Method for producing thermo-physical material data for polymer-based process simulation

Abstract

To provide a method for improving a polymer-based production process.SOLUTION: The method includes: classifying a plurality of known polymers into polymer types; determining a representative parameter for each set of thermo-physical properties for each polymer type, based on measured parameters of these thermo-physical properties and using a statistical method; and generating a dataset for an unknown polymer based on a choice of a polymer type of the unknown polymer. The generated dataset can be used as input for a simulation of a polymer-based production process.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The disclosure of the present application (hereinafter referred to as the present disclosure) relates to the field of computer-based simulation of processes in the manufacturing area, and more particularly to the improvement of polymer-based manufacturing processes such as polymer injection, compression, transfer molding or extrusion. The present disclosure further relates to computer-based systems and methods for designing, setting, or controlling manufacturing machines involved in the aforementioned processes. Background

[0002] Injection molding, compression molding, or transfer molding is a manufacturing process that repeatedly pushes a liquid material (especially a polymer) into a cavity. Inside the cavity surrounded by a mold, the liquid material solidifies by cooling, vulcanization, etc. During the solidification process, the liquid material changes into a solid. When the material solidifies, it is taken out of the mold, and a new manufacturing cycle starts by injecting new material into the empty cavity. The solidified polymer becomes the product.

[0003] On the other hand, the extrusion process is used to make products with a constant cross-section. In this process, the liquid material is continuously pressed through a die. Then, since the liquid material solidifies, the final shape of the product is fixed.

[0004] During the solidification process, the material accumulates internal stress, which may lead to deformation of the product.

[0005] For example, it is desirable to set and control the manufacturing process by controlling manufacturing machine parameters such as temperature, injection speed, or pressure. Also, it is desirable to design the mold or die so that the characteristics and shape of the product after all manufacturing processes meet specific quality criteria (e.g., part shape, surface quality, mechanical or chemical properties).

[0006] The correlations between the parameters of manufacturing machines, the behavior of materials, and the aforementioned criteria are very complex. Computer-based simulations are known to be used in various scenarios, such as mold design including cavity shape and cooling layout, initial setting of machine parameters, control of injection molding machines during manufacturing, or understanding the reasons why the process does not function as expected.

[0007] In all scenarios, these correlations are governed by the product materials used and their thermal properties. Therefore, the product materials used and their thermal properties are extremely important inputs for computer-based simulations.

[0008] Examples of thermal property data required for computer-based simulations include thermal conductivity, specific heat, density, viscosity, curing rate, crystallization rate, etc. A dataset is called a collection of all or a subset of the required thermal property data.

[0009] It is known to measure the required thermal property values. For each required thermal property, special experimental setups and appropriate interpretations of experimental data have been developed. For example, viscosity can be measured by pressing a polymer melt through a die at a certain temperature. The pressure required for a given volume flow rate is measured. From this pressure and volume flow rate data, the viscosity as a function of shear rate can be calculated.

[0010] There are two problems in simulating the aforementioned process on a computer.

[0011] One is that when the polymers are different, the deviations of all thermal property quantities related to the polymers can be very large. If specific information about the thermal property quantities cannot be obtained, the deviation between the computer-based simulation and the actual process behavior may also become significantly large. Such a deviation reduces the reliability of the simulation for improving the actual process, or may even be incorrect, thus reducing the advantages of the simulation.

[0012] On the one hand, it is very time-consuming and costly to perform accurate and complete characterization of a single material. Not only are there a wide variety of available materials, but also the combinations of materials and additives (such as fibers, carbon black, colorants, aging agents, etc.) and the use of recycled materials are diverse, so accurate knowledge about specific materials is often lacking.

[0013] Among existing methods, there are some that attempt to calculate thermal properties based on molecular structure. However, these methods are not yet accurate enough and require special knowledge about the molecular structure of the target polymer. Abstract

[0014] The present invention aims to overcome the lack of reliable material data for polymer materials by providing a method for creating a dataset of acceptable quality without performing the above-described characterization. Here, "acceptable" means that the results of computer-based simulations of manufacturing processes using the created dataset are accurate enough to obtain benefits in the aforementioned scenarios.

[0015] This goal is achieved by a method of creating thermal property data of a specific material based on known properties of the material (such as polymer type, etc.). The variation in these thermal property values among different polymers can generally be very large, but for specific types of polymers, this variation may be considerably smaller than the variation among all polymers.

[0016] According to a first aspect, a method for improving the simulation of a polymer-based manufacturing process, a method implemented by a computer, is provided. The method includes, for each of a plurality of polymers, obtaining measured parameters for a set of thermophysical properties based on physical measurements; clustering the plurality of polymers into polymer types based on at least one clustering criterion; using a statistical technique to determine representative parameters for each of the set of thermophysical properties for each polymer type; obtaining a selection of a polymer type from a user; generating a dataset based on the selection of the polymer type; and the dataset includes the determined representative parameters for the selected polymer type.

[0017] In some embodiments, at least one of the plurality of polymers can be clustered into multiple polymer types.

