Metallocene catalyst structure determination method and device and metallocene catalyst generation system

Abstract

The invention provides a metallocene catalyst structure determination method, a metallocene catalyst structure determination device and a metallocene catalyst generation system. The method comprises the steps that metal information is obtained, and the metal information is performance information of metal atoms; a structure generation model is constructed, the structure generation model is obtained through training of a particle swarm optimization algorithm by using multiple sets of training data, each set of training data in the multiple sets of training data comprises historical metal information obtained in a historical time period and historical structure information corresponding to the historical metal information, and the historical metal information is obtained according to the historical structure information; the historical structure information is historical information of historical metal atom combinations in the metallocene catalyst; and inputting the metal information into the structure generation model to obtain structure information corresponding to the metal information. According to the scheme, the problems of long time consumption and low efficiency of manually determining the structure of the metallocene catalyst in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of optimization algorithms, and in particular, to a method for determining the structure of a metallocene catalyst, an apparatus for determining the structure of a metallocene catalyst, a computer program product, and a system for generating the structure of a metallocene catalyst. Background Art

[0002] In the study of metallocene catalysts, accurately identifying and constructing the structure of metallocene catalysts is crucial to understanding the catalytic mechanism and optimizing catalyst design. Traditional methods often rely on expert experience for initial guesses and composition, which is not only time-consuming but also inefficient. Summary of the invention

[0003] The main purpose of the present application is to provide a method for determining the structure of a metallocene catalyst, an apparatus for determining the structure of a metallocene catalyst, a computer program product and a system for generating the structure of a metallocene catalyst, so as to at least solve the problem in the prior art that manual determination of the structure of a metallocene catalyst is time-consuming and inefficient.

[0004] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for determining the structure of a metallocene catalyst is provided, comprising: obtaining metal information, wherein the metal information is information on the performance of metal atoms; constructing a structure generation model, wherein the structure generation model is trained by a particle swarm optimization algorithm using multiple sets of training data, and each set of training data in the multiple sets of training data includes historical metal information obtained within a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts; and inputting the metal information into the structure generation model to obtain structure information corresponding to the metal information.

[0005] Optionally, before constructing the structure generation model, the method also includes: obtaining a solution space, wherein the total dimension of the solution space is the sum of the numbers of all particles, wherein the particles are structures of the metal atoms; using a particle swarm optimization algorithm to generate multiple initial structure information, wherein all the particles are combined and changed in the solution space, and the initial structure information is the structure information of the initial metallocene catalyst composed of the metal atoms; calculating all fitness of multiple initial structure information; and determining the initial structure information corresponding to the highest fitness as the target structure information.

[0006] Optionally, obtaining the solution space includes: obtaining degree of freedom information, wherein the degree of freedom information is information on the translation, rotation and spin changes of the particles participating in the reaction; generating a vector based on the degree of freedom information, wherein the vector is used to represent the position and orientation of the particles participating in the reaction in space; and generating the solution space based on at least the vector corresponding to the particles participating in the reaction.

[0007] Optionally, a particle swarm optimization algorithm is used to generate multiple initial structure information, including: an initialization step: generating a particle swarm, wherein the particle swarm includes multiple particles, initializing the positions of all the particles and the speeds of all the particles; an evaluation step: calculating the fitness of each particle respectively; an update step: updating the position of each particle until all the particles have reached the optimal position, and updating the speed of each particle until all the particles have reached the optimal speed; a repetition step: repeating the initialization step, the evaluation step and the update step at least once.

[0008] Optionally, the evaluation step includes: calculating the fitness according to a fitness formula, wherein the fitness formula is:

[0009]

[0010] represents the fitness, represents the vector of the particle, α represents the first weight coefficient, β represents the second weight coefficient, δ k represents the third weight coefficient, represents the penalty term for the distance of the particle, represents the compliance assessment item of the particle, Represents a custom item.

[0011] Optionally, after generating multiple initial structure information using the particle swarm optimization algorithm, the method further includes: stopping the repetitive step when a preset condition is met, wherein the preset condition includes that the number of iterations is greater than or equal to a preset number and the fitness is greater than or equal to a preset fitness.

[0012] Optionally, after inputting the metal information into the structure generation model to obtain the structural information corresponding to the metal information, the method further includes: generating a target image based on the structural information, wherein the target image is an image composed of the metal atoms of the metallocene catalyst; and displaying the target image in a display interface.

[0013] According to another aspect of the present application, a structure determination device for a metallocene catalyst is provided, comprising: a first acquisition unit, for acquiring metal information, wherein the metal information is information on the performance of metal atoms; a construction unit, for constructing a structure generation model, wherein the structure generation model is obtained by training with a particle swarm optimization algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes historical metal information obtained within a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts; and a first processing unit, for inputting the metal information into the structure generation model to obtain structure information corresponding to the metal information.

[0014] According to another aspect of the present application, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any one of the methods for determining the structure of the metallocene catalyst are implemented.

[0015] According to another aspect of the present application, a system for generating a structure of a metallocene catalyst is provided, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the structure determination of any one of the metallocene catalysts.

[0016] By applying the technical solution of the present application, the structural information of the metallocene catalyst is generated by machine learning. Compared with the method of manually judging and designing the metal atom combination in the prior art, the present solution can design the metal atom combination through model automation, which is faster and more efficient than manual design. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for determining the structure of a metallocene catalyst provided in an embodiment of the present application is shown;

[0019] Figure 2 A schematic flow chart of a method for determining the structure of a metallocene catalyst provided in an embodiment of the present application is shown;

[0020] Figure 3 A flow chart showing another method for determining the structure of a metallocene catalyst provided in an embodiment of the present application is shown;

[0021] Figure 4 A structural block diagram of a device for determining the structure of a metallocene catalyst provided according to an embodiment of the present application is shown.

[0022] The above drawings include the following reference numerals:

[0023] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0024] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The transition state is the molecular configuration corresponding to the highest energy point in the reaction coordinates of the elementary reaction. The molecule in the transition state is also called an activated complex. In theory, the activated complex is extremely unstable, and its probability of transformation into reactants and products is equal; in an absolutely irreversible reaction, at the moment of the transition state, all colliding molecules will be transformed into products. According to the theory of quantum mechanics, the activated complex has a zero population, and the transition state is the highest energy point. Any disturbance will cause it to change, so it cannot be separated and cannot be observed.

