural network model of an internal factor, an external factor and a comprehensive heat influence effect and training; and during on-line analysis, recalculating the external factors according to the size and position data of the modified cooling pipeline, and predicting the comprehensive heat influence effect by adopting the trained neural network model without recalculating the internal factors. The method is simple and easy to implement, and can realize the on-line analysis of the cooling system.

Description

Online analysis method and equipment for injection mold cooling system
Technical Field
The invention belongs to the field of mold design and evaluation, and particularly relates to an online analysis method and equipment for an injection mold cooling system.
Background
The injection molding is widely applied due to high production efficiency, good product quality, less material consumption and low production cost, and the output of the existing injection mold accounts for more than one third of the design and production of the whole plastic mold. With the development of the trend of replacing steel with plastic and replacing wood with plastic, plastic products are increasingly applied in daily life and industry, and the requirements of industries such as automobiles, energy sources, machinery, electronics, information, aerospace and the like on injection molding technology are continuously improved.
The injection molding process mainly comprises the stages of mold locking, plasticizing, injecting, pressure maintaining, cooling, mold opening, ejecting and the like, wherein the cooling time accounts for about 50-80% of the molding period. The injection mold cooling system is an important factor for determining the quality and the production efficiency of a plastic part, not only determines the molding performance, the dimensional precision and the mechanical property of the plastic part, avoids the defects of uneven wall thickness, warping deformation, dimensional change, residual stress and the like of the plastic part, but also influences the length of a molding cycle by determining a cooling process. The reasonable cooling system can adjust the temperature environment inside the injection mold, and realizes uniform and rapid cooling of the plastic part, thereby ensuring the quality of the plastic part, shortening the cycle of injection molding and improving the production efficiency. Therefore, the cooling system is rapidly analyzed in the mold design stage and then used as the basis for adjusting and optimizing the cooling system, and the method has great significance for shortening the design period of the whole mold and the molding period in the actual production of the mold.
The design and analysis of conventional injection mold cooling systems relies primarily on the experience and intuition of designers, which places high demands on the level of designers, and is detrimental to the work of primary designers. In addition, the design relying on experience lacks theoretical basis and scientific calculation, whether the design is reasonable or not needs to be judged through continuous mold testing, and the defect of the plastic part is mainly solved by mold repairing, so that the production efficiency of the plastic part is low, the cost is high, and the quality is difficult to guarantee. With the development of computer application technology, the application of Computer Aided Engineering (CAE) in the field of injection molds has enabled the design efficiency and quality of injection mold cooling systems to be greatly improved. The CAE mainly adopts numerical analysis methods such as Finite Element (FEM), finite Difference (FDM) and Boundary Element (BEM), the cooling process is analyzed through an iteration method, and the cooling time in the analysis result can reflect the comprehensive thermal influence effect of the plastic part. Although the CAE result is accurate, the CAD model needs to be reconstructed into the CAE model with the corresponding format, the grid needs to be divided again after the cooling system is modified, the operation is complex, multiple times of modification needs multiple times of analysis, the calculated amount is large, the time is long, and the requirements of online analysis are difficult to meet.
Disclosure of Invention
Aiming at the defects or the improvement requirements of the prior art, the invention provides an online analysis method for a cooling system of an injection mold, which aims to divide the factors influencing the cooling of plastic products into an internal factor and an external factor, respectively calculate the internal factor and the external factor according to the two factors, and calculate the comprehensive heat influence effect reflecting the actual cooling condition of the products by combining the internal factor and the external factor, thereby realizing the online analysis of the cooling system and solving the technical problems of long analysis time and low efficiency of the current cooling system.
To achieve the above object, according to an aspect of the present invention, there is provided an online analysis method of an injection mold cooling system, comprising the steps of:
an off-line training stage:
step 1: the influence of a cooling system is not considered, and the model of the mold and the plastic part is dispersed into a triangular unit; the triangular units comprising the same node are adjacent units, any triangular unit is selected as a central unit, the cooling time of the central unit and the adjacent unit is used as an internal factor for representing the influence of the internal factor on the cooling effect of the plastic part, and the internal factor is a matrix and is recorded as:
Figure BDA0002431813430000021