[0018] In an example implementation of the first aspect, the method includes using the generated dataset as an input to a simulation of the polymer-based manufacturing process for evaluating the use of the selected polymer type in the polymer-based manufacturing process.

[0019] In an example implementation of the first aspect, the at least one clustering criterion includes at least one of polymer morphology, polymer composition, T90 time, curing system, and polymer hardness.

[0020] In an example implementation of the first aspect, the at least one clustering criterion may be the presence or amount of an additive filler in the polymer. The additive filler may include glass fiber, carbon fiber, calcium carbonate, colorant, carbon black, silica, and / or oil.

[0021] In some embodiments, at least one of the polymer types may be further subdivided based on the atomic or chemical structure of the base polymer.

[0022] In some embodiments, the polymer type may include at least one of an elastomer, a thermosetting material, and a thermoplastic material.

[0023] In some embodiments, the measured parameter is one-dimensional and includes scalar measurements such as viscosity, and the determined representative parameter includes a scalar representative value.

[0024] In some embodiments, the measured parameter is multi-dimensional and may include multiple measurements of thermophysical properties for a particular variable, such as multiple values of thermal conductivity measured at various temperatures. In such embodiments, the determined representative parameter is the corresponding multi-dimensional representative parameter.

[0025] Also, multiple measurements may be interpolated or extrapolated to define, for example, a continuous line or curve (in the case of a two-dimensional parameter) or a surface (in the case of a three-dimensional parameter). In such cases, the determined representative parameter is the corresponding line, curve, or surface.

[0026] In some embodiments, the statistical method includes calculating a median (central value) parameter for each polymer type using the Gaussian distribution of the measured parameters of each set of thermophysical properties in each cluster.

[0027] In embodiments where the measured parameter is a scalar value, the median parameter is the average value of the scalar values, and when the measured parameter is multi-dimensional, the median parameter is also correspondingly multi-dimensional, such as a median curve.

[0028] Depending on the embodiment, the thermophysical set may include thermal conductivity, specific heat, density, viscosity, T90 time, curing kinetics or crystallization kinetics, solidification behavior, and / or mechanical properties (such as rigidity, shrinkage rate, curing shrinkage rate, etc.).

[0029] In an example of the implementation form of the first aspect, the method further includes, for at least one of the thermophysical sets, calculating a statistical deviation (σ) from the representative parameter using the statistical method, and determining a plurality of alternative representative parameters for the thermophysical property based on the representative parameter and the calculated statistical deviation (σ).

[0030] In an example of the implementation form of the first aspect, a plurality of alternative representative values (V var ) are calculated by subtracting or adding a multiple of the statistical deviation (σ) from or to the representative value (V). That is, V var = V + / - (n×σ).

[0031] In an example of the implementation form of the first aspect, the plurality of alternative representative values (V var ) include at least one of the lowest representative value (V min ), the lower representative value (V low ), the average representative value (V avg ), the upper representative value (V high ), and the highest representative value (V max ).

[0032] Depending on the embodiment, the lowest representative value (V min ) is calculated as V min = V - 2σ. Depending on the embodiment, the lower representative value (V low ) is calculated as V low = V - σ. Depending on the embodiment, the average representative value (V avg ) is V avgis given as V. In some embodiments, the upper representative value (V high ) is high calculated as V = V + σ. In some embodiments, the highest representative value (V max ) is max calculated as V = V + 2σ.

[0033] Even in the case of multi-dimensional representative parameters such as a median curve, corresponding methods can be implemented to calculate lower and upper representative parameters. In this method, the median curve is shifted in a positive or negative direction to determine a lower representative curve or an upper representative curve.

[0034] In an example of an implementation form of the first aspect, the step of creating the data set is further based on the selection of a representative parameter from a plurality of alternative representative parameters for at least one thermophysical property.

[0035] In an example of an implementation form of the first aspect, the method further includes defining a set of input parameters based on the selection of the polymer type and optionally the selection of representative parameters for at least one thermophysical property value, and changing the input parameters to create a plurality of data sets.

[0036] In an example of an implementation form of the first aspect, the method further includes performing a simulation of a manufacturing process for each created data set, the manufacturing process having at least one known output parameter measured in physical reality, and the created data set can be used as input thermophysical material data for a given simulation run to obtain the simulated output parameter (simulation output parameter) of the manufacturing process.

[0037] In an example of an implementation form of the first aspect, the method further includes identifying at least one acceptable data set that results in simulation output parameters that meet at least one quality criterion when compared to known output parameters.

[0038] In an example of an implementation form of the first aspect, the identification of the acceptable data set is based on the evaluation of a plurality of quality criteria that define multi-dimensional quality thresholds with respect to known output parameters, and the acceptable data set is selected as the data set that results in simulation output parameters that meet the conditions with respect to the quality thresholds.

[0039] In some embodiments, the condition is that the acceptable data set is selected as the data set where the simulation output parameters do not exceed the quality threshold.