[0028] The concept of transition state plays an important role in understanding the mechanism of organic reactions. The transition state theory holds that chemical reactions are not completed by simple collisions of reactant molecules, but rather that they pass through a high-energy transition state in the process from reactants to products. This is similar to climbing a mountain, where the highest point of the mountain is the transition state. The transition state is an unstable combination of reactant atoms that can quickly decompose into products in an irreversible reaction. Usually, the energy of the reaction intermediate is not much different from that of the transition state, and it is difficult to distinguish between the two. With the help of femtosecond infrared spectroscopy, it is possible to observe the molecular configuration structure close to the transition state.

[0029] In the study of transition states, accurately identifying and constructing the structure of transition states is crucial for understanding catalytic mechanisms and optimizing catalyst design. Traditional transition state search methods often rely on expert experience for initial guesses, which is not only time-consuming but also inefficient.

[0030] Particle Swarm Optimization (PSO), also known as particle swarm algorithm, is an evolutionary computing technology developed in 1995, which originated from the simulation of a simplified social model.

[0031] The PSO algorithm was originally developed to graphically simulate the graceful and unpredictable movement of bird flocks. Through observations of animal social behavior, it was found that social sharing of information in a group provides an evolutionary advantage, and this was used as the basis for developing the algorithm. The initial version of PSO was formed by adding speed matching of neighbors, considering multi-dimensional search and acceleration based on distance. Later, the inertia weight w was introduced to better control exploitation and exploration, forming the standard version.

[0032] As introduced in the background technology, there is a problem in the prior art that manually determining the structure of a metallocene catalyst is time-consuming and inefficient. To solve the above problem, the embodiments of the present application provide a method for determining the structure of a metallocene catalyst, an apparatus for determining the structure of a metallocene catalyst, a computer program product, and a system for generating the structure of a metallocene catalyst.

[0033] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0034] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal of a method for determining the structure of a metallocene catalyst according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0035] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0036] In this embodiment, a method for determining the structure of a metallocene catalyst running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0037] Figure 2 FIG. 1 is a schematic flow chart of a method for determining the structure of a metallocene catalyst according to an embodiment of the present application. Figure 2As shown, the method comprises the following steps:

[0038] Step S201, obtaining metal information, wherein the metal information is information on the properties of metal atoms;

[0039] Specifically, metal atoms can be iron, copper, zinc, gold, silver, tungsten, etc. Of course, it is not limited to the above-mentioned metals, but can also be any other feasible metal atoms. Metal information is the information of the properties of these metal atoms. For example, gold has extremely high corrosion resistance, good electrical conductivity and thermal conductivity, and the nucleus of gold has a large effective cross section for capturing neutrons, and its infrared reflection ability is close to 100%; iron is a silvery white and shiny metal with a density of 7.86g / cm 3 , melting point 1539℃, has electrical conductivity, thermal conductivity, ductility, can be attracted by magnets, and has ferromagnetism; copper is a soft metal, the surface is red-orange with metallic luster when just cut, and the single substance is purple-red. It has good ductility, high thermal conductivity and electrical conductivity; metal zinc has good rolling, wear resistance, corrosion resistance, castability, and good room temperature performance, and can be made into alloys with various metals; tungsten is a rare high-melting-point metal that can improve the high-temperature hardness of steel. Tungsten is a silvery-white metal that looks like steel. Tungsten has a high melting point, a very low vapor pressure, and a low evaporation rate.

[0040] Specifically, metal information is performance information about metal atoms, including the electron affinity, electronegativity, atomic radius, oxidation state, coordination number, and other physical and chemical properties of metal atoms that are related to catalyst performance.

[0041] Step S202, constructing a structure generation model, wherein the structure generation model is obtained by training with a particle swarm optimization algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes historical metal information obtained in a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts;

[0042] Specifically, through machine learning, at least two metal atoms can be combined without manual design. The combination of metal atoms can be designed automatically by machine, and through particle swarm optimization, the structure generation model can be faster, the global search capability can be stronger, and the efficiency of generating structural information can be higher.

[0043] Specifically, first, it is necessary to collect multiple sets of historical training data, each set of which contains historical metal information and corresponding historical structural information obtained during a historical period, that is, historical information on the combination of historical metal atoms. This may involve the design data of multiple known catalysts, including the composition and configuration of their metal cores, as well as performance data related to them.

[0044] Specifically, taking iron-based metallocene catalysts as an example, the historical metal information can be the performance information of iron atoms in different oxidation states, and the historical structural information can be the specific distribution and configuration of these iron atoms in the catalyst, as well as the performance indicators of the catalyst (such as catalytic activity, selectivity, etc.).

[0045] Specifically, next, the structure generation model is trained using a particle swarm optimization algorithm. The PSO algorithm finds the optimal solution by imitating the flight and social behavior of a flock of birds. In this scenario, each "particle" represents a possible metal atom configuration, and its position and velocity are adjusted in the solution space (i.e., the space of all possible metal atom configurations) to find the best configuration.

[0046] The training process is as follows:

[0047] Initialize the particle swarm: Create a set of randomly configured particles, where the position of each particle represents a specific metal atom configuration, and the velocity represents the rate of adjustment of the configuration.

[0048] Evaluate fitness: For each particle configuration, calculate its similarity to the high-performance catalyst structure in the historical training data, as well as the potential performance of the configuration (such as catalytic activity). Similarity and performance can be used as components of the fitness function to guide the movement of particles.

[0049] Update the best position: Particles update their own positions (i.e., metal atom configurations) based on the feedback of the fitness function, and move closer to the individual historical best configuration and the group historical best configuration.

[0050] Velocity and position update: Using the standard velocity and position update formulas of the PSO algorithm, the velocity and position of the particles are adjusted to explore the solution space and optimize the metal atom configuration.

[0051] Iterative optimization: Repeat the above process until the preset number of iterations is reached or the fitness is no longer significantly improved.