wherein, theta in Representing the internal factor, t represents the cooling time of the currently selected delta unit,
Figure BDA0002431813430000022
respectively represent 1 st turn 1 st to n 1 The cooling time of each of the triangular units,
Figure BDA0002431813430000023
respectively represent the 1 st to n nd circles 2 2 The cooling time of each of the triangular units,
Figure BDA0002431813430000024
and
Figure BDA0002431813430000025
respectively represent the 1 st to n th circles 3 3 The cooling time of each of the triangular units,
Figure BDA0002431813430000031
and
Figure BDA0002431813430000032
respectively represent the 1 st to n th circles m Cooling time of each triangular unit;
step 2: introducing a finite-length continuous-line heat source model of a constant temperature boundary into the model in the step 1 to cool the model in the step 1, thereby calculating an external factor representing the influence of the external factor on the cooling effect of the plastic part, wherein the formula is as follows:
Figure BDA0002431813430000033
wherein, theta ex Represents external factors, i =1 \ 8230q represents different boundary surfaces of the die, q represents the number of boundary surfaces of the die,
Figure BDA0002431813430000034
and
Figure BDA0002431813430000035
respectively represent the coordinates of the center cell centroid with respect to the source line heat source axis direction and the coordinates of the source line heat source axis direction with respect to the mirror image line of the boundary surface i,
Figure BDA0002431813430000036
and
Figure BDA0002431813430000037
respectively represents the coordinates of two end points of the source line heat source in the axial direction,
Figure BDA0002431813430000038
and
Figure BDA0002431813430000039
respectively represent the coordinates of the two end points of the source line heat source with respect to the mirror image line heat source of the boundary surface i in the axial direction,
r i 1 and r i 2 Respectively representing the distance from the center unit mass center relative to the source line heat source and the distance from the center unit mass center to the mirror image line heat source of the source line heat source relative to the boundary surface i;
and 3, step 3: taking the mould and the model of the plastic part in the step 1 and the initial cooling system model in the step 2 as a whole, and obtaining the cooling time of the plastic part through simulation, namely the comprehensive thermal influence effect of the plastic part;
and 4, step 4: establishing a neural network model, and using the internal factor theta of the step 1 in External factor θ of step 2 ex As input featuresTaking the comprehensive heat influence effect of the step 3 as expected output, and training the neural network model to obtain a comprehensive heat influence effect prediction model;
an online analysis stage:
and 5: in the adjusting and optimizing process of the cooling system of the injection mold, according to the size and position data of the adjusted cooling pipeline, the external factor is recalculated according to the method in the step 2, the internal factor directly follows the result in the step 1, and then the comprehensive heat influence effect prediction model obtained by training in the step 4 is adopted to predict the comprehensive heat influence effect.
Further, in step 1, m has a value of 1 to 5.
Further, the neural network model in step 4 is a BP neural network model, and the BP neural network model includes four layers: an input layer, a hidden layer 1, a hidden layer 2 and an output layer; the number m of the input layer nodes Into Between 1 and 60, equal to the total number of central and adjacent cells selected; the number h of nodes of the hidden layer 1 1 ≥2m Into +1, the number of nodes of the hidden layer 2 is between 2 and 20, the number of nodes of the output layer is 1, and the transfer function of the nodes selects a Sigmoid function.
Further, the BP neural network model in step four is iteratively optimized by adopting a method of variable step size adjustment and batch processing.
Further, in step 4, before training, the internal factors and the external factors are respectively used as raw data, and maximum and minimum normalization processing is respectively performed according to the following formula:
Figure BDA0002431813430000041
wherein x is 0 、x min 、x max And x norm Respectively representing the original data, the minimum value in the original data sequence, the maximum value in the original data sequence and the data after maximum and minimum normalization.
To achieve the above object, according to another aspect of the present invention, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method as described in any one of the preceding claims.
To achieve the above object, according to another aspect of the present invention, there is provided an online analysis device for an injection mold cooling system, which is characterized by comprising the computer-readable storage medium as described above and a processor for calling and processing a computer program stored in the computer-readable storage medium.
In general, compared with the prior art, the above technical solution contemplated by the present invention can obtain the following beneficial effects:
1. the analysis method can be directly operated in a CAE system, wherein the offline training stage only needs two CAE analyses at most, namely the analysis of internal factors and comprehensive thermal influence effects; while the on-line analysis phase can directly follow the internal factors of the off-line training phase, and the external factors can be directly recalculated without re-dividing the finite element mesh. Therefore, the method greatly reduces the iterative computation times, has simple process, small computation amount and time saving, can realize the on-line analysis of the cooling system, and can realize the rapid and even real-time computation under the condition of ensuring that the cooling analysis result meets the use precision.