[0040] In some embodiments, when the acceptable data set is based on the evaluation of two quality criteria, the multi-dimensional quality threshold can be represented by a two-dimensional straight line or curve. In other embodiments where the acceptable data set is based on the evaluation of more than two quality criteria, the multi-dimensional quality threshold can be represented by a surface expressed in three or more dimensions.

[0041] In an example of an implementation form of the first aspect, at least one of the thermophysical properties is a thermophysical property of a plurality of parameters (multi-parameter thermophysical property), such as the degree of cure over time at a given temperature, and the method further includes: · obtaining a plurality of measurement values regarding the multi-parameter thermophysical property based on physical measurements of a plurality of polymers; and · determining a master curve or constitutive equation that defines a master curve that best approximates the plurality of measurement values (for the multi-parameter thermophysical property) for a given polymer type; including To create a dataset, representative parameters of the multi-parameter thermophysical properties of the selected polymer type are determined using the master curve or its constitutive equation.

[0042] In an example of an implementation form of the first aspect, the constitutive equation is a combination of a mathematical formula and related parameters, and the method further includes optimizing the related parameters, and the optimizing includes · calculating an error as the difference between a plurality of measured values and the corresponding calculated values given by the constitutive equation; · calculating an error measure; · determining a set of related parameters that minimizes the error measure; and is performed by

[0043] In some embodiments, the error measure is the sum of all squared errors.

[0044] In an example of an implementation form of the first aspect, the method further includes · obtaining input values of the multi-parameter thermophysical properties of the selected polymer type; · selecting a master curve or a constitutive equation that defines a master curve of the multi-parameter thermophysical property values based on the selected polymer type; · adjusting the master curve or the constitutive equation to generate an operated master curve (operated master curve) that conforms to the obtained input values; and includes

[0045] In an example of an implementation form of the first aspect, to create a dataset, representative values of the multi-parameter thermophysical properties of the selected polymer type are determined using the operated master curve or the adjusted constitutive equation.

[0046] In an example of the implementation form of the first aspect, generating the manipulated master curve includes stretching the master curve in a certain direction or applying an offset until the obtained input value conforms to the master curve.

[0047] In another example of the implementation form of the first aspect, generating the manipulated master curve includes inverse fitting the relevant parameters of the corresponding constitutive equation that defines the master curve until the obtained input value conforms to the master curve.

[0048] In an example of the implementation form of the first aspect, the method further includes setting or controlling a manufacturing machine (such as an injection molding machine) involved in the polymer-based manufacturing process using the result of the simulation of the polymer-based manufacturing process.

[0049] In some embodiments, the result of the simulation may be used to control parameters of the manufacturing machine, such as temperature, injection speed, pressure, etc.

[0050] In an example of the implementation form of the first aspect, the method further includes using the result of the simulation of the polymer-based manufacturing process in the process of designing components used in the manufacturing process, such as the mold in the injection molding process or the die in the extrusion molding process.

[0051] In some embodiments, the result of the simulation may be used in the design of the component shape, surface quality, mechanical properties, and chemical properties of the mold or die.

[0052] According to a second aspect, a computer system is provided. The computer system includes an input / output interface; a display; a non-volatile machine-readable storage medium including a computer program product; and at least one processor operable to execute the computer program product, interact with the input / output interface and the display, and perform a process according to any of the forms of the first approach. It comprises.

[0053] According to a third aspect, there is provided a computer program product encoded on a non-volatile machine-readable storage medium and operable to cause a processor to perform a process of a method according to any of the above implementation forms of the first approach.

[0054] These aspects and other aspects will become more apparent from the embodiments described below.

Brief Description of the Drawings

[0055] Hereinafter, various aspects, embodiments, and implementation examples will be described in detail with reference to the exemplary embodiments shown in the drawings.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

[0056] FIG. 1 shows a possible method for clustering polymer 5 into polymer type 6 according to one embodiment of the present disclosure. The polymer material or polymer 5 (simplified and represented by letters A to H in FIG. 1) can be clustered at various levels based on various clustering criteria.

[0057] Such clustering criteria can be based on polymer morphology, which represents the overall form of the polymer structure, such as crystallinity, branching, molecular weight, cross-linking, etc. Small molecules usually have a crystalline solid, which is a highly ordered three-dimensional array of molecules. Solid polymers can be either crystalline or amorphous (a disordered arrangement where the chains are randomly coiled or intertwined). Thermoplastic plastics are usually semi-crystalline, with a combination of crystalline and amorphous regions. Therefore, the properties of thermoplastic plastics are strongly influenced by their morphology.

[0058] The clustering criteria can also be based on polymer composition. Since polymers are not pure materials, many additives can be added to improve the functionality and stability of the polymers. Such additives include, for example, plasticizers, anti-aging stabilizers, flame retardants, colorants, etc. The presence, type, and amount of each additive can be used individually as clustering criteria.

[0059] The clustering criteria can also be based on the T90 time of elastomers and thermosetting resins. The T90 time in the technical field of this application refers to the time required for 90% of the maximum cross-linking between polymer chains to be generated at a predetermined temperature during vulcanization. This value indicates how fast the vulcanization reaction is trying to occur and can thus serve as a clustering criterion.