[0052] Step S203: input the metal information into the structure generation model to obtain the structure information corresponding to the metal information.

[0053] Specifically, by inputting the current metal information into a trained structure generation model, the structural information corresponding to the metal information can be obtained. The combination of metal atoms can be automatically designed without manual design.

[0054] Specifically, once the model training is completed, new metal information can be input into the model. The model will use the particle swarm optimization algorithm to generate the structural information that best matches the input metal information based on the trained data and optimization strategy, that is, the predicted structural configuration of the metallocene catalyst.

[0055] For example, if you are designing a new iron-based metallocene catalyst and you know the specific performance metrics of the iron atom, you can input that information into the model. The model will begin the optimization process, generating a range of possible structural configurations. Through iteration and optimization, the model will output a predicted structure that has the optimal configuration of metal atoms to maximize catalyst performance.

[0056] After outputting the predicted structural configuration, the configuration can be further verified experimentally or evaluated for performance using computational chemistry methods such as density functional theory (DFT) calculations to confirm its potential as a metallocene catalyst.

[0057] Through this embodiment, the structural information of the metallocene catalyst is generated by machine learning. Compared with the method of manually judging and designing the metal atom combination in the prior art, this solution can design the metal atom combination through model automation, which is faster and more efficient than manual design.

[0058] Specifically, the transition state guess structure generation method based on the particle swarm optimization algorithm of this scheme uses particle swarm optimization technology to explore and optimize the guessed structure in the potential reaction coordinate space, effectively automates the transition state search process, and improves the efficiency of identifying and constructing the transition state structure.

[0059] In a specific implementation process, the above step S201 can be performed before constructing the structure generation model, and the above method also includes the following steps: obtaining a solution space, wherein the total dimension of the above solution space is the sum of the numbers of all particles, wherein the above particles are structures of the above metal atoms; using a particle swarm optimization algorithm to generate multiple initial structure information, wherein all the above particles are combined and changed in the above solution space, and the above initial structure information is the structure information of the initial metallocene catalyst composed of the above metal atoms; calculating all fitness of the multiple initial structure information; determining the above initial structure information corresponding to the highest fitness as the target structure information.

[0060] In this scheme, the solution space is an N-dimensional space. Particles move and combine in the solution space. The particle swarm optimization algorithm is used to generate initial structural information, and the fitness function is used to measure the quality of the initial structure. The higher the fitness, the better the quality. Therefore, the particle swarm optimization algorithm can effectively automate the transition state search process and improve the efficiency of structural information generation.

[0061] Specifically, the above content is explained. In this embodiment, the solution space refers to the space of all possible metal atom configurations. A particle swarm is created, and each particle represents a possible metal atom configuration in the solution space. The position of the particle can be randomly initialized to cover a wide area of ​​the solution space, and the velocity of the particle is initialized to a smaller value to control the initial intensity of the search. Generate multiple initial structural information: This information reflects the initial position and orientation of the metal atoms in the metallocene catalyst. After the particle swarm is initialized, the configuration of each particle can be regarded as an initial structural information.

[0062] Specifically, the fitness function needs to be designed according to the performance target of the catalyst. For metallocene catalysts, the performance targets may include catalytic activity, selectivity, stability, etc. The fitness function can be combined with the following key evaluation indicators:

[0063] Interatomic distance: ensure that metal atoms and ligands do not come into close contact, which would lead to structural instability;

[0064] Metal-ligand bond energy: Optimizing the bonding between metals and organic ligands to improve catalyst stability;

[0065] Theoretical catalytic performance: The performance of a catalyst is predicted based on computational chemistry models (such as DFT calculations).

[0066] For each particle in the particle swarm, the fitness value is calculated using the fitness function defined above according to the metal atom configuration it represents.

[0067] Specifically, for each particle, if the fitness of its current configuration is higher than the best fitness previously recorded, its individual best position is updated. Traverse the particle group to find the particle with the highest current fitness, and use its configuration as the global best structural information. According to the standard update rule of the PSO algorithm, adjust the position and speed of each particle to move toward the individual and global best positions. Repeat this process until the preset number of iterations is reached or the fitness is no longer significantly improved. After the optimization process is completed, the structural information with the highest fitness is selected as the target structural information, that is, the optimal metallocene catalyst structural configuration.

[0068] For example, designing an iron-based metallocene catalyst with high catalytic activity and stability. Through the particle swarm optimization algorithm, the position and orientation of metal atoms are continuously adjusted to find a structure that can optimize the catalyst performance. At the end of the optimization process, the structural information with the highest fitness will be selected as the final structure of the designed metallocene catalyst. This step provides the basis for subsequent experimental verification or further theoretical calculations.

[0069] In some embodiments, obtaining the solution space can be achieved through the following steps: obtaining degree of freedom information, wherein the degree of freedom information is information about the translation, rotation and spin changes of the particles participating in the reaction; generating vectors based on the degree of freedom information, wherein the vectors are used to represent the positions and orientations of the particles participating in the reaction in space; and generating the solution space at least based on the vectors corresponding to the particles participating in the reaction.

[0070] In this scheme, the particles participating in the reaction are regarded as asymmetric rigid bodies. These asymmetric rigid bodies have degrees of freedom in space and can translate, rotate and spin. The position and orientation of the particles in space are represented by vectors. By combining these vectors, a more accurate solution space can be obtained.

[0071] Specifically, initial conformation optimization can be performed to perform initial spatial conformation optimization on each particle structure participating in the reaction, ensuring that the starting point of each particle structure participating in the reaction is physically reasonable in theory.

[0072] Specifically, the solution space can be defined, and the total dimension N of the solution space will be the sum of the number of non-fixed degrees of freedom of all particle structures:

[0073]

[0074] N represents the total dimension of the solution space, m represents the number of particle structures in the reaction system, and i represents the particle structure. i represents the number of non-fixed degrees of freedom of the ith metal atom. For a metal atom, its degrees of freedom may include translational freedom in three-dimensional space and rotational freedom around three axes. If the movement or rotation of the metal atom in certain directions is restricted (for example, due to ligand constraints), n i will be less than 7.