2. The number of adjacent cells selected in step 1 is preferably the number of grid cells contained in 1 to 5 circles of outward expansion centered on the calculation cell, which can further reduce the calculation amount and the calculation time on the premise of ensuring the calculation accuracy.
3. By selecting the BP neural network model and designing the specific structure and the node number of the BP neural network model, the accuracy can be ensured and the training efficiency can be improved.
4. The BP neural network model adopts a variable step length adjustment and batch processing method to improve the algorithm, and can overcome the defects of long training time and slow convergence of the traditional algorithm.
5. Before training, the maximum and minimum normalization is carried out on the internal factors and the external factors, so that the accuracy and the convergence speed of the model can be further improved.
Drawings
FIG. 1 is a schematic flow chart of the algorithm of the present invention;
FIG. 2 is a schematic heat transfer diagram of a grid cell;
FIG. 3 is a schematic diagram of a finite-length continuous-line heat source for a constant temperature boundary;
FIG. 4 is a schematic diagram of a BP neural network of the present invention;
fig. 5 (a) and (b) are schematic diagrams comparing the predicted results and the actual results of the embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
The basic principle of the invention is as follows: and (3) taking the unit cooling time of CAE analysis when a cooling system is not designed as an internal factor for representing the influence on the cooling of the plastic part, calculating an external factor for representing the influence on the cooling of the plastic part by using a simplified finite-length continuous-line heat source model of a constant temperature boundary, and obtaining the comprehensive heat influence effect on the plastic part through the internal factor and the external factor.
The method divides the factors determining the cooling result into two types: internal factors including the material, shape, size of the molding and mold, and the mold temperature, and external factors including the size and position of the cooling ducts and the temperature, flow rate, etc. of the cooling medium. The internal factors are constant for the same pair of molds, and the influence of the internal factors on the cooling process, also called internal factors, is characterized by the size of the cooling time obtained by CAE analysis when the cooling system is not designed. The cooling system is modified and optimized, the size and the position of the cooling pipeline are changed, generally, the length of the cooling pipeline is 1-2 orders of magnitude larger than the diameter of the cooling pipeline, therefore, the cooling pipeline can be approximated to a linear heat source, and the influence effect of the cooling pipeline, which is also called an external factor, is obtained by solving a Fourier heat conduction differential equation. According to the method, only two CAE analyses are needed, a prediction model is constructed through a BP neural network, the cooling time (internal factor) of each grid unit of the plastic part obtained through CAE analysis when a cooling system is not designed and the external factor obtained through calculation are used as the input of a training model, and the cooling time of each grid unit of the plastic part obtained through CAE analysis including the initial cooling system is used as the output of the training model. In the adjusting and optimizing stage of the cooling system, only the external factor needs to be recalculated, and then the comprehensive heat influence effect is obtained by utilizing the BP neural network prediction model, so that the online analysis of the cooling system is realized.
The invention adopts the following technical scheme:
step 1: and (4) carrying out cooling analysis on the model only comprising the mould and the plastic part by using CAE software to obtain the cooling time of each grid unit of the plastic part. Because the temperature of each unit of the plastic part is different, heat transfer caused by temperature difference exists, and the cooling process of the unit is further influenced. After the surface of the plastic part is dispersed into triangular units, the units have geometrical topological relation, the triangular units comprising the same node are adjacent units, so that the cooling time of the unit and the adjacent units is selected as the internal factor of the unit together, and the internal factor is a matrix at the moment.
Preferably, the cooling process of each unit in step 1 is affected by all other units, increasing the number of adjacent units can better characterize internal factors, and further improve the accuracy of the comprehensive thermal influence effect, but the calculation cost and time are increased, and generally 1 to 5 circles of adjacent units are selected.
Step 2: designing an initial cooling system, calculating the centroid coordinates of all grid units and the endpoint coordinates of cooling pipelines, and calculating external factors through a finite-length continuous-line heat source model of a constant-temperature boundary, wherein the used formula is as follows:
Figure BDA0002431813430000071
theta is described ex Indicating the external factor, i =1 \ 82306 indicating the six boundary surfaces of the mold,
Figure BDA0002431813430000072
and
Figure BDA0002431813430000073