[0060] The clustering criteria can also be based on the vulcanization system used for elastomers. The curing of elastomers occurs during so-called vulcanization. Single polymer chains bind to other chains to form a network. This chemical reaction is promoted by the vulcanization system. Therefore, different curing systems (e.g., peroxide-based or sulfur-based) can also function as clustering criteria.

[0061] The clustering criterion can also be based on hardness (Shore or Vickers). In this case, the hardness represents the plastic deformation of the part when the indenter is pressed with a standardized pressure and time. In the technical field of the present application, since it is common to estimate the thermophysical properties from this characteristic, such hardness may also serve as a clustering criterion.

[0062] Depending on the embodiment, it is possible to distinguish between elastomers, thermosetting materials, and thermoplastic materials. These polymers 5 can each also be characterized by the chemical structure of the base polymer. For example, elastomers can be subclassified by their atomic structure (heteroatoms). For example, natural rubber, polybutadiene, ethylene propylene rubber, etc. The presence or amount of added fillers can also be used for subclassification. Examples of fillers include glass fibers, carbon fibers, calcium carbonate, colorants, carbon black, silica and / or oil. Such subclassification is referred to in the present disclosure as polymer type 6.

[0063] Knowledge of the specific thermophysical properties 4 of such polymers 5 is essential for performing computer-based simulations of the manufacturing process 2 using such polymers 5 (for example, for mold design including cavity shape and cooling layout, initial setting of machine parameters, control of injection molding machines during manufacturing, or understanding the reasons why the process does not function as expected). Examples of thermophysical property data required for computer-based simulations include thermal conductivity, specific heat, density, viscosity, T90 time, curing rate or crystallization rate, solidification behavior, and / or mechanical properties (such as rigidity, shrinkage rate, curing shrinkage rate, etc.). The collection of all or a subset of the necessary data regarding the thermophysical properties 4 is called a dataset 8.

[0064] For each of the necessary thermophysical properties 4, dedicated experimental setups and appropriate interpretations of experimental data have been developed to collect measured values 3A. The measured values 3A are collected in the dataset 8 stored in the database.

[0065] When starting a simulation, the user usually selects dataset 8 from the database based on certain values, such as the expected or measured values of the starting materials. However, if specific information about the thermal properties 4 of a certain polymer 5 is not provided as input, there is a high possibility that the deviation between the computer-based simulation and the actual process behavior will become very large. Such a deviation will reduce the reliability of the simulation for improving the actual process, or may even lead to incorrect simulation results, thus reducing the advantages of the simulation.

[0066] This problem frequently occurs in polymer-based simulations because it is very time-consuming and costly to obtain an accurate and complete characterization of the specific polymers used. Not only are there a wide variety of available materials, but also the combinations of materials and additives (such as fibers, carbon black, colorants, aging agents, etc.) and the use of recycled materials are diverse, so there is often a lack of accurate knowledge about specific polymers.

[0067] In particular, recycled materials are combinations of materials with different histories, so there are significant differences in their thermomechanical properties. Therefore, since the properties vary from batch to batch, they cannot be accurately measured, or may not be measurable at all.

[0068] The disclosed method is intended to significantly improve the quality of the simulation in scenarios where there is no existing accurate dataset for the polymer 5 intended to be used in the simulated manufacturing process and the user needs to select the dataset that is most likely to match the properties of the polymer 5 used.

[0069] Also, the disclosed method is also intended to be used in DOE (Design of Experiments) to find a stable process for changes in the thermal properties of materials.

[0070] The disclosed method may also be able to improve the initial design stage of a polymer-based manufacturing process where a specific polymer material has not yet been selected for the product to be manufactured. Even at such a stage, it is required that the material meets one or more criteria or that the final product has certain predefined qualities. The method proposed here can be used to check what characteristics the polymer material must have in order to be able to set up the working process. With this knowledge, a specific polymer material can be selected.

[0071] Furthermore, in the case of elastomers, the polymer material is generally made to order for a specific application. In this case, for example, with the help of the method proposed in relation to DOE, the requirements for the material must be specified. Then, a polymer material that meets the specifications can be manufactured.

[0072] Figure 2 shows in a graph the large variation in measured values 3 of a certain thermophysical property 4 for a plurality of polymers 5. As also shown in Figure 2, among these polymers 5, within a selected cluster of the same polymer type 6, these variations between the measured values 3 are much smaller. This gave the inventor the idea of an invention that uses a technique based on clustering polymers 5 into polymer types 6 in order to generate a data set 8 for a computer-based simulation 1, as will be explained below.

[0073] Figure 3 is a flowchart showing a method for improving a simulation 11 of a polymer-based manufacturing process according to an embodiment of the present disclosure.

[0074] In a first step 101, for each of a plurality of known polymers 5, measured parameters 3 for a thermophysical property 4 are obtained using known test methods.

[0075] In the next step 102, as shown in FIG. 1, a plurality of polymers 5 are clustered into polymer types 6 based on at least one clustering criterion. This clustering criterion may be, as described above, a molecular structure, a filler component, or other criteria.