[0075] For example, when designing a metallocene catalyst containing two metal atoms, assuming that the position and orientation of the metal atoms in space are completely free, then each metal atom has 7 degrees of freedom, and the total dimension N of the solution space will be 14 (2*7).

[0076] Specifically, the solution space is defined as an N-dimensional space, where the non-fixed degree of freedom vector of each particle structure i is is part of this space; the state of the entire system can be represented as an N-dimensional vector Therefore, there is Where R N Represents the N-dimensional real number space.

[0077] The common solution space S is an N-dimensional space composed of the combination of the non-fixed degree of freedom vectors of each particle structure. The common solution space S contains all combinations of spatial positions and orientations that all structures may occupy under the constraints.

[0078] Specifically, the particle structure i is an asymmetric rigid body, which has seven degrees of freedom in three-dimensional space. Specifically, i = [X i ,Y i ,Z i ,θ Xi ,θ Yi ,θ Zi ,ω i ],in:

[0079] (X i ,Y i ,Z i ) represents the translational freedom of the particle structure i on the three orthogonal axes of X-axis, Y-axis and Z-axis;

[0080] θ Xi represents the rotational freedom of particle structure i around the X axis;

[0081] θ Yi represents the rotational freedom of particle structure i around the Y axis;

[0082] θ Zi represents the rotational freedom of particle structure i around the Z axis;

[0083] ω i : represents the spin degree of particle structure i.

[0084] Specifically, for different reaction types, some degrees of freedom may be fixed, that is, they do not participate in the change. In this case, for the particle structure i participating in the reaction, it is assumed that it has ni non-fixed degrees of freedom (0≤ni≤7). Therefore, the non-fixed degree of freedom vector for the particle structure i participating in the reaction is The dimension is ni.

[0085] Specifically, to explain the above, first, determine the degrees of freedom of the particles involved in the reaction. Particles may include metal atoms, organic molecules, or other chemicals. The degree of freedom information mainly focuses on the changes in the particles in translation, rotation, and spin (if applicable). For example, when designing a metallocene catalyst containing an iron atom and a cyclopentadiene ligand, pay attention to the iron atom and the translational degree of freedom, rotational degree of freedom, and possibly spin degree of freedom. If the iron atom is regarded as a rigid body, it has three translational degrees of freedom and three rotational degrees of freedom in three-dimensional space. Assuming that the spin degree of freedom is also considered, the iron atom will have a total of 7 degrees of freedom.

[0086] Specifically, for each particle, the information of its translation, rotation and spin degrees of freedom is converted into a vector. Each element of the vector represents the value of the particle's degree of freedom in a specific direction.

[0087] Specifically, the solution space is a multidimensional space that contains all possible combinations of spatial positions and orientations that all particles can occupy under the constraints. To construct the solution space, the vectors representing the degrees of freedom of each particle need to be combined.

[0088] In this example, assuming there are three particles (an iron atom and two cyclopentadiene ligands), each with 7 degrees of freedom, the total dimension N of the solution space will be the sum of the number of degrees of freedom of all particles: N = 7 + 7 + 7 = 21; this means that the solution space is a 21-dimensional space, which contains the iron atom and its translational, rotational, and spin degrees of freedom, as well as the translational and rotational degrees of freedom of the two ligands.

[0089] Specifically, the solution space may be constructed by the following steps: initializing vector combinations, defining constraint conditions, and generating grids or random points in the solution space.

[0090] Initialization vector combination: To understand each point in space, create a vector containing information about the degrees of freedom of all particles. For example, you can create a 21-dimensional vector, the first 7 dimensions represent the degrees of freedom of the iron atom, and the next two 7 dimensions represent the degrees of freedom of the two cyclopentadiene ligands.

[0091] Define constraints: Define constraints based on the actual situation of the chemical reaction, for example, limit the distance between particles to ensure that they do not overlap or maintain a reasonable chemical bond distance. These constraints will limit the distribution of valid points in the solution space.

[0092] Generate a grid or random points in the solution space: To explore the solution space, you can generate a dense grid of points or random points that represent possible particle configurations. Each point is formed by the vectors defined above and represents a specific combination of particle spatial position and orientation.

[0093] For example, when creating the grid points of the solution space, you can evenly distribute the points within the possible range of each degree of freedom to ensure that all regions of the solution space are covered. If the range of each degree of freedom is - arrive Can be The grid points are created in increments of , which means that each degree of freedom will have 21 possible values, for a total of 2121 points (a very large number for a 21-dimensional space). However, in practice, due to limited computing resources, smart search methods such as random sampling or particle swarm optimization (PSO) are usually used to generate points in the solution space instead of creating a complete grid.

[0094] In the specific implementation process, a particle swarm optimization algorithm is used to generate multiple initial structure information, which can be achieved through the following steps: an initialization step: generating a particle swarm, wherein the particle swarm includes multiple particles, initializing the positions of all the particles and the speeds of all the particles; an evaluation step: calculating the fitness of each of the particles respectively; an update step: updating the position of each particle until all the particles have reached the optimal position, and updating the speed of each particle until all the particles have reached the optimal speed; a repetition step: repeating the initialization step, the evaluation step and the update step at least once.

[0095] In this scheme, the position and velocity of particles can be randomly generated, the position and velocity of each particle can be initialized, and the fitness value of each particle can be evaluated. The position and velocity of the particle can be updated until the optimal position and optimal velocity are reached. These steps are repeated to find the optimal solution and obtain more accurate and optimal initial structure information.

[0096] Specifically, in the common solution space S mentioned above, a particle swarm optimization algorithm is used to generate guessed structures, and the structure with the highest fitness is selected as the guessed result output of the transition state.

[0097] Specifically, initialization: generate a particle swarm, each particle p in the above particle swarm r represents a potential solution, i.e., a point in the common solution space S. Each particle p r Location and speed are randomly initialized;

[0098] Specifically, evaluate the fitness: for each particle p r , according to the above fitness function Calculate fitness;

[0099] Specifically, update the individual and global best positions: for each particle p r , update its individual position to If the particle p r Current location If it is better than the previous individual position, the current position is selected is the best position; traverse each particle p in the above particle group r , update the global best position This is the optimal position achieved so far by all particles in the group;

[0100] Specifically, update the velocity and position: for each particle p r , update its velocity and position according to the following formula:

[0101]

[0102]

[0103] ω is the inertia weight, c1 and c2 are learning factors, λ1 and λ2 are random numbers between [0,1];

[0104] represents the updated particle velocity, Indicates the position of the particle after the update, and those without +1 indicate the data before the update.

