Polyolefin performance parameter prediction method and device and storage medium

Abstract

The invention provides a polyolefin performance parameter prediction method, a polyolefin performance parameter prediction device and a storage medium. The polyolefin performance parameter prediction method comprises the following steps: obtaining a to-be-predicted polyolefin structure parameter; according to the structural parameters, performing dynamic simulation on polyolefin molecules under various process conditions to generate microstructure data of the polyolefin molecules; and predicting the performance parameters of the polyolefin according to the microstructure data. The performance parameters comprise a plurality of mechanical property indexes of the polyolefin. According to the method, the microstructure data of polyolefin molecules can be generated through molecular dynamics simulation, and then the polyolefin performance parameters are predicted by constructing the prediction model representing the mapping relation between the polyolefin molecule microstructure data and the macroscopic mechanical properties, so that the accuracy and adaptability of the polyolefin performance parameter prediction result are improved, and the prediction accuracy and adaptability of the polyolefin performance parameter prediction result are improved. Therefore, the development efficiency and the performance optimization level of the polyolefin material are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of polypropylene product quality prediction, and particularly to a method for predicting polyolefin performance parameters, a device for predicting polyolefin performance parameters, and a computer-readable storage medium. Background Art

[0002] Polyolefin is one of the most widely used polymer materials, with excellent mechanical properties and processing adaptability. It is widely used in fields such as packaging, automotive, and construction. Its mechanical properties directly affect the performance and market competitiveness of the material in industrial applications. However, in the prior art, the methods for predicting the mechanical property parameters of polyolefin mainly rely on experimental tests and physical property models based on empirical formulas. Due to the complex, costly, and time-consuming experimental test process, it is difficult to meet the requirements of efficient screening for the development of new materials. At the same time, the prediction method based on the empirical model relies on idealized assumptions and cannot accurately describe the non-linear influence of complex molecular structures on mechanical properties, resulting in limited prediction accuracy.

[0003] In order to overcome the above-mentioned defects existing in the prior art, there is an urgent need in the art for an improved method for predicting polyolefin performance parameters, which is used to improve the accuracy and adaptability of the prediction results of polyolefin performance parameters, thereby improving the development efficiency and performance optimization level of polyolefin materials. Summary of the Invention

[0004] A brief overview of one or more aspects is given below to provide a basic understanding of these aspects. This overview is not an exhaustive survey of all contemplated aspects, and is neither intended to identify key or decisive elements of all aspects nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to a more detailed description to follow.

[0005] In order to overcome the above-mentioned defects existing in the prior art, the present invention provides a method for predicting polyolefin performance parameters, a device for predicting polyolefin performance parameters, and a computer-readable storage medium. First, through molecular dynamics simulation, microscopic structure data of polyolefin molecules can be generated, and then by constructing a prediction model representing the mapping relationship between the microscopic data of polyolefin molecules and macroscopic mechanical properties, the polyolefin performance parameters can be predicted, which is used to improve the accuracy and adaptability of the prediction results of polyolefin performance parameters, thereby improving the development efficiency and performance optimization level of polyolefin materials.

[0006] Specifically, the method for predicting the performance parameters of polyolefins provided by the first aspect of the present invention includes the following steps: obtaining the structural parameters of the polyolefin to be predicted; performing kinetic simulations on the polyolefin molecules under various process conditions according to the structural parameters to generate the microscopic structure data of the polyolefin molecules; and predicting the performance parameters of the polyolefin according to the microscopic structure data. The performance parameters include various mechanical property indexes of the polyolefin.

[0007] Further, in some embodiments of the present invention, the step of performing kinetic simulations on the polyolefin molecules under various process conditions according to the structural parameters to generate the microscopic structure data of the polyolefin molecules includes: defining the key parameters in the kinetic simulation process according to the process conditions. The key parameters at least include temperature, pressure, simulation time step, boundary conditions, initial molecular configuration, and the type of force field selected; adjusting the atomic positions within the polyolefin molecules according to the key parameters to minimize the energy of the polyolefin molecular system to eliminate the overlapping and high-potential energy regions in the polyolefin molecular system; and performing kinetic simulations on the polyolefin molecules in the equilibrium stage and the non-equilibrium stage, and collecting the microscopic structure data of the polyolefin molecules during the kinetic simulation process.