[0076] In the next step 103, for all polymer types 6, a statistical method is applied to identify one representative parameter 7 for each thermal property 4. This may be performed, as shown in FIG. 5 and described in more detail below, by calculating the mean value 9 of the Gaussian distribution 10 of the measured values 3A of a specific thermal property 4.

[0077] Once the polymer types 6 are identified and the above representative parameters 7 are calculated, in the next step 104, a dataset 8 can be generated based on the selection of the polymer types 6. This dataset 8 includes the determined representative parameters 7 of the selected polymer types 6.

[0078] In the last step 105, the created dataset 8 can be used as an input for the simulation 1 of the polymer-based manufacturing process 2.

[0079] FIG. 4 shows a method of clustering some polymers 5 into some polymer types 6 according to an embodiment of the present disclosure. As shown in this figure, a certain polymer 5 can be clustered into only one polymer type 6 based on the clustering criterion as described above, while other polymers 5 (shown as polymer F in the figure) can be clustered into multiple polymer types 6.

[0080] FIGS. 5 and 6 show a statistical method for determining the representative value 7 of the thermal property 4 for a specific polymer type 6, which is used to generate an extended dataset 8 for the simulation 1 of the polymer-based manufacturing process 2 according to an embodiment of the present disclosure.

[0081] In the embodiment shown in FIG. 5, the statistical method includes calculating the mean value 9 of the Gaussian distribution 10 of the scalar values 3A measured for a specific thermophysical property 4. The mean value 9 can be used as the only representative parameter 7. Alternatively, the mean value 9 can be selected and used as one of the representative parameters 7 when generating the data set 8 as shown in the flowchart of FIG. 6, and the results of the steps shown in FIG. 6 are depicted in the diagram of FIG. 5. Particularly in the latter case, the statistical method further includes the next step 106 of calculating the statistical deviation 11 from the representative parameter 7 for the selected thermophysical property 4 after determining the representative parameter 7.

[0082] In the next step 107, based on the representative parameter 7 and the calculated statistical deviation 11, a plurality of alternative representative parameters 71, 72, 73, 74, 75 for the selected thermophysical property 4 can be calculated. The plurality of alternative representative parameters 71, 72, 73, 74, 75 can be calculated by subtracting multiples of the statistical deviation 11 from the representative parameter 7 or adding multiples of the statistical deviation 11 to the representative parameter 7. As a result, as shown in FIG. 5, five alternative representative parameters such as the lowest representative parameter (LOWEST) 71, the lower representative parameter (LOWER) 72, the average representative parameter (AVG) 73, the higher representative parameter (HIGHER) 74, and the highest representative parameter (HIGHEST) 73 can be selected.

[0083] According to an embodiment in which the representative parameter is a scalar value, the alternative representative values can be calculated as follows: · The lowest representative value (V min ) is calculated as V min = V - 2σ; · The lower representative value (V low ) is calculated as V low = V - σ; · The average representative value (V avg ) is given as V avg = V; · The higher representative value (V high ) is calculated as V high = V + σ. · The highest representative value (V max ) is calculated as V max = V + 2σ.

[0084] As shown in Figure 3 and as described above, step 104 of generating dataset 8 is based on one or more selections among alternative representative parameters 71, 72, 73, 74, 75 for at least one thermophysical property 4.

[0085] This selection is further shown in Figure 7. Here, the selection of one polymer type 6 from among a plurality of polymer types 6 and the selection of representative parameters 7 for various thermophysical properties 4 are provided by the user via a graphical user interface (GUI) 21 as input parameter set 12. This input parameter set 12 is an input parameter set for generating dataset 8 used as input for simulation 1 of polymer-based manufacturing process 2.

[0086] This embodiment is particularly useful when the trend of thermophysical property 4 is known for a specific dataset 8 compared to other types of polymers 6. For example, when it is known that the conductivity is particularly low compared to other polymers 5 of the same polymer type 6 using a Gaussian distribution, the value obtained by subtracting the statistical deviation 11 of conductivity from the average value 9 can be used. And by changing which of the plurality of alternative representative parameters 7 to select, a plurality of datasets 8 can be generated. And using these various datasets 8, a manufacturing process that is already well understood can be simulated. And as shown in detail in Figure 8, it is possible to identify which dataset 8 produced the best results.

[0087] Figure 8 is a flow diagram showing a method of determining an acceptable dataset 15 by performing a plurality of simulations 1 based on a set of input parameters 12 that can be provided via a GUI as shown in Figure 7 and varying them.

[0088] In an initial step 108, a set of input parameters 12 is defined to enable generation of a plurality of data sets 8. The input parameters 12 are based on the selection of a polymer type 6 and the selection of representative parameters 7 for at least one thermophysical property 4. The number of these alternative parameters can be varied for each thermophysical property 4 based on known trends regarding the specific thermophysical property 4. Depending on the thermophysical property 4, only one representative parameter 7 may be provided, while for other thermophysical properties 4, a lower parameter 72 and an upper parameter 74 may be provided as alternative options, and an average parameter 73 may also be provided as an alternative option in the same way. For other thermophysical properties 4, a minimum parameter 71 and a maximum parameter 75 may further be provided as alternative options.