[0105] Specifically, the above contents are explained as follows:

[0106] Initialization step: Generate a particle swarm. First, define the size of the particle swarm, that is, the number of particles, which is usually based on the complexity of the problem and computing resources. Each particle represents a point in the solution space, that is, a possible particle structure configuration. For example, if computing resources allow, you may choose to generate a particle swarm containing 100 particles. The position of the particle is represented by the vector of the particle structure configuration, including translation, rotation and spin degrees of freedom. In this embodiment, these positions are randomly initialized to cover a wide area in the solution space, but chemical and physical constraints (for example, the distance between atoms cannot be too close) must be observed. The particle speed initialization is usually set to a smaller random value, which affects the speed and direction of the particle moving in the solution space. When initializing the speed, make sure that the speed is small enough to avoid particles jumping over possible optimal solution areas in the initial stage.

[0107] Evaluation step: Calculate the fitness of each particle. The fitness function is the core of the PSO algorithm and is used to evaluate the quality of the particle's position in the solution space. In chemical reaction transition state optimization, the fitness function may include the following key components: interatomic distance penalty, compliance assessment, and other custom items. The interatomic distance penalty ensures that the atoms in the particle structure are not too close to avoid unreasonable chemical bond formation. The compliance assessment item evaluates whether the particle structure meets the expected reaction mechanism or stability criteria based on prior knowledge of a specific reaction. Other custom items may include additional performance indicators such as catalytic activity, selectivity, etc. For each particle in the particle swarm, the fitness value is calculated using the fitness function defined above according to its corresponding particle structure configuration.

[0108] Update step: Update the position and velocity of the particles. After calculating the fitness of each particle, the PSO algorithm updates the position and velocity of the particles. Each particle remembers its best position in history (i.e., the configuration with the highest fitness) and moves toward this position. The entire particle swarm remembers the optimal position reached by all particles in the swarm, and each particle is guided to move toward this global optimal position. According to the current velocity of the particle, the personal best position, and the global best position, the velocity update formula of PSO is used to adjust the velocity of each particle. Use the updated velocity to adjust the position of each particle, thereby changing the configuration of the particle structure.

[0109] Repeating steps: Repeat the initialization, evaluation, and update steps. The PSO algorithm is an iterative process that requires repeated initialization, evaluation, and update steps until a stopping condition is met. Stopping conditions may include reaching a maximum number of iterations, the fitness value reaching a predetermined threshold, or no significant change in fitness over a period of time. In each iteration, the particle swarm explores the solution space and gradually concentrates on structural configurations with higher fitness. This process may take hundreds or thousands of iterations to converge to a region of the solution space that is close to the optimal solution.

[0110] For example, when searching for the transition state configuration of a new iron-based metallocene catalyst, one might need to perform the following iterative process:

[0111] Iteration 1: Initialize the particle swarm and randomly generate 100 particle structure configurations, each of which includes the position and orientation information of the iron atom and the cyclopentadiene ligand. Calculate the fitness of each particle based on the fitness function defined above.

[0112] Iteration 2: Based on the results of the previous iteration, update the particle speed and position to move the particle group toward a configuration with higher fitness. Calculate the fitness of each particle again.

[0113] Iteration 3: Repeat iteration 2 until the particle swarm converges to the configuration with the highest fitness or reaches the preset upper limit of the number of iterations.

[0114] In some embodiments, the evaluation step may be implemented by the following steps: calculating the fitness according to a fitness formula, wherein the fitness formula is:

[0115]

[0116] represents the above fitness, represents the vector of the above particles, α represents the first weight coefficient, β represents the second weight coefficient, δ k represents the third weight coefficient, represents the penalty term for the distance of the above particles, Indicates the compliance assessment items of the above particles, Represents a custom item.

[0117] In this solution, a relatively accurate fitness can be calculated through the fitness formula of this solution, so that optimization can be performed based on the fitness later. The above formula is only exemplary, and any variation of the formula falls within the protection scope of this application.

[0118] Specifically, in the optimization process of the particle swarm algorithm, the fitness function plays a key guiding role. This function measures the quality of the guessed structure based on prior knowledge (such as reaction mechanism). Setting the fitness function in Represents the position vector of all atoms in the reaction system, expressed as a weighted sum of several key parts, and the weight coefficient is used to adjust the relative importance of different parts.

[0119] Penalty term for interatomic distance In order to avoid unreasonable bonding caused by too close distance between atoms, in the fitness function A penalty term is set in . According to the distance d between atoms p and q ij and a reasonable minimum distance d min To define, its use

[0120]

[0121] to penalize pairs of atoms that are too close.

[0122] Specifically, the conformity assessment items It is to evaluate the conformity of the particle structure according to the a priori reaction mechanism or stability criteria of a specific reaction.

[0123] Specifically, the conformity assessment items is based on the ideal distance between the specified atoms p and q:

[0124]

[0125] Evaluate the conformity of the particle structure, where: It is the distance between the atomic pair (p,q) in the guessed structure in the prior reaction mechanism.

[0126] Specifically, More evaluation items can be added according to the needs of specific application scenarios.

[0127] Specifically, the distance penalty term, conformity assessment term, and custom term in the above formula need to be determined based on the specific atomic composition of the catalyst, the empirical reaction mechanism of this type of metallocene catalyst, and the specific application needs. Specifically:

[0128] The penalty term for the interatomic distance is to avoid unreasonable bonding situations caused by the atoms in the generated structure being too close.

[0129] The formula is:

[0130]

[0131] in, - represents the sum of the distances between atom p and atom q among all atoms. min It represents the reasonable minimum distance, which is usually set according to the atom type, such as the van der Waals radius, covalent radius of different atoms, or the minimum interaction distance observed in reaction experience.