respectively representing the coordinates of the cell centroid with respect to the source line heat source axis direction and the coordinates of the source line heat source axis direction with respect to the mirror image line of the boundary surface i,
Figure BDA0002431813430000074
and
Figure BDA0002431813430000075
coordinates of two end points of the source line heat source in the axis direction are shown,
Figure BDA0002431813430000076
and
Figure BDA0002431813430000077
coordinates in the axial direction, r, of both end points of the source line heat source with respect to the mirror image line of the boundary surface i i 1 And r i 2 Respectively, represent the cell centroid distance relative to the source line heat source and the cell centroid distance to the mirror line heat source of the source line heat source with respect to boundary surface i.
And 3, step 3: and analyzing the scheme comprising the initial cooling system design by using CAE software to obtain the cooling time of the grid unit, namely the comprehensive thermal influence effect.
And 4, step 4: and (3) establishing a neural network model of an internal factor, an external factor and a comprehensive heat influence effect according to the data obtained in the steps 1, 2 and 3, wherein the internal factor and the external factor are used as input characteristics, and the comprehensive heat influence effect is used as expected output.
Preferably, the neural network model in step 4 adopts a BP neural network, and the model comprises four layers: input layer, hidden layer 1, hidden layer 2 and output layer. The number m of the nodes of the input layer Into Between 1 and 60, the number depending on the number of adjacent cells selected; the number h1 of the nodes of the hidden layer 1 is more than or equal to 2m Into +1, hiding layer 2The number of nodes is between 2 and 20, and the number of nodes of the output layer is 1. The node transfer function selects a Sigmoid function, and the formula is as follows:
Figure BDA0002431813430000078
the Levenberg-Marquardt algorithm is selected in the training method, and the network training speed is fastest for medium-scale networks. And (3) training the neural network, wherein the proper maximum training times, precision requirements, learning rate and minimum gradient requirements are selected according to the size of the data volume obtained in the step one.
Preferably, the maximum and minimum normalization processing is performed on the data obtained in step 1 and step 2, and each row of the internal factor matrix needs to be normalized separately for the external factors.
The maximum and minimum normalization method is as follows:
Figure BDA0002431813430000081
said x 0 、x min 、x max And x norm Respectively representing the original data, the minimum value in the original data sequence, the maximum value in the original data sequence and the normalized data.
Preferably, the BP neural network model adopts a variable step size adjustment and batch processing method to improve the algorithm, so as to overcome the defects of long training time and slow convergence of the traditional algorithm.
And 5: and in the adjustment and optimization process of the cooling system of the injection mold, recalculating the external factors according to the size and position data of the modified cooling pipeline, wherein the internal factors are not recalculated, and predicting the comprehensive heat influence effect by adopting the BP neural network model trained in the step 4.
The above method is more specifically described below by taking a plastic product as an example:
step 1: fig. 2 shows a part of the plastic product grid in this case, in which the units with common nodes are adjacent to each other, and in the drawing, two types of units with different shading, namely, a unit marked with a "(1)" in the center of fig. 2, of the unit 1 from the near side to the far side respectively represent a first circle of adjacent units and a second circle of adjacent units. Since unit 1 (i.e., the unit labeled "(1)" in the center of fig. 2) has a temperature difference from the adjacent units, there is heat transfer (as indicated by the arrows in fig. 2), and thus each unit cannot be calculated in isolation when the internal factors are considered.
The cooling of the plastic product is closely related to the shape and size of the plastic product, specifically, each unit is affected by all other units, but when the internal factors are calculated, a large amount of redundancy occurs by including all the units, the calculation amount and the calculation time are greatly increased, and therefore, the embodiment is preferably only required to have 1-5 layers of adjacent units. The expression of the internal factor is as follows:
Figure BDA0002431813430000082
theta is described in Represents an internal factor, t represents a cooling time of the current cell (i.e., the currently selected center cell),
Figure BDA0002431813430000091
and
Figure BDA0002431813430000092
respectively representing the 1 st adjacent cell and the n-th adjacent cell of the 1 st turn 1 The cooling time of the individual cells is,
Figure BDA0002431813430000093
and
Figure BDA0002431813430000094
respectively representing the 1 st adjacent cell and the n-th adjacent cell of the 2 nd turn 2 The cooling time of the individual units is,
Figure BDA0002431813430000095
and
Figure BDA0002431813430000096
respectively represent the 1 st adjacent cell and the n-th adjacent cell of the 3 rd turn 3 The cooling time of the individual units is,
Figure BDA0002431813430000097
and
Figure BDA0002431813430000098