[0089] In the next step 109, a plurality of data sets 8 are generated by varying these input parameters 12.

[0090] In the next step 110, for each generated data set 8, a simulation 1 of the manufacturing process 2 is executed. Each created data set 8 is used as input thermophysical material data for a specific simulation 1, resulting in simulation output parameters 14 of the manufacturing process 2.

[0091] An important condition for this method is that the manufacturing process 2 has at least one known output parameter 13 measured in physical reality.

[0092] In this way, in a final step 111, based on the simulation 1 resulting in simulation output parameters 14 that meet quality criteria 16 when compared with the known output parameter 13, at least one acceptable data set 15 can be identified. This step may be based on only one quality criterion 16, such as the deviation of the simulation output parameter 14 from the known output parameter 13 being below a predetermined threshold, or as further shown in FIG. 9, it may be determined based on the evaluation of a plurality of quality criteria 16.

[0093] FIG. 9 is a diagram showing the selection of an acceptable data set 15 based on a plurality of quality criteria 16 according to an embodiment of the present disclosure. In the illustrated embodiment, identifying the acceptable data set 15 is based on a plurality of quality criteria 16 that define a multi-dimensional quality threshold 20 with respect to known output parameters 13, and the acceptable data set 15 is selected as the data set 8 that results in simulation output parameters 14 that meet the criteria with respect to the quality threshold 20.

[0094] In some embodiments, the condition is that the acceptable data set 15 is selected as the data set 8 in which the simulation output parameters 14 do not exceed the quality threshold 20.

[0095] As shown in FIG. 9, in embodiments where the acceptable data set 15 is based on the evaluation of two quality criteria 16, the multi-dimensional quality threshold 20 is represented by a two-dimensional line or curve. Each of the simulation output parameters 14 that meet the criteria of the quality threshold may represent a plurality of acceptable data sets 15, and there is no distinction in the quality of these acceptable data sets 15. The sum of the acceptable data sets 15 represents a Pareto set. These are indicated by the simulation output parameters 14 located on the curve 20 in FIG. 9.

[0096] In other embodiments where the acceptable data set 15 is based on the evaluation of more than two quality criteria 16, the multi-dimensional quality threshold 20 can also be represented by a surface represented in three or more dimensions.

[0097] Also, the thermophysical property 4 used to generate the data set 8 may be a multi-parameter thermophysical property 4A, such as the degree of cure over time at a specific temperature. FIGS. 10-13 show the steps of the method in such a case. In this case, at least one multi-parameter thermophysical property 4A in a set of thermophysical properties 4 is involved.

[0098] Figure 10 is a diagram showing the steps for determining a master curve 18 of multi-parameter thermophysical properties 4A for a given polymer type 6 according to an embodiment of the present disclosure.

[0099] In this case, the method includes a first step 101A of obtaining a plurality of measured values 3A for the multi-parameter thermophysical properties 4A, that is, the degree of cure over time at a specific temperature, based on physical measurements of a plurality of polymers 5. In the example of Figure 10, the measured values 3A are obtained for the degree of cure from 0 to 1 (or 0 to 100%) over a time period of 0 to 1000 seconds at temperatures of 160 °C, 175 °C, and 190 °C.

[0100] Based on these obtained measured values 3A, in the next step 112, a master curve 18 (or a constitutive equation defining the master curve 18) for the multi-parameter thermophysical properties 4A is determined based on the best approximation of the plurality of measured values 3A.

[0101] And by using the master curve 18 or its constitutive equation to determine the representative parameter 7 of this multi-parameter thermophysical property 4A, a data set 8 of the selected polymer type 6 can be created.

[0102] This constitutive equation is a combination of a mathematical formula and related parameters, and these related parameters can be further optimized. In such a case, the method further includes calculating an error as the difference between the plurality of measured values 3A and the corresponding calculated values given by the constitutive equation.

[0103] Next, an error measure such as the sum of all squared errors is defined, and a set of optimized related parameters is calculated as the related parameters that result in the minimum value of this error measure.

[0104] If a specific input value 17 for the multi-parameter thermophysical property 4A is known, the master curve 18 or its constitutive equation can be adjusted to create a data set that better matches the specific input value 17.

[0105] Figures 11 and 12 are consecutive figures, and Figure 13 is a flowchart showing the steps of operating a master curve 18 to conform to a given input value 17 according to an embodiment of the present disclosure.

[0106] As shown in Figure 11 and the flowchart (Figure 13), in the first step 113, an input value 17 of the multi-parameter thermophysical property 4A of the selected polymer type 6 is obtained. As an example, this input value 17 defines the curing rate, as shown in the above example of Figure 10. In this particular example, the input value 17 is a degree of cure of 90% or 0.9 at a specific temperature of 160°C over a time of 750 seconds.