[0132] For example, for an iron-based metallocene catalyst, it is desirable to avoid an unreasonable bonding situation where the distance between the iron atom and the carbon atom in the metallocene ring is too short. min can be set as the minimum reasonable distance between the iron atom and the carbon atom. Considering the sum of the covalent radius or van der Waals radius of iron and carbon, the distance between iron and carbon in the cyclopentadienyl ring should be no less than In addition, adjacent carbon atoms in the cyclopentadienyl ring have specific ideal bond lengths (approximately ), the bond length cannot be too short or too long.

[0133] The reaction mechanism-based conformity assessment item is to ensure that the generated transition state guess structure conforms to the known reaction mechanism, and is usually evaluated based on the ideal distance or geometric relationship between specific atoms.

[0134] The formula is:

[0135]

[0136] Among them, d pq Represents the actual distance between atoms p and q in the guessed structure. Ideal distance based on reaction mechanism.

[0137] For example, for an iron-based metallocene catalyst, in the transition state of the iron-based metallocene catalyst, the iron atom should form a specific geometric relationship with the cyclopentadienyl ring to achieve appropriate electron transfer and ligand exchange. For example, in the transition state, the iron atom may need to deviate from the cyclopentadienyl ring plane by a certain distance in order to form a new interaction with a certain group in the reactant. Here It can be set as the expected ideal distance between iron and a ligand atom in the transition state (for example, based on the quantitative calculation results of the same type of reaction), and the fitness function can be used to ensure that the structure approaches the ideal state.

[0138] Custom items are used to add other important structural requirements and optimization goals for specific reaction types or catalyst application scenarios. These items can be flexibly set according to specific reaction requirements and application goals.

[0139] The formula is:

[0140]

[0141] Among them, each Can represent different constraints or optimization objectives and vary depending on the specific catalyst reaction requirements.

[0142] For example, for an iron-based metallocene catalyst, certain ligands must form a channel wide enough for a particular molecule to pass through. In the customization, you can add a constraint to ensure a minimum channel width, preventing the generation of structures that are too crowded or block the reaction channel.

[0143] In addition, for an iron-based metallocene catalyst, in order to achieve higher reaction activity, a higher electron density near the metal atoms may be required, and the introduction of constraints related to the electron density distribution can be considered.

[0144] In the specific implementation process, after using the particle swarm optimization algorithm to generate multiple initial structure information, the above method also includes the following steps: when the preset conditions are met, stop the above repetitive steps, wherein the above preset conditions include the number of iterations being greater than or equal to the preset number, and the above fitness being greater than or equal to the preset fitness.

[0145] In this solution, by pre-setting preset conditions, when the preset conditions are met, the repeated search for the optimal solution can be stopped, thereby avoiding falling into a loop, avoiding unnecessary calculations, and improving the efficiency of the overall algorithm.

[0146] Specifically, taking the application of PSO algorithm in designing transition state configuration of iron-based metallocene catalysts as an example, this paper discusses how to set and monitor preset conditions to decide when to stop the optimization process.

[0147] Default condition settings:

[0148] Iterations: The preset maximum number of iterations is 1000, which is usually based on a balance between computing resources and expected optimization accuracy. If the algorithm has not converged after 1000 iterations, it may be necessary to re-evaluate the fitness function or particle swarm parameters.

[0149] Fitness threshold: The preset fitness threshold is a performance indicator, for example, the catalytic activity is greater than or equal to a certain value. This depends on the performance requirements of the target catalyst, for example, the catalytic activity may be required to be greater than or equal to 1000 mol / h / g.

[0150] At the end of each iteration, the state of the particle swarm is checked to determine whether any of the stopping conditions are met:

[0151] The number of iterations reaches the preset number: When the 1000th iteration is reached, the algorithm will stop regardless of whether the particle swarm finds a configuration that meets the fitness threshold.

[0152] Fitness reaches a preset threshold: Even during the iteration process, if any particle in the particle swarm reaches the preset fitness threshold (for example, the catalytic activity is greater than or equal to 1000 mol / h / g), the algorithm will stop immediately.

[0153] For example, assuming that at the 500th iteration, the algorithm finds a transition state configuration with a catalytic activity greater than 1000 mol / h / g, then according to the preset condition of "fitness is greater than or equal to the preset fitness", the algorithm will stop immediately and no further iterations will be performed.

[0154] In some embodiments, after inputting the above-mentioned metal information into the above-mentioned structure generation model to obtain the structural information corresponding to the above-mentioned metal information, the above-mentioned method also includes the following steps: generating a target image based on the above-mentioned structural information, wherein the above-mentioned target image is an image composed of the above-mentioned metal atoms of the above-mentioned metallocene catalyst; and displaying the above-mentioned target image in a display interface.

[0155] In this solution, a target image can be generated through structural information and then displayed in a display interface, so that the structural information of the metallocene catalyst can be displayed intuitively, making it convenient for staff to view the structural information of the metallocene catalyst.

[0156] Specifically, after the particle swarm optimization algorithm (PSO) or other optimization process is completed, a set of optimized structural information will be obtained, which describes the best configuration of metal atoms in the metallocene catalyst. The structural information may include the precise coordinates of each metal atom, the connection method of the ligands, and the relative position and orientation of all atoms.

[0157] Suppose you are designing an iron-based metallocene catalyst, and the optimization process has determined the best position and orientation of the iron atom and the cyclopentadienyl ligand. The structural information might include the coordinates of the iron atom in three-dimensional space, the connection pattern of the cyclopentadienyl ligand, and the relative positions of all other atoms.

[0158] Use chemical visualization software or programming libraries such as PyMOL, Jmol, or visualization modules in chemical computing packages to convert structural information into intuitive images. These tools can generate accurate chemical structure images based on atomic coordinates, chemical bond information, and how ligands are connected.

[0159] For example, after optimization, the structural information of the iron atom and two cyclopentadiene ligands was obtained. You can use PyMOL software to input the optimized coordinate data, set the atomic color, display mode (such as ball-and-stick model, wireframe model or space-filling model), and adjust the viewing angle to generate a detailed metallocene catalyst structure image.

[0160] The generated metallocene catalyst structure images are embedded into a user interface for easy viewing and analysis. This could be a specially designed chemical computing software interface, a web application, or a mobile application.