respectively represent the 1 st adjacent unit and the n-th adjacent unit of the m-th circle m The cooling times of the individual units, which were calculated when the cooling system was not designed, were used. The nature of the cooling system which is not designed does not consider the influence of the cooling system, and the influence is reflected in the actual simulation calculation process, the modeling of the cooling system is not carried out firstly, and the processing is simpler. In other embodiments, the cooling system may be designed and then tuned to 0 for its effect on unit cooling during the simulation, but the model tuning process may be somewhat more complex.
Step 2: fig. 3 is a schematic diagram of a finite-length continuous-line heat source at a constant temperature boundary, an analytic solution of heat transfer of the heat source at an instantaneous point can be obtained according to a fourier heat conduction differential equation, and the analytic solution of the finite-length continuous-line heat source can be obtained by integrating in time and space.
Because the boundary is constant in temperature, a mirror image heat source is added at the position where the source heat source and the boundary are symmetrical, the difference between the source heat source and the boundary is an external factor, the module in the embodiment has 6 surfaces including an upper surface, a lower surface, a left surface, a right surface, a front surface and a rear surface, and 6 mirror image heat sources need to be arranged. Therefore, the formula for calculating the extrinsic factor is as follows:
Figure BDA0002431813430000099
theta is described ex Indicating the external factor, i =1 \ 82306 indicating the six boundary surfaces of the mold,
Figure BDA00024318134300000910
and
Figure BDA00024318134300000911
respectively representing the coordinates of the center cell centroid with respect to the source line heat source axis direction and the source line heat source with respect to the boundary surface iThe coordinates of the axis direction of the heat source of the mirror line,
Figure BDA00024318134300000912
and
Figure BDA00024318134300000913
coordinates of two end points of the source line heat source in the axis direction are shown,
Figure BDA00024318134300000914
and
Figure BDA00024318134300000915
coordinates in the axial direction, r, of both end points of the source line heat source with respect to the mirror image line of the boundary surface i i 1 And r i 2 Respectively representing the distance of the center cell centroid relative to the source line heat source and the distance of the cell centroid to the mirror line heat source of the source line heat source relative to boundary surface i. The above formula is applicable to molds having different numbers of boundary surfaces.
And 3, step 3: after obtaining the internal factor and the external factor, performing CAE analysis on the cooling system scheme containing the initial design to obtain a comprehensive thermal influence effect;
and 4, step 4: and then establishing a neural network model of the internal factors, the external factors and the comprehensive heat influence effect, taking the internal factors and the external factors as input characteristics of the neural network model, and taking the comprehensive heat influence effect as expected output of the neural network model.
Preferably, a BP neural network model is adopted in this case, the maximum iteration step number of model training is 1000, the learning rate is 0.1, and the allowable iteration error is 0.0001. As a simple exemplary demonstration, the BP neural network model designed in this example includes four layers: an input layer, a hidden layer 1, a hidden layer 2 and an output layer; the input layer represents an internal factor and an external factor and comprises 57 nodes; hidden layer 1 and hidden layer 2 contain 120 and 10 nodes, respectively; the output layer represents the composite thermal influence effect and comprises 1 node. The node transfer function selects a Sigmoid function, and the formula is as follows:
Figure BDA0002431813430000101
wherein x represents input data of a node and S (x) represents output data of the node;
preferably, the Levenberg-Marquardt algorithm is selected in the training method, and the algorithm is improved by adopting a variable step size adjusting and batch processing method in the training process, so that the defects of long training time and slow convergence of the traditional algorithm are overcome.
Preferably, before training, the internal factors and the external factors are subjected to maximum and minimum normalization processing. The normalization method comprises the following steps:
Figure BDA0002431813430000102
said x 0 、x min 、x max And x norm Respectively representing the original data, the minimum value in the original data sequence, the maximum value in the original data sequence and the maximum and minimum normalized data.
For the trained BP model, the method for online detection by using the model is as follows:
and 5: in the adjusting and optimizing process of the injection mold cooling system, according to the size and position data of the modified cooling pipeline, recalculating the external factors according to the method in the step 2, directly using the result in the step 1 without recalculating the internal factors, and then predicting the comprehensive heat influence effect by adopting the BP neural network model obtained by training in the step 4. The prediction result and the actual measurement result of the case are shown in fig. 5, and it can be known from fig. 5 that the variation rule of the prediction result and the actual result is highly consistent, so that the difference that each part of the plastic product is actually influenced by the cooling system due to different positions and structures can be accurately reflected, and the defects of the product, such as hot spots and the like, can be accurately predicted.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (7)