[0107] In the next step 114, a master curve 18 (or the constitutive equation defining the master curve 18) of the multi-parameter thermophysical property 4A is selected based on the selected polymer type 6. In this case, it is the master curve 18 at a specific temperature of 160°C. As can be seen from Figure 11, the input value 17 does not match this master curve 18.

[0108] Figure 12 shows the next steps of operating the master curve 18 to conform to the input value 17. This is also explained in the flowchart of Figure 13.

[0109] In the next step 115, the master curve 18 (or the constitutive equation) is adjusted to generate an operated master curve 19 (operated master curve 19) that conforms to the obtained input value 17.

[0110] The generation of the operated master curve 19 can include various techniques. For example, simply stretching the master curve 18 in a certain direction, applying an offset, or inverse fitting the relevant parameters of the constitutive equation until the obtained input value 17 fits on the master curve 19.

[0111] As shown in FIG. 13, once the processed master curve 19 is generated, it can be used in the generation of the data set 8, and at this time, the representative parameter 7 of the multi-parameter thermophysical property 4A of the selected polymer type 6 is determined using the processed master curve 19 (or its adjusted constitutive equation).

[0112] FIG. 14 is a flowchart showing a method for setting or controlling a manufacturing machine 22 used in a polymer-based manufacturing process 2 or a method for designing components used in a polymer-based manufacturing process 2 according to an embodiment of the present disclosure.

[0113] As shown in this flowchart, the above-described method further includes a step 112 of setting or controlling a manufacturing machine 22 (such as an injection molding machine) involved in the polymer-based manufacturing process 2 using the simulation output parameter 14 of the simulation 1 of the polymer-based manufacturing process 2, following a step 104 of creating a data set and a step 105 of executing a simulation 105 of the polymer-based manufacturing process 2.

[0114] According to some embodiments, the simulation result 14 may be used for parameter control of the manufacturing machine 22, such as temperature, injection speed, pressure, etc., as shown in FIG. 15.

[0115] As further shown in the flowchart of FIG. 14, the above-described method further includes a step 113 of designing components used in the manufacturing process 2, such as a mold for an injection molding process or a die for an extrusion process, using the simulation output parameter 14 of the simulation 1 of the polymer-based manufacturing process 2, following a step 104 of creating a data set and a step 105 of executing a simulation 1 of the polymer-based manufacturing process 2.

[0116] According to some embodiments, the simulation result 14 may be used for designing the component shape, surface quality, mechanical properties, and chemical properties of the mold or die.

[0117] FIG. 15 is a schematic block diagram showing an example of the hardware configuration of a computer-based system 28 for performing a simulation 1 of a polymer-based manufacturing process 2 and, in some embodiments, setting or controlling a manufacturing machine 22 involved in the polymer-based manufacturing process 2, in accordance with the present disclosure.

[0118] The computer system 28 may include one or more processors (CPUs) 32 configured to execute instructions that cause the computer system to perform a method according to any of the above-described embodiments.

[0119] The computer system 28 may include a computer-readable storage medium 31 configured to store software-based instructions as part of a program product executed by the CPU 32.

[0120] The computer system 28 may include a memory 33 configured to (temporarily) store data of applications and processes.

[0121] The computer system 28 may include an input / output interface 29 for interaction between the system and a user 38. The input / output interface 29 may be connected to, or may include, an input interface (such as a keyboard and / or a mouse) for receiving input from the user 38. The computer system may also further include an output device (such as an electronic display 30) for communicating information to the user 38 through a graphical user interface (GUI) 21 such as the GUI 21 of FIG. 7.

[0122] The computer system 28 may further include a communication interface 34 for communicating directly or indirectly with an external device such as a remote client 37 via a computer network 35.

[0123] The foregoing hardware elements within the computer system may be connected via an internal bus configured to process data communication and processing operations.

[0124] The computer-based system 28 may further be connected to a database 36. The database 36 is configured to store data used as input for the methods described above (such as measurement parameters 3 of thermophysical properties 4 for various polymers 5, or a set of measured values 3A of multi-parameter thermophysical properties 4A). The type of connection between the system 28 and the database 36 can be direct or indirect. The computer system 28 and the database 36 may both be included in the same physical device and connected via an internal bus, or may be part of physically different devices and connected directly via a communication interface 34 or indirectly via a computer network 35.

[0125] As described above, the computer-based system 28 may be connected to the manufacturing machine 22 of the polymer-based manufacturing process 2, and the simulation results 14 generated by the computer-based system 28 may be used to control parameters of the manufacturing machine 22 such as temperature, injection speed, or pressure. Also, as described above in connection with FIG. 8, parameters may be obtained from the manufacturing machine 22 to improve the simulation 1 and / or the data set 8 used in the simulation 1.

[0126] Various aspects and implementations of the invention have been described with several examples. However, upon examining the specification, drawings, and claims of this application, those skilled in the art will understand that there are many variations in addition to the described examples when implementing the invention described in the claims and will also be able to embody them. The terms "comprising", "having", "including" used in the claims do not exclude the existence of elements or steps not described. Even if it is not explicitly stated that the number of elements described in the claims is plural, the existence of a plurality of such elements is not excluded. The functions of several elements described in the claims may be performed by a single processor or other units. Even if several matters are described in separate dependent claims, this does not exclude implementing them in combination, and benefits can be obtained by implementing them in combination. A computer program may be stored or distributed on a suitable medium such as an optical storage medium or a solid medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless electrical communication systems.

[0127] The reference signs used in the claims shall not be construed as limiting the scope of the invention.

Claims

1. 1. A computer-implemented method for improving a polymer-based manufacturing process, comprising: obtaining, for each of the plurality of polymers, measured parameters for a set of thermal properties based on physical measurements; clustering the plurality of polymers into polymer types based on at least one clustering criterion; applying a statistical method to the measured parameters for a particular thermal property of a particular polymer type to determine a representative parameter for each of the set of thermal properties for each polymer type; obtaining a selection of a polymer type from a user; generating a data set based on the selection of polymer types; wherein the dataset includes the determined representative parameters for the selected polymer type.

2. 2. The method of claim 1, further comprising using the generated data set as an input for a simulation of a polymer-based manufacturing process to evaluate use of the selected polymer type in the polymer-based manufacturing process.

3. The method of claim 1 , wherein the at least one clustering criterion comprises at least one of polymer morphology, polymer composition, T90 time, cure system, and polymer hardness.

4. The method of claim 1 , wherein at least one of the polymer types is further subdivided based on the atomic or chemical structure of a base polymer.

5. The method of claim 1 , wherein the statistical approach comprises calculating, for each polymer type, a median parameter using a Gaussian distribution of the measured parameters for each of the set of thermal properties in each cluster.

6. 10. The method of claim 1, further comprising: calculating, for at least one of the set of thermal properties, a statistical deviation from the representative parameter using a statistical method; determining a plurality of surrogate representative parameters for the at least one thermophysical property based on the representative parameters and the calculated statistical deviation; Including, The step of creating the data set further comprises: selecting a representative parameter from the plurality of alternative representative parameters for at least one thermophysical property; method.

7. 10. The method of claim 1, further comprising: defining a set of input parameters based on a selection of a polymer type and, optionally, also based on a selection of a parameter representative of at least one thermophysical property; Varying the input parameters to generate a plurality of data sets; performing a simulation of a manufacturing process for each of the created data sets; the manufacturing process having at least one known output parameter measured in physical reality, the created data set being used as input thermophysical material data for a given simulation run to obtain simulation output parameters of the manufacturing process; The method further includes identifying at least one acceptable data set that, when compared to the known output parameters, results in simulation output parameters that meet a quality criterion.

8. 8. The method of claim 7, wherein the step of identifying at least one acceptable data set is based on a number of quality criteria defining multi-dimensional quality thresholds with respect to the known output parameters, the acceptable data set being selected as a data set resulting in simulation output parameters that satisfy conditions with respect to the quality thresholds.

9. 10. The method of claim 1, wherein at least one of the set of thermal properties is a multi-parameter thermal property, such as degree of cure over time at a particular temperature, the method further comprising: obtaining a plurality of measurements of the multi-parameter thermal properties based on a plurality of physical measurements of the polymer; determining a master curve or constitutive equation; wherein the master curve or constitutive equation defines, for a given polymer type, a master curve of the multi-parameter thermophysical properties that best approximates a plurality of measured values; wherein the representative parameters of the multi-parameter thermophysical properties of a selected polymer type are determined using the master curve or constitutive equation to generate the data set. method.

10. The constitutive equation is a combination of a mathematical expression and relevant parameters, and the method further includes optimizing the relevant parameters, the optimizing comprising: calculating an error as the difference between a plurality of measurements and corresponding calculated values ​​given by said constitutive equation; Calculating an error measure such as the sum of all squared errors; determining a set of relevant parameters that minimizes said error measure; The method of claim 9, wherein the method is carried out by

11. obtaining input values ​​for multi-parameter thermal properties of a selected polymer type; selecting a master curve or constitutive equation that defines a master curve for the multi-parameter thermal properties based on the selected polymer type; adjusting the master curve or the constitutive equation to generate a manipulated master curve that fits the obtained input values; 10. The method of claim 9, further comprising: determining the representative values ​​of the multi-parameter thermophysical properties of a selected polymer type using the manipulated master curve or adjusted constitutive equation to create the data set.

12. The method of any of claims 1 to 11, further comprising using simulation output parameters of the simulation to configure or control a manufacturing machine, such as an injection molding machine, involved in a polymer-based manufacturing process.

13. The method of claim 1 , further comprising using simulation output parameters of the simulation to design a part to be used in a manufacturing process, such as a mould for an injection moulding process or a die for an extrusion moulding process.

14. An input / output interface; A display; a non-volatile machine-readable storage medium containing a computer program product; At least one processor operable to execute the computer program product, to interact with the input / output interface and the display, and to perform the method according to any of claims 1 to 11; A computer system comprising:

15. A computer program product encoded on a non-volatile machine-readable storage medium and operable to cause a processor to perform operations according to the method of any of claims 1 to 11.

Images