[0161] The display interface may include the following functions:

[0162] Zoom and Pan Tools: Users can zoom in, zoom out, or move the image to view specific areas in more detail.

[0163] Rotation tool: Users can rotate the image to observe the relative positions of metal atoms and ligands from different perspectives.

[0164] Atom and Bond Annotation: Users can view or hide the names or types of atoms and bonds to help identify key features in the structure.

[0165] Performance data display: The optimized structural information may also contain predicted data of catalyst performance, such as catalytic activity, selectivity, etc. These data can be displayed next to the image to provide users with additional reference information.

[0166] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the method for determining the structure of the metallocene catalyst of the present application will be described in detail below in conjunction with specific embodiments.

[0167] This embodiment relates to a specific method for determining the structure of a metallocene catalyst, such as Figure 3 As shown, the following steps are included:

[0168] Step (1), initial conformation optimization; step (2), definition of solution space; step (3), setting of fitness function; step (4), particle swarm optimization; step (5), selection and output of transition state structure.

[0169] The present application embodiment also provides a kind of structure determination device of metallocene catalyst, it should be noted that the structure determination device of metallocene catalyst of the present application embodiment can be used to execute the structure determination method for metallocene catalyst provided by the present application embodiment. The device is used to implement the above-mentioned embodiment and preferred embodiment, and the description has been made and will not be repeated. As used below, the term "module" can realize the combination of software and / or hardware of predetermined functions. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or the combination of software and hardware is also possible and conceived.

[0170] The following is an introduction to the structure determination device of the metallocene catalyst provided in the examples of the present application.

[0171] Figure 4 is a structural block diagram of a device for determining the structure of a metallocene catalyst according to an embodiment of the present application. Figure 4 As shown, the device comprises:

[0172] A first acquisition unit 10 is used to acquire metal information, wherein the metal information is information on properties of metal atoms;

[0173] A construction unit 20 is used to construct a structure generation model, wherein the structure generation model is obtained by training with a particle swarm optimization algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes historical metal information obtained in a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts;

[0174] The first processing unit 30 is used to input the metal information into the structure generation model to obtain the structure information corresponding to the metal information.

[0175] Through this embodiment, the structural information of the metallocene catalyst is generated by machine learning. Compared with the method of manually judging and designing the metal atom combination in the prior art, this solution can design the metal atom combination through model automation, which is faster and more efficient than manual design.

[0176] In the specific implementation process, the above-mentioned device also includes a second acquisition unit, a first generation unit, a calculation unit and a determination unit. The second acquisition unit is used to obtain a solution space before constructing a structure generation model, wherein the total dimension of the above-mentioned solution space is the sum of the number of all particles, wherein the above-mentioned particles are the structures of the above-mentioned metal atoms; the first generation unit is used to generate multiple initial structure information using a particle swarm optimization algorithm, wherein all of the above-mentioned particles are combined and changed in the above-mentioned solution space, and the above-mentioned initial structure information is the structure information of the initial metallocene catalyst composed of the above-mentioned metal atoms; the calculation unit is used to calculate all fitness of the above-mentioned multiple initial structure information; the determination unit is used to determine that the above-mentioned initial structure information corresponding to the highest fitness is the target structure information.

[0177] In this scheme, the solution space is an N-dimensional space. Particles move and combine in the solution space. The particle swarm optimization algorithm is used to generate initial structural information, and the fitness function is used to measure the quality of the initial structure. The higher the fitness, the better the quality. Therefore, the particle swarm optimization algorithm can effectively automate the transition state search process and improve the efficiency of structural information generation.

[0178] In some embodiments, the second acquisition unit includes an acquisition module, a first generation module and a second generation module, the acquisition module is used to acquire degree of freedom information, wherein the above-mentioned degree of freedom information is information on the translation, rotation and spin changes of the above-mentioned particles participating in the reaction; the first generation module is used to generate vectors based on the above-mentioned degree of freedom information, wherein the above-mentioned vectors are used to represent the position and orientation of the above-mentioned particles participating in the reaction in space; the second generation module is used to generate the above-mentioned solution space based on at least the above-mentioned vectors corresponding to the above-mentioned particles participating in the reaction.

[0179] In this scheme, the particles participating in the reaction are regarded as asymmetric rigid bodies. These asymmetric rigid bodies have degrees of freedom in space and can translate, rotate and spin. The position and orientation of the particles in space are represented by vectors. By combining these vectors, a more accurate solution space can be obtained.

[0180] In the specific implementation process, the first generation unit includes a third generation module, a calculation module, a first processing module and a second processing module, the third generation module is used to perform an initialization step: generate a particle group, wherein the particle group includes a plurality of the particles, and initialize the positions of all the particles and the speeds of all the particles; the calculation module is used to perform an evaluation step: calculate the fitness of each of the particles respectively; the first processing module is used to perform an update step: update the position of each of the particles until all the particles have reached the optimal position, and update the speed of each of the particles until all the particles have reached the optimal speed; the second processing module is used to perform a repetition step: repeat the initialization step, the evaluation step and the update step at least once.

[0181] In this scheme, the position and velocity of particles can be randomly generated, the position and velocity of each particle can be initialized, and the fitness value of each particle can be evaluated. The position and velocity of the particle can be updated until the optimal position and optimal velocity are reached. These steps are repeated to find the optimal solution and obtain more accurate and optimal initial structure information.

[0182] In some embodiments, the calculation module includes a calculation submodule, and the calculation submodule is used to calculate the fitness according to the fitness formula, wherein the fitness formula is:

[0183]

[0184] represents the above fitness, represents the vector of the above particles, α represents the first weight coefficient, β represents the second weight coefficient, δ k represents the third weight coefficient, represents the penalty term for the distance of the above particles, Indicates the compliance assessment items of the above particles, Represents a custom item.

[0185] In this solution, a relatively accurate fitness can be calculated through the fitness formula of this solution, so that optimization can be performed based on the fitness later. The above formula is only exemplary, and any variation of the formula falls within the protection scope of this application.