1. An online analysis method for an injection mold cooling system is characterized by comprising the following steps:
an off-line training stage:
step 1: the influence of a cooling system is not considered, and the model of the mold and the plastic part is dispersed into a triangular unit; the triangular units comprising the same node are adjacent units, any triangular unit is selected as a central unit, the cooling time of the central unit and the adjacent unit is used as an internal factor for representing the influence of the internal factor on the cooling effect of the plastic part, and the internal factor is a matrix and is recorded as:
Figure FDA0003796127710000011
wherein, theta in Representing the internal factor, t represents the cooling time of the currently selected delta unit,
Figure FDA0003796127710000012
respectively represent the 1 st turn 1 to n 1 The cooling time of each of the triangular units,
Figure FDA0003796127710000013
respectively represent the 1 st to n nd circles 2 2 The cooling time of each of the triangular units,
Figure FDA0003796127710000014
and
Figure FDA0003796127710000015
respectively represent the 1 st to n th circles 3 3 The cooling time of each of the triangular units,
Figure FDA0003796127710000016
and
Figure FDA0003796127710000017
respectively represent the 1 st to n th circles m The cooling time of each triangular unit;
step 2: introducing a finite-length continuous-line heat source model of a constant temperature boundary into the model in the step 1 to cool the model in the step 1, thereby calculating an external factor representing the influence of the external factor on the cooling effect of the plastic part, wherein the formula is as follows:
Figure FDA0003796127710000018
wherein, theta ex Represents external factors, i =1 \ 8230q represents different boundary surfaces of the die, q represents the number of boundary surfaces of the die,
Figure FDA0003796127710000019
and
Figure FDA00037961277100000110
respectively represent the coordinates of the center cell centroid with respect to the source line heat source axis direction and the coordinates of the source line heat source axis direction with respect to the mirror image line of the boundary surface i,
Figure FDA00037961277100000111
and
Figure FDA00037961277100000112
respectively represents the coordinates of two end points of the source line heat source in the axial direction,
Figure FDA00037961277100000113
and
Figure FDA00037961277100000114
respectively representing the coordinates of the two end points of the source line heat source relative to the mirror image line heat source of the boundary surface i in the axial direction，
r i 1 And r i 2 Respectively representing the distance from the center unit mass center relative to the source line heat source and the distance from the center unit mass center to the mirror image line heat source of the source line heat source relative to the boundary surface i;
and 3, step 3: taking the mould and the model of the plastic part in the step 1 and the initial cooling system model in the step 2 as a whole, and obtaining the cooling time of the plastic part through simulation, namely the comprehensive thermal influence effect of the plastic part;
and 4, step 4: establishing a neural network model, and determining the internal factor theta of the step 1 in External factor θ of step 2 ex As an input characteristic, taking the comprehensive thermal influence effect of the step 3 as an expected output, and training the neural network model to obtain a comprehensive thermal influence effect prediction model;
an online analysis stage:
and 5: in the adjusting and optimizing process of the injection mold cooling system, according to the size and position data of the adjusted cooling pipeline, the external factor is recalculated according to the method in the step 2, the internal factor directly continues the result in the step 1, and then the comprehensive heat influence effect prediction model obtained by training in the step 4 is adopted to predict the comprehensive heat influence effect.
2. An on-line analysis method for a cooling system of an injection mold according to claim 1, wherein m in step 1 has a value of 1 to 5.
3. The online analysis method for the cooling system of the injection mold according to claim 1, wherein the neural network model in the step 4 is a BP neural network model, and the BP neural network model comprises four layers: an input layer, a hidden layer 1, a hidden layer 2 and an output layer; the number m of the nodes of the input layer Into Between 1 and 60, equal to the total number of central and adjacent cells selected; the number h of nodes of the hidden layer 1 1 ≥2m Go into +1, the number of nodes of the hidden layer 2 is between 2 and 20, the number of nodes of the output layer is 1, and the transfer function of the nodes selects a Sigmoid function.
4. The on-line analysis method for the cooling system of the injection mold according to any one of claims 1 to 3, wherein the BP neural network model in the fourth step is iteratively optimized by adopting a variable step size adjustment and batch processing method.
5. An on-line analysis method for a cooling system of an injection mold according to any one of claims 1 to 3, wherein in step 4, before training, the internal factors and the external factors are respectively used as raw data, and maximum and minimum normalization processing is respectively performed according to the following formula:
Figure FDA0003796127710000031
wherein x is 0 、x min 、x max And x norm Respectively representing the original data, the minimum value in the original data sequence, the maximum value in the original data sequence and the maximum and minimum normalized data.
6. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method according to any one of the claims 1 to 5.
7. An on-line analysis apparatus for an injection mold cooling system, comprising the computer-readable storage medium of claim 6 and a processor for invoking and processing a computer program stored in the computer-readable storage medium.
CN202010238565.9A 2020-03-30 2020-03-30 Online analysis method and equipment for injection mold cooling system Active CN111444619B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010238565.9A CN111444619B (en) 2020-03-30 2020-03-30 Online analysis method and equipment for injection mold cooling system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010238565.9A CN111444619B (en) 2020-03-30 2020-03-30 Online analysis method and equipment for injection mold cooling system