[0008] Further, in some embodiments of the present invention, the step of performing kinetic simulations on the polyolefin molecules in the equilibrium stage and the non-equilibrium stage includes: adjusting the temperature in the kinetic simulation process and determining whether the polyolefin molecules reach thermodynamic equilibrium; and in response to the polyolefin molecules not reaching thermodynamic equilibrium, changing the initial structure of the polyolefin molecules and re-performing the kinetic simulation.

[0009] Further, in some embodiments of the present invention, the step of performing kinetic simulations on the polyolefin molecules in the equilibrium stage and the non-equilibrium stage further includes: in response to the polyolefin molecules reaching thermodynamic equilibrium, applying an external perturbation to the polyolefin molecules and collecting the molecular structure and dynamic behavior information of the polyolefin molecules to generate the microscopic data structure. The external perturbation at least includes tensile behavior and compressive stress.

[0010] Further, in some embodiments of the present invention, the microscopic structure data at least includes the bond lengths, bond angles, intramolecular interactions, and intermolecular interactions at the atomic scale of the polyolefin molecules.

[0011] Further, in some embodiments of the present invention, the step of predicting the performance parameters of the polyolefin based on the microstructure data includes: performing feature processing on the microstructure data to obtain a representative structural descriptor of the polyolefin molecule. The representative structural descriptor at least includes the radial distribution of the polyolefin molecule, the end-to-end distance of the molecular chain, and the number of hydrogen bonds; and inputting the representative structural descriptor into a pre-trained prediction model to predict the performance parameters of the polyolefin.

[0012] Further, in some embodiments of the present invention, the step of performing feature processing on the microstructure data to obtain a representative structural descriptor of the polyolefin molecule includes: using a feature screening method to analyze the microstructure data to generate a molecular structural descriptor representing the relationship between the performance and microstructure of the polyolefin; and performing normalization processing on the molecular structural descriptor to generate the representative structural descriptor.

[0013] Further, in some embodiments of the present invention, the step of training the prediction model includes: obtaining multiple sets of historical sample data of the structural parameters and performance parameters of the polyolefin under the various process conditions; constructing a prediction model to be trained and initializing the prediction model; and training the prediction model according to the historical sample data.

[0014] Further, in some embodiments of the present invention, the prediction model includes an aggregation layer, an attention layer, a normalization layer, a splicing layer, and a fully connected layer. The aggregation layer is used to convert the representative structural descriptor into a continuous high-dimensional embedding vector. The high-dimensional embedding vector includes a plurality of nodes and edges. The nodes represent atoms within the polyolefin molecule. The edges represent chemical bonds between atoms within the polyolefin molecule. The attention layer is used to perform information aggregation on the high-dimensional embedding vector according to the neighborhood information of each node and edge to update the features of the nodes and edges in the high-dimensional embedding vector. The neighborhood information at least includes features such as the type of the node, the type of the edge, the distance and angle between the nodes within the neighborhood. The normalization layer is used to normalize the features in the high-dimensional embedding vector to ensure the uniformity of the data distribution and the stability of the prediction model. The splicing layer is used to splice the features of multiple normalization layers to generate high-dimensional features. The fully connected layer is used to map the relationship between the high-dimensional features and the property parameters of the polyolefin.

[0015] Further, in some embodiments of the present invention, the step of training the prediction model according to the historical sample data includes: dividing the historical sample data into a training set and a test set; training the prediction model using the training set, and testing the prediction model using the test set to determine the model loss value of the prediction model; in response to the model loss value being greater than a preset loss threshold, redefining the hyperparameters of the prediction model, and re-training the prediction model according to the redefined hyperparameters; and in response to the model loss value being less than or equal to the loss threshold, determining that the training of the prediction model is completed.

[0016] In addition, the above-mentioned prediction device for polyolefin performance parameters provided by the second aspect of the present invention includes a memory and a processor. A computer instruction is stored on the memory. The processor is connected to the memory and is used to execute the computer instruction stored on the memory to implement the prediction method for polyolefin performance parameters provided by the first aspect of the present invention.

[0017] In addition, the above-mentioned computer-readable storage medium provided by the third aspect of the present invention stores a computer instruction, and is characterized in that when the computer instruction is executed by a processor, it implements the prediction method for polyolefin performance parameters provided by the first aspect of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] After reading the detailed description of the embodiments of the present disclosure in conjunction with the following drawings, the above features and advantages of the present invention can be better understood. In the drawings, the components are not necessarily drawn to scale, and components with similar relevant characteristics or features may have the same or similar reference numerals.

[0019] Figure 1 A flowchart showing the prediction method for polyolefin performance parameters provided by some embodiments of the present invention is shown.

[0020] Figure 2 A flowchart showing the kinetic simulation provided by some embodiments of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The following specific embodiments illustrate the implementation manners of the present invention, and those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Although the description of the present invention will be introduced in combination with preferred embodiments, this does not mean that the features of this invention are limited to this implementation manner. On the contrary, the purpose of introducing the invention in combination with the implementation manner is to cover other alternatives or modifications that may extend based on the claims of the present invention. In order to provide a deep understanding of the present invention, many specific details will be included in the following description. The present invention can also be implemented without using these details. In addition, in order to avoid confusing or obscuring the key points of the present invention, some specific details will be omitted in the description.

[0022] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "mounted", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0023] In addition, the "upper", "lower", "left", "right", "top", "bottom", "horizontal", and "vertical" used in the following description should be understood as the orientations shown in this section and the relevant drawings. This relative term is only for convenience of description and does not mean that the device described needs to be manufactured or operated in a specific orientation, so it should not be understood as a limitation to the present invention.

[0024] It can be understood that although terms such as "first", "second", and "third" can be used herein to describe various components, regions, layers, and / or parts, these components, regions, layers, and / or parts should not be limited by these terms, and these terms are only used to distinguish different components, regions, layers, and / or parts. Therefore, the first component, region, layer, and / or part discussed below can be referred to as the second component, region, layer, and / or part without departing from some embodiments of the present invention.

[0025] As described above, polyolefin is one of the most widely used polymer materials, with excellent mechanical properties and processing adaptability. It is widely used in fields such as packaging, automotive, and construction. Its mechanical properties directly affect the performance and market competitiveness of the material in industrial applications. However, in the prior art, the prediction methods for the mechanical property parameters of polyolefins mainly rely on experimental tests and physical property models based on empirical formulas. Since the experimental test process is complex, costly, and time-consuming, it is difficult to meet the requirements of efficient screening for the development of new materials. At the same time, the prediction method based on the empirical model relies on idealized assumptions and cannot accurately describe the non-linear influence of complex molecular structures on mechanical properties, resulting in limited prediction accuracy.

[0026] To overcome the above-mentioned defects existing in the prior art, the present invention provides a method for predicting polyolefin property parameters, a device for predicting polyolefin property parameters, and a computer-readable storage medium. First, through molecular dynamics simulation, microscopic structure data of polyolefin molecules can be generated, and then by constructing a prediction model representing the mapping relationship between the microscopic data of polyolefin molecules and macroscopic mechanical properties, the polyolefin property parameters can be predicted, which is used to improve the accuracy and adaptability of the prediction results of polyolefin property parameters, thereby improving the development efficiency and performance optimization level of polyolefin materials.

[0027] In some non-limiting embodiments, the above-mentioned device for predicting polyolefin property parameters provided in the second aspect of the present invention includes a memory and a processor. Here, the memory includes but is not limited to the computer-readable storage medium provided in the foregoing third aspect, on which computer instructions are stored. The processor is connected to the memory and is configured to execute the computer instructions stored on the memory to implement the method for predicting polyolefin property parameters provided in the first aspect of the present invention.

[0028] The working principle of the above-mentioned device for predicting polyolefin property parameters will be described below in conjunction with some embodiments of the method for predicting polyolefin property parameters. Those skilled in the art can understand that these embodiments of the prediction methods are only some non-limiting implementation manners provided by the present invention, aiming to clearly show the main concept of the present invention and provide some specific solutions convenient for the public to implement, rather than limiting all the functions or all the working modes of the prediction device. Similarly, the device for predicting polyolefin property parameters is also a non-limiting implementation manner provided by the present invention, and does not limit the execution subject and execution order of each step in these methods for predicting polyolefin property parameters.

[0029] Specifically, please refer to Figure 1 。 Figure 1 Fig. shows a schematic flow chart of the method for predicting polyolefin property parameters provided by some embodiments of the present invention.

[0030] As Figure 1As shown, the processor can first obtain the structural parameters of the polyolefin to be predicted. Here, the above-mentioned structural parameters include the multi-scale characteristics of the polyolefin, specifically including atomic-level characteristics and segment-level characteristics. The processor can generate a polyolefin molecular model with a diverse segment structure based on the key structural parameters of the polyolefin polymer.

[0031] After that, the processor can perform kinetic simulations on the polyolefin molecules under various process conditions according to the above-mentioned structural parameters to generate the microscopic structure data of the polyolefin molecules.

[0032] Please refer further to Figure 2 。 Figure 2 FIG. shows a schematic flow diagram of kinetic simulation provided according to some embodiments of the present invention.

[0033] Specifically, as Figure 2 shown, the processor can define the key parameters in the kinetic simulation process according to the process conditions. Here, the above-mentioned key parameters at least include temperature, pressure, simulation time step, boundary conditions, initial molecular configuration, and the type of force field selected. In this way, the above-mentioned prediction method for polyolefin performance parameters provided by the first aspect of the present invention can cover various process conditions such as different temperature ranges, pressure conditions, shear rates, stress fields existing during processing, and different compositions of the polyolefin system, and enhance the generality and adaptability of the prediction method through molecular dynamics simulations that comprehensively cover the above-mentioned process conditions.

[0034] After that, the processor can adjust the atomic positions within the polyolefin molecules according to the above-mentioned key parameters to minimize the energy of the polyolefin molecular system, so as to eliminate the overlapping and high-potential energy regions in the polyolefin molecular system.

[0035] Specifically, in some embodiments, during the process of minimizing the energy of the polyolefin molecular system, the processor can first calculate the total energy of the polyolefin molecular system according to the force field. After that, the processor can adjust the atomic positions or bond angles within the polyolefin molecules to lower the total energy until the convergence criterion is reached. The result of energy minimization is a thermodynamically stable initial molecular configuration, providing a reliable starting point for kinetic simulation to ensure model stability.

[0036] Then, the processor can perform kinetic simulations on the polyolefin molecules in the equilibrium stage and non-equilibrium stage, and collect the microscopic structure data of the polyolefin molecules during the kinetic simulation process.

[0037] Furthermore, as Figure 2 shown, during the process of performing kinetic simulation on the polyolefin molecules, the processor can adjust the temperature in the kinetic simulation process and determine whether the polyolefin molecules reach thermodynamic equilibrium.

[0038] Subsequently, in response to the polyolefin molecules not reaching thermodynamic equilibrium, the processor can change the initial structure of the polyolefin molecules in the polyolefin molecule model and re - perform kinetic simulations.

[0039] Alternatively, in response to the polyolefin molecules reaching thermodynamic equilibrium, the processor can perform non - equilibrium molecular dynamics simulations on the polyolefin molecular system. Specifically, the processor can apply external perturbations to the polyolefin molecules and collect information on the molecular structure and dynamic behavior of the polyolefin molecules to generate a microscopic data structure. Here, the external perturbations include at least tensile behavior and compressive stress.

[0040] Furthermore, in some embodiments, the obtained microscopic structure data at least includes bond lengths, bond angles, intramolecular interactions, and intermolecular interactions at the atomic scale of the polyolefin molecules.

[0041] In addition, in some preferred embodiments, the processor can also utilize high - throughput simulation techniques based on a distributed computing architecture or parallel computing technology to achieve parallel execution of tasks by allocating multiple simulation tasks to different nodes in a computing cluster.

[0042] Specifically, the processor can batch - generate input files, schedule molecular dynamics simulation tasks to multiple computing nodes, monitor the task running status, and automatically collect and organize output data.

[0043] In this way, the above - mentioned method for predicting polyolefin performance parameters provided by the first aspect of the present invention can parallel - simulate multiple molecular models under different conditions, thereby quickly generating a large - scale microscopic structure data set. Finally, an overall evaluation of the simulation results is carried out, and the reliability of the results is verified through multiple indicators, providing high - quality data support for further performance modeling.

[0044] After that, the processor can predict the performance parameters of the polyolefin based on the microscopic structure data. Here, the performance parameters include various mechanical property indicators of the polyolefin.

[0045] Specifically, the processor can first perform feature processing on the microscopic structure data to obtain a representative structural descriptor of the polyolefin molecules. Here, the representative structural descriptor at least includes the radial distribution of the polyolefin molecules, the end - to - end distance of the molecular chains, and the number of hydrogen bonds.

[0046] Further, in some embodiments, the processor can use feature screening methods such as principal component analysis and feature importance evaluation to analyze the microscopic structure data to generate a molecular structural descriptor representing the relationship between the performance and the microscopic structure of the polyolefin.

[0047] After that, the processor can perform standardization processing on the molecular structural descriptor, such as unit conversion and range normalization, to generate a representative structural descriptor.

[0048] Thus, the method for predicting the above-mentioned polyolefin performance parameters provided by the first aspect of the present invention can ensure that the above-mentioned representative structure descriptors can comprehensively and accurately express the influence of molecular characteristics and process conditions on performance, thereby providing high-quality input data for the subsequent prediction model.

[0049] In addition, in some preferred embodiments, the processor can represent the above-mentioned representative structure descriptors in the form of SMILES (Simplified Molecular Input Line Entry System) characters. Specifically, SMILES characters are a linear string format for representing molecular structures, which concisely describe the atoms and their bonding modes in a molecule through regular symbols and syntax. Subsequently, the processor can convert the SMILES characters into multi-dimensional vectors of atomic features and bond features, specifically including parsing the SMILES characters of polymer repeating units into atomic features and bond features, and converting each atom and bond into a unique digital code through a feature function, that is, for each atom type (such as C, O, N, H, etc.) or bond type (such as single bond, double bond, aromatic bond, etc.), an independent dimension is assigned to support the performance prediction of complex structures.

[0050] After that, the processor can construct a prediction model representing the mapping relationship between the microscopic data and macroscopic mechanical properties of polyolefin molecules for predicting the polyolefin performance parameters.

[0051] Specifically, the processor can first obtain multiple sets of historical sample data of the structural parameters and performance parameters of polyolefin under various process conditions.

[0052] After that, the processor can construct a prediction model to be trained and initialize the prediction model.

[0053] Furthermore, in some embodiments, the above-mentioned prediction model includes an aggregation layer, an attention layer, a normalization layer, a splicing layer, and a fully connected layer.

[0054] Specifically, this aggregation layer is used to transform the above-mentioned representative structure descriptors into continuous high-dimensional embedding vectors. Here, the high-dimensional embedding vectors include multiple nodes and edges. The nodes represent atoms within the polyolefin molecule, and the edges represent chemical bonds between atoms within the polyolefin molecule. This attention layer is used to perform information aggregation on the high-dimensional embedding vectors based on the neighborhood information of each node and edge to update the features of the nodes and edges in the high-dimensional embedding vectors. Here, the neighborhood information at least includes features such as the type of nodes, the type of edges, the distance between nodes within the neighborhood, and the angle. This normalization layer is used to standardize the features in the high-dimensional embedding vectors to ensure the uniformity of the data distribution and the stability of the prediction model. This concatenation layer is used to concatenate the features of multiple normalization layers to generate high-dimensional features. This fully connected layer is used to map the relationship between the high-dimensional features and the property parameters of the polyolefin.

[0055] After that, the processor can train the prediction model based on historical sample data.

[0056] Specifically, the processor can first divide the historical sample data into a training set and a test set. After that, the processor can use the training set to train the prediction model and use the test set to test the prediction model to determine the model loss value of the prediction model. Here, the model loss value is the mean absolute error (MAE), the mean squared error (MSE), and the coefficient of determination (R 2 ), to comprehensively measure the adaptability and stability of the prediction model on different data sets.

[0057] After that, in response to the model loss value being greater than the preset loss threshold, the processor can redefine the hyperparameters of the prediction model and re-train the prediction model according to the redefined hyperparameters.

[0058] After that, in response to the model loss value being less than or equal to the loss threshold, the processor can determine that the training of the prediction model is completed.

[0059] In addition, in some preferred embodiments, the processor can divide the large-scale data set into multiple subsets based on batch training and cross-validation techniques and sequentially use each subset as the test set to train the model, so that the above-mentioned prediction model can achieve excellent generalization ability and prediction accuracy on the large-scale data set.

[0060] After that, as Figure 1 shown, the processor can input the representative structure descriptors into the above-mentioned pre-trained prediction model to predict the performance parameters of the polyolefin.

[0061] Specifically, the prediction model can generate numerical predictions of mechanical property indicators and corresponding error analysis reports. The prediction results include performance trend analysis, error distribution statistics, and interpretation of the important features of the model, enabling users to deeply understand the relationship between the molecular structure and performance parameters of polyolefins, and thus helping to improve the development efficiency and performance optimization level of polyolefin materials.

[0062] In summary, the above-mentioned prediction method for polyolefin performance parameters, the prediction device for polyolefin performance parameters, and the computer-readable storage medium provided by the present invention can all first generate microscopic structure data of polyolefin molecules through molecular dynamics simulation, and then predict the polyolefin performance parameters by constructing a prediction model representing the mapping relationship between the microscopic data of polyolefin molecules and the macroscopic mechanical properties, which is used to improve the accuracy and adaptability of the prediction results of polyolefin performance parameters, thereby improving the development efficiency and performance optimization level of polyolefin materials.

[0063] Although the above methods are illustrated and described as a series of actions for simplicity of explanation, it should be understood and appreciated that these methods are not limited by the order of the actions, because according to one or more embodiments, some actions may occur in a different order and / or concurrently with other actions not illustrated and described herein but understood by those skilled in the art.

[0064] The steps of the methods or algorithms described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read from and write to the storage medium. In an alternative, the storage medium can be integrated into the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In an alternative, the processor and the storage medium can reside in the user terminal as discrete components.

[0065] In one or more exemplary embodiments, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or code. The computer-readable medium includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. The storage media may be any available media that can be accessed by a computer. By way of example and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a web site, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks typically reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0066] The foregoing description of the disclosure has been provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting polyolefin performance parameters, characterized in that: The following steps are involved: Obtaining structural parameters of the polyolefin to be predicted; According to the structural parameters, a dynamic simulation is performed on the polyolefin molecule under various process conditions to generate microstructure data of the polyolefin molecule; as well as The performance parameters of the polyolefin are predicted according to the microstructure data, wherein the performance parameters include a plurality of mechanical property indicators of the polyolefin.

2. The prediction method according to claim 1, characterized in that: The step of performing dynamic simulation on the polyolefin molecules under various process conditions according to the structural parameters to generate microstructure data of the polyolefin molecules comprises: According to the process conditions, defining key parameters in the kinetic simulation process, wherein the key parameters at least include temperature, pressure, simulation time step, boundary conditions, initial molecular configuration and selected force field type; According to the key parameters, adjusting the atomic positions within the polyolefin molecules to minimize the energy of the polyolefin molecular system to eliminate overlapping and high potential energy regions in the polyolefin molecular system; and The kinetic simulation of the polyolefin molecules in the equilibrium stage and the non-equilibrium stage is performed, and the microstructure data of the polyolefin molecules during the kinetic simulation are collected.

3. The prediction method according to claim 2, characterized in that: The step of performing kinetic simulation on the polyolefin molecules in equilibrium stage and non-equilibrium stage comprises: adjusting the temperature during the kinetic simulation process and determining whether the polyolefin molecules have reached thermodynamic equilibrium; and In response to the polyolefin molecule not reaching thermodynamic equilibrium, the initial structure of the polyolefin molecule is changed, and the kinetic simulation is re-performed.

4. The prediction method according to claim 3, characterized in that: The step of performing kinetic simulation on the polyolefin molecules in equilibrium stage and non-equilibrium stage also includes: In response to the polyolefin molecules reaching thermodynamic equilibrium, external disturbances are applied to the polyolefin molecules, and molecular structure and dynamic behavior information of the polyolefin molecules are collected to generate the microscopic data structure, wherein the external disturbances include at least tensile behavior and compressive stress.

5. The prediction method according to claim 4, characterized in that: The microstructure data at least include the bond length, bond angle, intramolecular interaction force and intermolecular interaction force of the polyolefin molecules at the atomic scale.

6. The prediction method according to claim 1, characterized in that: The step of predicting the performance parameters of the polyolefin according to the microstructure data comprises: Performing feature processing on the microstructure data to obtain representative structural descriptors of polyolefin molecules, wherein the representative structural descriptors at least include radial distribution, molecular chain end spacing, and hydrogen bond quantity of the polyolefin molecules; and The representative structural descriptors are input into a pre-trained prediction model to predict the performance parameters of the polyolefin.

7. The prediction method according to claim 6, characterized in that: The step of performing feature processing on the microstructure data to obtain a representative structural descriptor of a polyolefin molecule comprises: Analyzing the microstructure data using a feature screening method to generate a molecular structure descriptor representing the relationship between the performance and microstructure of the polyolefin; and The molecular structure descriptors are normalized to generate the representative structure descriptors.

8. The prediction method according to claim 6, characterized in that: The steps of training the prediction model include: Acquire multiple groups of historical sample data of structural parameters and performance parameters of polyolefins under the multiple process conditions; Constructing a prediction model to be trained and initializing the prediction model; and The prediction model is trained based on the historical sample data.

9. The prediction method according to claim 8, characterized in that: The prediction model includes: A polymerization layer, configured to convert the representative structural descriptor into a continuous high-dimensional embedding vector, wherein the high-dimensional embedding vector includes a plurality of nodes and edges, wherein the nodes represent atoms in the polyolefin molecule, and the edges represent chemical bonds between atoms in the polyolefin molecule; An attention layer is used to aggregate information of the high-dimensional embedding vector according to neighborhood information of each node and edge, so as to update the features of the nodes and edges in the high-dimensional embedding vector, wherein the neighborhood information at least includes features such as the type of the node, the type of the edge, and the distance and angle between the nodes in the neighborhood; A normalization layer, used to standardize the features in the high-dimensional embedding vector to ensure the uniformity of data distribution and the stability of the prediction model; A concatenation layer, used to concatenate features of multiple normalization layers to generate high-dimensional features; and The fully connected layer is used to map the relationship between the high-dimensional features and the property parameters of the polyolefin.

10. The prediction method according to claim 8, characterized in that: The step of training the prediction model according to the historical sample data comprises: Dividing the historical sample data into a training set and a test set; Using the training set to train the prediction model, and using the test set to test the prediction model to determine a model loss value of the prediction model; In response to the model loss value being greater than a preset loss threshold, redefine the hyperparameters of the prediction model, and retrain the prediction model according to the redefined hyperparameters; as well as In response to the model loss value being less than or equal to the loss threshold, it is determined that the training of the prediction model is completed.

11. A device for predicting polyolefin performance parameters, characterized in that: include: a memory having computer instructions stored thereon; and A processor is connected to the memory and is used to execute computer instructions stored in the memory to implement the method for predicting polyolefin performance parameters as described in any one of claims 1 to 10.

12. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the method for predicting a polyolefin performance parameter according to any one of claims 1 to 10 is implemented.

Images