[0186] In the specific implementation process, the above-mentioned device also includes a second processing unit, which is used to stop the above-mentioned repetitive steps after generating multiple initial structure information using the particle swarm optimization algorithm when preset conditions are met, wherein the above-mentioned preset conditions include that the number of iterations is greater than or equal to the preset number and the above-mentioned fitness is greater than or equal to the preset fitness.

[0187] In this solution, by pre-setting preset conditions, when the preset conditions are met, the repeated search for the optimal solution can be stopped, thereby avoiding falling into a loop, avoiding unnecessary calculations, and improving the efficiency of the overall algorithm.

[0188] In some embodiments, the above-mentioned device also includes a second generation unit and a display unit, the second generation unit is used to generate a target image according to the above-mentioned structural information after inputting the above-mentioned metal information into the above-mentioned structure generation model to obtain the structural information corresponding to the above-mentioned metal information, wherein the above-mentioned target image is an image composed of the above-mentioned metal atoms of the above-mentioned metallocene catalyst; the display unit is used to display the above-mentioned target image in a display interface.

[0189] In this solution, a target image can be generated through structural information and then displayed in a display interface, so that the structural information of the metallocene catalyst can be displayed intuitively, making it convenient for staff to view the structural information of the metallocene catalyst.

[0190] The above-mentioned metallocene catalyst structure determination device includes a processor and a memory, and the above-mentioned first acquisition unit, construction unit and first processing unit are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0191] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of long time and low efficiency in manually determining the structure of the metallocene catalyst in the prior art can be solved by adjusting the kernel parameters.

[0192] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0193] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the structure determination method of the metallocene catalyst.

[0194] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the method for determining the structure of the metallocene catalyst is executed when the program is run.

[0195] An embodiment of the present invention provides a device, the device including a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the following steps are implemented:

[0196] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0197] A computer program product includes a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method in each embodiment of the present application are implemented.

[0198] An embodiment of the present invention provides a structure generation system for a metallocene catalyst, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the above-mentioned metallocene catalyst structure determination methods.

[0199] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0200] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0201] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0202] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0203] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0204] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0205] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0206] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0207] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0208] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0209] 1) A method for determining the structure of a metallocene catalyst in the present application generates structural information of the metallocene catalyst by machine learning. Compared with the method of manually judging and designing metal atom combinations in the prior art, this scheme can design metal atom combinations through model automation, which is faster and more efficient than manual design.

[0210] 2) A structure determination device for a metallocene catalyst of the present application generates structural information of the metallocene catalyst by machine learning. Compared with the method of manually judging and designing metal atom combinations in the prior art, this solution can design metal atom combinations through model automation, which is faster and more efficient than manual design.

[0211] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for determining the structure of a metallocene catalyst, characterized in that: include: Acquiring metal information, wherein the metal information is information about properties of metal atoms; Constructing a structure generation model, wherein the structure generation model is obtained by training with a particle swarm optimization algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes historical metal information obtained in a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts; The metal information is input into the structure generation model to obtain the structure information corresponding to the metal information.

2. The method according to claim 1, characterized in that Before constructing the structure generation model, the method further includes: Obtaining a solution space, wherein the total dimension of the solution space is the sum of the numbers of all particles, wherein the particles are structures of the metal atoms; A particle swarm optimization algorithm is used to generate a plurality of initial structure information, wherein all the particles are combined and changed in the solution space, and the initial structure information is the structure information of the initial metallocene catalyst composed of the metal atoms; Calculating all fitness of the plurality of initial structure information; The initial structure information corresponding to the highest fitness is determined as the target structure information.

3. The method according to claim 2, characterized in that Get the solution space, including: Acquiring degree of freedom information, wherein the degree of freedom information is information on the translation, rotation and spin changes of the particles participating in the reaction; Generate a vector according to the degree of freedom information, wherein the vector is used to represent the position and orientation of the particle participating in the reaction in space; The solution space is generated at least according to the vector corresponding to the particle participating in the reaction.

4. The method according to claim 2, characterized in that: The particle swarm optimization algorithm is used to generate multiple initial structure information, including: Initialization step: generating a particle swarm, wherein the particle swarm includes a plurality of the particles, and initializing positions and velocities of all the particles; Evaluation step: calculating the fitness of each particle respectively; Updating step: updating the position of each particle until all the particles have reached the optimal position, and updating the speed of each particle until all the particles have reached the optimal speed; Repeating step: repeating the initialization step, the evaluation step and the updating step at least once.

5. The method according to claim 4, characterized in that The evaluation steps include: The fitness is calculated according to a fitness formula, wherein the fitness formula is: represents the fitness, represents the vector of the particle, α represents the first weight coefficient, β represents the second weight coefficient, δ k represents the third weight coefficient, represents the penalty term for the distance of the particle, represents the compliance assessment item of the particle, Represents a custom item.

6. The method according to claim 4, characterized in that After the particle swarm optimization algorithm is used to generate a plurality of initial structure information, the method further comprises: When a preset condition is met, the repetitive step is stopped, wherein the preset condition includes that the number of iterations is greater than or equal to a preset number and the fitness is greater than or equal to a preset fitness.

7. The method according to any one of claims 1 to 6, characterized in that: After inputting the metal information into the structure generation model to obtain the structure information corresponding to the metal information, the method further includes: Generate a target image according to the structural information, wherein the target image is an image composed of the metal atoms of the metallocene catalyst; The target image is displayed in a display interface.

8. A device for determining the structure of a metallocene catalyst, characterized in that: include: A first acquisition unit, used to acquire metal information, wherein the metal information is information on properties of metal atoms; A construction unit, used for constructing a structure generation model, wherein the structure generation model is obtained by training with a particle swarm optimization algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes historical metal information obtained in a historical time period and historical structure information corresponding to the historical metal information, wherein the historical structure information is historical information of historical metal atom combinations in metallocene catalysts; The first processing unit is used to input the metal information into the structure generation model to obtain the structure information corresponding to the metal information.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for determining the structure of the metallocene catalyst according to any one of claims 1 to 7 are implemented.

10. A structure generation system of a metallocene catalyst, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the structure determination method of the metallocene catalyst according to any one of claims 1 to 7.

Images