Publications (2)

Publication Number Publication Date
CN111444619A CN111444619A (en) 2020-07-24
CN111444619B true CN111444619B (en) 2022-12-02

Family

ID=71649328

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010238565.9A Active CN111444619B (en) 2020-03-30 2020-03-30 Online analysis method and equipment for injection mold cooling system

Country Status (1)

Country Link
CN (1) CN111444619B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023024052A1 (en) * 2021-08-27 2023-03-02 Siemens Industry Software Inc. Construction of conformal cooling channels for injection mold designs
CN115890987B (en) * 2022-12-19 2023-09-05 苏州可米可酷食品有限公司 Intelligent flexible mold and application production system thereof

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1485195A (en) * 2002-08-16 2004-03-31 ����ʵҵ��ʽ���� Shut-off nozzle with heating unit and cooling unit for hot runner systems of injection molding machines, and method of controlling the same
CN101853318A (en) * 2010-05-11 2010-10-06 华中科技大学 Method for quickly forecasting cooling time in injection molding of plastics
CN103020361A (en) * 2012-12-17 2013-04-03 华南理工大学 Method for extracting no-checkerboard topological diagram from compliant mechanism
CN103266272A (en) * 2013-04-16 2013-08-28 安徽省宁国市宁沪钢球有限公司 Preparation method of cylinder liner of ball mill
KR20140022251A (en) * 2012-08-13 2014-02-24 유도썬스(주) Injection mold cooling systems valves unit device
CN110826225A (en) * 2019-11-06 2020-02-21 安徽工业大学 Vertical single-U-shaped buried pipe heat exchanger temperature field analysis method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130110306A1 (en) * 2011-10-26 2013-05-02 Zhikui Wang Managing multiple cooling systems in a facility

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1485195A (en) * 2002-08-16 2004-03-31 ����ʵҵ��ʽ���� Shut-off nozzle with heating unit and cooling unit for hot runner systems of injection molding machines, and method of controlling the same
CN101853318A (en) * 2010-05-11 2010-10-06 华中科技大学 Method for quickly forecasting cooling time in injection molding of plastics
KR20140022251A (en) * 2012-08-13 2014-02-24 유도썬스(주) Injection mold cooling systems valves unit device
CN103020361A (en) * 2012-12-17 2013-04-03 华南理工大学 Method for extracting no-checkerboard topological diagram from compliant mechanism
CN103266272A (en) * 2013-04-16 2013-08-28 安徽省宁国市宁沪钢球有限公司 Preparation method of cylinder liner of ball mill
CN110826225A (en) * 2019-11-06 2020-02-21 安徽工业大学 Vertical single-U-shaped buried pipe heat exchanger temperature field analysis method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Eddy Current Computation in Translational Motion Conductive Plate of an Induction Heater With Consideration of Finite Length Extremity Effects;M. Messadi等;《IEEE Transactions on Magnetics》;20151106;第1-4页 *
地源热泵竖直埋管数值线源综合模型;陆志等;《上海交通大学学报》;20080315(第03期);第109-414页 *
地源热泵竖直埋管的有限长线热源模型;曾和义等;《热能动力工程》;20030331;第166-169页 *
注射模冷却过程CAE技术;申长雨等;《模具工业》;20010425(第04期);第52-56页 *
边界元矩阵稀疏化算法及其应用;崔树标等;《上海交通大学学报》;20081031;第11618-1621页 *

Also Published As

Publication number Publication date
CN111444619A (en) 2020-07-24

Similar Documents

Publication Publication Date Title
Cao et al. Multi-objective optimization of injection-molded plastic parts using entropy weight, random forest, and genetic algorithm methods
CN108304657B (en) Finite element-based multi-process continuous modeling simulation method for residual stress of machine tool weight parts
CN113051831B (en) Modeling method and thermal error control method for thermal error self-learning prediction model of machine tool
CN104298818B (en) A kind of end mill processing surface error prediction and emulation mode
US20230252203A1 (en) Multidisciplinary structural design optimization method for fuel assembly based on co-simulation
US11328107B2 (en) Hybrid measurement and simulation based distortion compensation system for additive manufacturing processes
CN111444619B (en) Online analysis method and equipment for injection mold cooling system
CN112069610B (en) Injection molding process parameter optimization method for transparent complex multi-cavity plastic part
CN111723440B (en) Thin-wall part machining precision prediction hybrid modeling method
CN112926152B (en) Digital twin-driven thin-wall part clamping force precise control and optimization method
CN111605199A (en) Finite element analysis and forming precision method for FDM3D printer
Hwang et al. Injection mold design of reverse engineering using injection molding analysis and machine learning
Tlija et al. A novel model for the tolerancing of nonrigid part assemblies in computer aided design
CN115795933A (en) Metal insert injection molding optimization method and device and storage medium
Yang et al. Thermal error modeling by integrating GWO and ANFIS algorithms for the gear hobbing machine
Sun et al. Toward axial accuracy prediction and optimization of metal tube bending forming: A novel GRU-integrated Pb-NSGA-III optimization framework
Sun et al. Uncertainty calibration and quantification of surrogate model for estimating the machining distortion of thin-walled parts
Yang et al. Study of Residual Wall Thickness and Multiobjective Optimization for Process Parameters of Water‐Assisted Injection Molding
Wu et al. Adaptive sampling method for thin-walled parts based on on-machine measurement
CN114919181B (en) Continuous fiber 3D printing process dynamic simulation and printed part buckling deformation prediction method
Zhang et al. Optimized design method for profile extrusion die based on NURBS modeling
Yang et al. Research on Optimization of Injection Molding Process Parameters of Automobile Plastic Front‐End Frame
Li et al. Clamping force prediction based on deep spatio-temporal network for machining process of deformable parts
CN111210877A (en) Method and device for deducing physical property parameters
CN114372381A (en) Quality control method based on quadratic polynomial stepwise regression proxy model

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant