Material particle size grading method and device, storage medium and equipment

Abstract

The present disclosure relates to a material particle size grading method and device, a storage medium and equipment, by acquiring a first particle size grading data set, inputting the particle size grading data in the first particle size grading data set into a first generation model, a second particle size grading data set containing particle size grading data not covered by the first particle size grading data set can be generated, which is beneficial to expand the combination space of particle size and ratio; on this basis, by determining target particle size grading data from the first particle size grading data set and the second particle size grading data set that can make the material performance after N times of sintering meet the preset condition, it is beneficial to improve the accuracy of particle size grading data optimization and the reliability of material performance optimization; and the material particle size grading method provided in the embodiments of the present disclosure does not need to rely on artificial experience exploration, saves the labor cost and improves the efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of new energy technology, and in particular to a method and apparatus for material particle size distribution, a storage medium and equipment. Background Technology

[0002] In related technologies, the performance of battery electrode materials is one of the key factors affecting the performance of new energy vehicles and energy storage devices. Therefore, optimizing the performance of battery electrode materials is an important approach to optimizing the performance of new energy vehicles and energy storage devices. Currently, the optimization methods for battery electrode material performance mainly focus on atomic-level doping, micro-particle design, particle size distribution control, and sintering process improvement. Among these, particle size distribution methods mostly rely on empirical exploration, resulting in low experimental efficiency and difficulty in fully covering the space of particle size and ratio combinations. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this disclosure provides a material particle size distribution method and apparatus, storage medium and equipment.

[0004] According to a first aspect of the present disclosure, a method for material particle size distribution is provided, comprising: acquiring a first particle size distribution dataset, the first particle size distribution dataset including at least one set of particle size distribution data; inputting the particle size distribution data in the first particle size distribution dataset into a first generation model, generating a second particle size distribution dataset based on the first generation model, wherein the particle size distribution data included in the second particle size distribution dataset is not included in the first particle size distribution dataset; determining target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset, such that the material properties obtained after N sinterings based on the target particle size distribution data meet preset conditions, where N is an integer greater than or equal to 1.

[0005] By acquiring a first particle size distribution dataset and inputting the particle size distribution data from the first particle size distribution dataset into a first generation model, a second particle size distribution dataset containing particle size distribution data not covered by the first particle size distribution dataset can be generated, which is beneficial to expanding the combination space of particle size and ratio. On this basis, by determining the target particle size distribution data that can make the material properties after N sintering meet the preset conditions from the first particle size distribution dataset and the second particle size distribution dataset, it is beneficial to improve the accuracy of particle size distribution data optimization and the reliability of material property optimization. Furthermore, the material particle size distribution method provided in this embodiment does not need to rely on manual experience exploration, saving labor costs and improving efficiency.

[0006] In some exemplary embodiments of this disclosure, the particle size distribution data in the first particle size distribution dataset and the second particle size dataset includes the following data: particle size ratio, sintering process parameters, and particle size distribution information after N sintering cycles.

[0007] This disclosure provides sufficient data for particle size distribution data optimization and material performance optimization by incorporating multi-dimensional information such as particle size ratio, sintering process parameters, and particle size distribution information after N sinterings into the particle size distribution data. This is beneficial to improving the accuracy of particle size distribution data optimization and material performance optimization.

[0008] In some exemplary embodiments of this disclosure, obtaining the first particle size distribution dataset includes: obtaining at least one set of sintering conditions, the sintering conditions including particle size distribution and sintering process parameters; for each set of sintering conditions, inputting the sintering conditions into a second generation model, predicting particle size distribution information after N sinterings based on the second generation model; and generating the first particle size distribution dataset based on each set of sintering conditions and the particle size distribution information after N sinterings under each set of sintering conditions.

[0009] By inputting the obtained sintering conditions into the second generation model, and using the second generation model to predict the particle size distribution information after N sinterings, the efficiency of obtaining particle size distribution information can be improved. Thus, when generating the first particle size distribution dataset, the first particle size distribution dataset can be quickly generated based on at least one set of sintering conditions and the particle size distribution information corresponding to each set of sintering conditions.

[0010] In some exemplary embodiments of this disclosure, determining the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset includes: for the particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, using a preset performance prediction model to predict the material properties corresponding to the particle size distribution data, wherein the material properties refer to the material properties obtained after N sintering based on the particle size distribution data; and determining the particle size distribution data whose material properties are closest to the target material properties as the target particle size distribution data based on the material properties corresponding to each particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset.

[0011] By predicting the material properties corresponding to the particle size distribution data contained in the first and second particle size distribution datasets using a pre-defined performance prediction model, the time required to obtain material properties experimentally can be saved, and the efficiency of obtaining material properties corresponding to particle size distribution data can be improved. Based on this, and using the material properties corresponding to each particle size distribution data in the first and second particle size distribution datasets, the particle size distribution data whose material properties are closest to the target material properties can be quickly obtained, and this particle size distribution data can be identified as the target particle size distribution data.

[0012] In some exemplary embodiments of this disclosure, determining the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset includes: using a Bayesian optimization algorithm with the target material properties as the optimization objective to determine the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset.

[0013] Using the target material properties as the optimization objective, a Bayesian optimization algorithm can be employed to quickly determine the target particle size distribution data from the first and second particle size distribution datasets, thereby improving the efficiency of determining the target particle size distribution data.

[0014] According to a second aspect of the present disclosure, a material particle size distribution apparatus is provided, comprising: The acquisition module is configured to acquire a first particle size distribution dataset, which includes at least one set of particle size distribution data.

[0015] The generation module is configured to input the particle size distribution data in the first particle size distribution dataset into the first generation model, and generate a second particle size distribution dataset based on the first generation model, wherein the particle size distribution data in the second particle size distribution dataset is not included in the first particle size distribution dataset.

[0016] The determining module is configured to determine target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset, such that the material properties obtained after N sinterings based on the target particle size distribution data meet preset conditions, where N is an integer greater than or equal to 1.

[0017] In some exemplary embodiments of this disclosure, the particle size distribution data in the first particle size distribution dataset and the second particle size dataset includes the following data: particle size ratio, sintering process parameters, and particle size distribution information after N sintering cycles.

[0018] In some exemplary embodiments of this disclosure, the acquisition module is configured to: acquire at least one set of sintering conditions, the sintering conditions including particle size distribution and sintering process parameters; for each set of sintering conditions, input the sintering conditions into a second generation model, predict particle size distribution information after N sinterings based on the second generation model; and generate a first particle size distribution dataset based on each set of sintering conditions and the particle size distribution information after N sinterings under each set of sintering conditions.

[0019] In some exemplary embodiments of this disclosure, the determining module is configured to: for the particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, use a preset performance prediction model to predict the material properties corresponding to the particle size distribution data, wherein the material properties refer to the material properties obtained after N sintering based on the particle size distribution data; and based on the material properties corresponding to each particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, determine the particle size distribution data whose material properties are closest to the target material properties as the target particle size distribution data.

[0020] In some exemplary embodiments of this disclosure, the determining module is configured to: determine target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset using a Bayesian optimization algorithm, with the target material properties as the optimization objective.

[0021] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the method of any of the embodiments of the first aspect described above.

[0022] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the method of any of the embodiments of the first aspect described above.

[0023] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0025] Figure 1 This is a flowchart illustrating a material particle size distribution method according to some embodiments of the present disclosure.

[0026] Figure 2 This is a schematic diagram illustrating a method for generating particle size distribution data according to some embodiments of the present disclosure.

[0027] Figure 3 This is a schematic diagram illustrating a method for generating particle size distribution data according to some embodiments of the present disclosure.

[0028] Figure 4 This is a schematic diagram illustrating a method for determining target particle size distribution data according to some embodiments of the present disclosure.

[0029] Figure 5 This is a schematic diagram illustrating another method for determining target particle size distribution data according to some embodiments of the present disclosure.

[0030] Figure 6 This is a flowchart illustrating a method for obtaining a particle size distribution dataset according to some embodiments of the present disclosure.

[0031] Figure 7 This is a flowchart illustrating a method for determining target particle size distribution data according to some embodiments of the present disclosure.

[0032] Figure 8 This is a block diagram illustrating a material particle size distribution device according to some embodiments of the present disclosure.

[0033] Figure 9 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0034] Some embodiments of this disclosure will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a particular order. Furthermore, for clarity and brevity, descriptions of features known in the art may be omitted.

[0035] In view of the relevant technology, this disclosure provides a material particle size distribution scheme. The technical solution of this disclosure will be described below with reference to exemplary embodiments.

[0036] The embodiments described in the following examples of this disclosure are not representative of all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0037] Figure 1 This is a flowchart illustrating a material particle size distribution method according to some embodiments of the present disclosure. This material particle size distribution method can be used in electronic devices, such as desktop computers, laptops, distributed processing nodes, servers, etc., but is not limited to the devices listed herein. Figure 1 As shown, in some embodiments, the material particle size distribution method provided in this disclosure may include the following steps.

[0038] In step S101, a first particle size distribution dataset is obtained, which includes at least one set of particle size distribution data.

[0039] In this embodiment of the disclosure, the first particle size distribution dataset refers to a set that includes particle size distribution data.

[0040] Particle size distribution data should include at least the particle size distribution ratio. The particle size distribution ratio is used to characterize the proportional relationship between powder raw materials of different particle size grades, including but not limited to the mass or volume percentage of coarse, medium, and fine particles.

[0041] In some examples, the particle size distribution data referred to in this disclosure may also include sintering process parameters. These sintering process parameters are used to control the process conditions of the sintering process, including but not limited to sintering temperature, heating rate, holding time, sintering atmosphere, and sintering pressure, to achieve precise control over the material densification process and grain growth process.

[0042] In some examples, the particle size distribution data referred to in this disclosure may also include particle size distribution information obtained after N sintering processes under the corresponding particle size distribution and sintering process parameters. For example, 3D point clouds of sintered particles, coordinates of particles of different sizes, volume percentage, mass percentage, etc.

[0043] In some examples, the first particle size distribution dataset can be obtained experimentally or through other feasible methods, and this disclosure does not impose specific limitations.

[0044] This disclosure provides sufficient data for particle size distribution data optimization and material performance optimization by including multi-dimensional information such as particle size ratio, sintering process parameters, and particle size distribution information after N sinterings in the particle size distribution data. This is beneficial to improving the accuracy of particle size distribution data optimization and material performance optimization.

[0045] In some examples, the N sinterings referred to in the embodiments of this disclosure can be any number of sinterings. For example, when N equals 1, the particle size distribution data in the first particle size distribution dataset can be understood as the relevant data of the precursor raw materials used for one sintering. When N equals 2, the particle size distribution data in the first particle size distribution dataset can be understood as the relevant data of the precursor raw materials used for a second sintering, and so on. This disclosure will not provide further examples.

[0046] In step S103, the particle size distribution data in the first particle size distribution dataset is input into the first generation model, and a second particle size distribution dataset is generated based on the first generation model. The particle size distribution data in the second particle size distribution dataset is not included in the first particle size distribution dataset.

[0047] The first generative model referred to in the embodiments of this disclosure can be any artificial intelligence model with open generative capabilities. In some applications of the embodiments of this disclosure, the first generative model is used to generate a set of particle size distribution data. The input of the first generative model is the particle size distribution data in the first particle size distribution dataset, and the output is a second particle size distribution data set. The second particle size distribution data set consists of particle size distribution data that are not covered by the first particle size distribution dataset.

[0048] For example, in some examples, the first generative model may include an encoder and a decoder. The encoder generates a latent variable space based on the particle size distribution data in the first particle size distribution dataset and generates a set of candidate latent variables using Latin hypercube sampling. The decoder decodes the set of candidate latent variables to obtain the second particle size distribution dataset.

[0049] for example, Figure 2 This is a schematic diagram illustrating a method for generating particle size distribution data according to some embodiments of this disclosure. Figure 2 As shown, assume that the first particle size distribution dataset includes particle size distribution data A, particle size distribution data B, and particle size distribution data C. In some examples, particle size distribution data A, B, and C can be input into a first generation model, which generates particle size distribution data D, E, F, and G, thus obtaining a second particle size distribution dataset containing particle size distribution data D, E, F, and G.

[0050] For example, Figure 3 This is a schematic diagram illustrating a method for generating particle size distribution data according to some embodiments of this disclosure. Figure 3 As shown, in some examples, the first generation model referred to in this disclosure includes a data generation network and a data filtering network. The data generation network is used to generate particle size distribution data. In this example, the input to the data generation network is particle size distribution data A, B, and C contained in the first particle size distribution dataset. The output is particle size distribution data B, C, D, E, F, and G. The data filtering network is used to filter out data from the output of the data generation network that is not covered by the input of the data generation network. The input to the data filtering network is the input and output of the data generation network, and the output is the particle size distribution data not covered by the input of the data generation network. In this example, the output of the data filtering network is particle size distribution data D, E, F, and G.

[0051] Of course, of course Figure 2 and Figure 3This is merely an illustrative example of a method for generating particle size distribution data, and not the only one.

[0052] In step S105, target particle size distribution data is determined from the first particle size distribution dataset and the second particle size distribution dataset, so that the material properties obtained after N sintering based on the target particle size distribution data meet the preset conditions, where N is an integer greater than or equal to 1.

[0053] In this context, "material performance meeting preset conditions" can be understood as the material performance reaching or exceeding the target performance. Material performance and target performance can be represented by parameters of one or more dimensions, such as the compaction density of the electrode material, discharge efficiency, and capacity ratio. In this case, "material performance meeting target performance" can be understood, for example, as satisfying at least one of the following conditions: the material's compaction density is greater than or equal to the target compaction density; the material's discharge efficiency is greater than or equal to the target discharge efficiency; and the material's capacity ratio is greater than or equal to the target capacity ratio.

[0054] This disclosure provides various methods for determining target particle size distribution data from a first particle size distribution dataset and a second particle size distribution dataset. For ease of understanding, exemplary methods are described below; however, it should be noted that these methods are merely exemplary and not the only methods.

[0055] Method 1 Figure 4 This is a schematic diagram illustrating a method for determining target particle size distribution data according to some embodiments of this disclosure. Figure 4 As shown, in some examples, a first optimization model is provided. This first optimization model searches from a particle size distribution dataset for target particle size distribution data that optimizes the material properties after N sintering cycles. In this example, the input to the first optimization model is particle size distribution data from a first and a second particle size distribution dataset, and the output is target particle size distribution data that ensures the material properties after N sintering cycles meet preset conditions. In this disclosure, the first optimization model can be trained based on any artificial intelligence model; its training method can be found in relevant technologies and will not be elaborated here.

[0056] Method 2 Figure 5 This is a schematic diagram illustrating another method for determining target particle size distribution data according to some embodiments of this disclosure. Figure 5As shown, in some examples, a second optimization model is provided. This second optimization model searches the particle size distribution dataset for particle size distribution data that enables the material properties after N sinterings to meet preset conditions. In this example, the input to the second optimization model is the performance parameters corresponding to the target material properties (e.g., compaction density, 1C (coulomb) discharge, 1C 3.2V capacity ratio, etc.), and the particle size distribution data from the first and second particle size distribution datasets. The output is the particle size distribution data (i.e., the target particle size distribution data) whose material properties after N sinterings are closest to the target material properties. In this disclosure, the second optimization model can be trained based on any artificial intelligence model; its training method can be found in relevant technologies and will not be elaborated here.

[0057] This disclosure, by acquiring a first particle size distribution dataset and inputting the particle size distribution data from the first particle size distribution dataset into a first generation model, can generate a second particle size distribution dataset containing particle size distribution data not covered by the first particle size distribution dataset, which is beneficial for expanding the combination space of particle size and ratio. On this basis, by determining the target particle size distribution data that can make the material properties after N sintering meet the preset conditions from the first and second particle size distribution datasets, it is beneficial for improving the accuracy of particle size distribution data optimization and the reliability of material property optimization. Furthermore, the material particle size distribution method provided in the embodiments of this disclosure does not rely on manual experience exploration, saving labor costs and improving efficiency.

[0058] Figure 6 This is a flowchart illustrating a method for obtaining a particle size distribution dataset according to some embodiments of this disclosure. Figure 6 As shown, in some embodiments, the first particle size distribution dataset can be obtained by the following method.

[0059] In step S601, at least one set of sintering conditions is obtained, including particle size distribution and sintering process parameters.

[0060] The sintering conditions referred to in this disclosure can be understood, for example, as the sintering conditions used when performing N sintering operations. For instance, when N equals 2, the sintering conditions referred to in this disclosure can be understood as the sintering conditions for two sintering operations, or the conditions for one and two sintering operations; when N equals 3, the sintering conditions referred to in this disclosure can be understood as the sintering conditions for three sintering operations, or the sintering conditions for one, two, and three sintering operations, and so on, without further listing here.

[0061] In some examples, when multiple sets of sintering conditions are obtained, these multiple sets of sintering conditions can be different. For example, the particle size distribution and / or sintering process parameters may be different. Different sintering process parameters include differences in some parameters or differences in all sintering process parameters.

[0062] The following examples illustrate how to obtain sintering conditions. However, it is important to note that the methods described below are not the only ways to obtain sintering conditions.

[0063] Taking N=2 as an example, when N=2, the objective is to determine the particle size distribution data that optimizes the material properties after two sintering processes. In this case, the sintering conditions described in this disclosure can be obtained through steps S1-S4.

[0064] In step S1, multiple sintering processes are used to sinter the precursor raw material (such as lithium iron phosphate) once to obtain the sintering product corresponding to each process.

[0065] In step S2, for each process, the primary sintering product is graded and ground using a grinding equipment to obtain a grinding product containing multiple particle sizes (such as 3μm, 8μm, 15μm, etc.).

[0066] The grinding and grading of the primary sintering products for each process can be the same or different.

[0067] In step S3, the mass or volume percentage of each particle size product is determined as needed to obtain the particle size distribution.

[0068] In step S4, the particle size distribution obtained in step S3 is combined with one or more pre-configured sintering process parameters as needed to obtain one or more sintering conditions.

[0069] In step S603, for each set of sintering conditions, the sintering conditions are input into the second generation model, and the particle size distribution information after N sinterings is predicted based on the second generation model.

[0070] The second generation model described in this embodiment is used to predict particle size distribution information after N sintering cycles. The particle size distribution information includes at least one of the following: 3D point cloud of particles, coordinates of particles of each size, mass percentage, volume percentage, etc.

[0071] In cases where the particle size distribution information includes the 3D point cloud of the particles, the mass percentage and / or volume percentage of each particle size in the sintered material can be calculated using methods such as segmented statistics and voxel method.

[0072] One example of the voxel method is to divide the 3D space containing the 3D point cloud into voxel grids of equal volume, count the total number of voxel grids occupied by particles of each size, and then calculate the mass percentage and / or volume percentage of particles of that size in the 3D point cloud based on the total number of voxel grids occupied by particles of that size and the total number of voxel grids.

[0073] For example, a segmented statistical method could be to use particle size as a reference, count the total number of particles of each particle size in the 3D point cloud, and calculate the volume percentage and / or mass percentage of particles of each particle size based on the total number of particles of each particle size and the total number of particles in the 3D point cloud.

[0074] In this embodiment of the disclosure, the input to the second generation model is the sintering conditions, and the output is the particle size distribution information obtained after sintering according to the sintering conditions.

[0075] In some examples, the second generative model can be any type of model, such as a neural network model, a machine learning model, or a deep neural network model, but it is not limited to the models listed here.

[0076] Taking a generative neural network as an example, in some examples, the second generative model referred to in the embodiments of this disclosure may include an encoder and a decoder.

[0077] The encoder input can be, for example, [d1, …, d]. n , s1, …, s n ,c1 …, c m n and m are integers greater than or equal to 1. The values ​​of n and m can be the same or different.

[0078] Where d1, …, d n This represents the particle size values ​​corresponding to each particle size classification. s1, …, s n This indicates the mass or volume percentage corresponding to each particle size fraction. d1 corresponds to s1, d2 to s2, and so on. n and s n Corresponding. c1 …, c m Let m be the parameters of the sintering process. The encoder is used to encode the input into a feature vector.

[0079] The decoder takes the feature vector from the encoder as input and outputs particle size distribution information.

[0080] In some examples, a second generative model can be trained based on pre-set sample data. The sample data may include sintering conditions and the actual particle size distribution information obtained under those sintering conditions, such as 3D point clouds, mass percentage, or volume percentage.

[0081] During training, various loss functions can be used to calculate the model's overall loss. For example, in some implementations, the reconstruction loss and KL divergence loss can be calculated separately. The overall loss is obtained by weighted summation of the reconstruction loss and KL divergence loss, and the model parameters are then adjusted based on this overall loss. The expression for calculating the overall loss can be represented as follows: L_total = λ1 L_recon + λ2 KL(q(z|X)||p(z)) (1) Where L_total is the total loss value, λ1 and λ2 are the weight values, L_recon is the reconstruction loss, and KL(q(z|X)||p(z)) is the KL divergence loss. q(z|X) is the posterior distribution of the encoder output (which can be understood as the vector corresponding to the particle size distribution information predicted by the model), and p(z) is the prior distribution (which can be understood as the vector corresponding to the true particle size distribution information).

[0082] In step S605, a first particle size distribution dataset is generated based on each set of sintering conditions and the particle size distribution information after N sinterings under each set of sintering conditions.

[0083] For example, based on sintering condition 'a', the predicted particle size distribution information is 'a1', and based on sintering condition 'b', the predicted particle size distribution information is 'b1'. Combining sintering condition 'a' and particle size distribution information 'a1' yields particle size gradation data A, and combining sintering condition 'b' and particle size distribution information 'b1' yields particle size gradation data B. Aggregating particle size gradation data A and particle size gradation data B into a single set yields the first particle size gradation data set.

[0084] This disclosure improves the efficiency of obtaining particle size distribution information by inputting the obtained sintering conditions into a second generation model and predicting the particle size distribution information after N sinterings through the second generation model. Thus, when generating the first particle size distribution dataset, the first particle size distribution dataset can be quickly generated based on at least one set of sintering conditions and the particle size distribution information corresponding to each set of sintering conditions.

[0085] Figure 7 This is a flowchart illustrating a method for determining target particle size distribution data according to some embodiments of this disclosure. Figure 7 As shown, in some examples, the target particle size distribution data can be determined using the following method.

[0086] In step S701, for the particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, a preset performance prediction model is used to predict the material properties corresponding to the particle size distribution data.

[0087] The material properties corresponding to the particle size distribution data refer to the material properties obtained after N sintering processes based on the particle size distribution data.

[0088] In this disclosure, a performance prediction model is trained to predict the material properties after sintering. The input to the performance prediction model is particle size distribution data, and the output is the material properties that can be obtained after N sinterings based on the particle size distribution data, such as compaction density, 1C discharge, and 1C 3.2V capacity ratio.

[0089] In the embodiments of this disclosure, the performance prediction model can be any type of model, such as a machine learning model, a deep learning model, or a neural network model.

[0090] In step S703, based on the material properties corresponding to each particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, the particle size distribution data whose material properties are closest to the target material properties is determined as the target particle size distribution data.

[0091] For example, in some cases, the particle size distribution data and their corresponding material properties, as well as the target material properties, can be input into a pre-defined comparison model. The comparison model then determines the similarity between the material properties corresponding to each particle size distribution data and the target material properties. The particle size distribution data corresponding to the material property with the highest similarity is then identified as the target particle size distribution data. The training method for the comparison model can be found in relevant technologies and will not be elaborated upon here.

[0092] This disclosure uses a pre-defined performance prediction model to predict the material properties corresponding to the particle size distribution data contained in the first and second particle size distribution datasets. This saves time in obtaining material properties experimentally and improves the efficiency of acquiring material properties corresponding to particle size distribution data. Based on this, and using the material properties corresponding to each particle size distribution data in the first and second particle size distribution datasets, the particle size distribution data whose material properties are closest to the target material properties can be quickly obtained, and this particle size distribution data can be identified as the target particle size distribution data.

[0093] In some embodiments of this disclosure, the target material performance can also be used as the optimization objective, and a Bayesian optimization algorithm can be employed to determine the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset. For example, in some examples, the first particle size distribution dataset and the second particle size distribution dataset can be merged into a candidate dataset. Initially, a set of particle size distribution data is randomly selected from the candidate dataset and input into the performance prediction model. The performance prediction model predicts the material performance corresponding to the particle size distribution data. Based on this, the difference between the material performance corresponding to the particle size distribution data and the target material performance is determined, and the next particle size distribution data is determined from the candidate dataset accordingly. Performance prediction is then performed. Based on the difference between the predicted material performance and the target material performance, the next particle size distribution data is re-determined, and so on, until particle size distribution data corresponding to material performance that is close to the target material performance is obtained. It should be noted that this is only a brief description of the application of the Bayesian optimization algorithm in this disclosure. Based on this, the specific execution method of the Bayesian optimization algorithm can be found in related technologies and will not be repeated here.

[0094] This disclosure takes the target material properties as the optimization objective and adopts a Bayesian optimization algorithm to quickly determine the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset, thereby improving the efficiency of determining the target particle size distribution data.

[0095] Figure 8 This is a block diagram illustrating a material particle size distribution device according to some embodiments of the present disclosure. (Refer to...) Figure 8 The material particle size distribution device 800 includes: The acquisition module 801 is configured to acquire a first particle size distribution dataset, which includes at least one set of particle size distribution data.

[0096] The generation module 802 is configured to input the particle size distribution data in the first particle size distribution dataset into the first generation model, and generate a second particle size distribution dataset based on the first generation model, wherein the particle size distribution data in the second particle size distribution dataset is not included in the first particle size distribution dataset.

[0097] The determining module 803 is configured to determine target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset, so that the material properties obtained after N sintering based on the target particle size distribution data meet preset conditions, where N is an integer greater than or equal to 1.

[0098] In some exemplary embodiments of this disclosure, the particle size distribution data in the first particle size distribution dataset and the second particle size dataset includes the following data: particle size ratio, sintering process parameters, and particle size distribution information after N sintering cycles.

[0099] In some exemplary embodiments of this disclosure, the acquisition module 801 is configured to: acquire at least one set of sintering conditions, the sintering conditions including particle size distribution and sintering process parameters; for each set of sintering conditions, input the sintering conditions into a second generation model, predict particle size distribution information after N sinterings based on the second generation model; and generate a first particle size distribution dataset based on each set of sintering conditions and the particle size distribution information after N sinterings under each set of sintering conditions.

[0100] In some exemplary embodiments of this disclosure, the determining module 803 is configured to: for the particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, use a preset performance prediction model to predict the material properties corresponding to the particle size distribution data, wherein the material properties refer to the material properties obtained after N sintering based on the particle size distribution data; and based on the material properties corresponding to each particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, determine the particle size distribution data whose material properties are closest to the target material properties as the target particle size distribution data.

[0101] In some exemplary embodiments of this disclosure, the determining module 803 is configured to: determine target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset using a Bayesian optimization algorithm with the target material properties as the optimization objective.

[0102] Regarding the above Figure 8 The specific manner in which the various modules perform their operations in the embodiments of the apparatus have been described in detail in the embodiments relating to the method, and will not be elaborated upon here.

[0103] In some embodiments, an electronic device is also provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to: implement the method of any of the above method embodiments.

[0104] Figure 9 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Figure 9 As shown, the electronic device 900 is manifested in the form of a general-purpose computing device. The components of the electronic device 900 may include, but are not limited to: at least one processing unit 910, at least one storage unit 920, and a bus 930 connecting different system components (including storage unit 920 and processing unit 910).

[0105] The storage unit stores program code that can be executed by the processing unit 910, causing the processing unit 910 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 910 can perform the method of any of the above-described method embodiments.

[0106] Storage unit 920 may include readable media in the form of volatile storage units, such as random access memory (RAM) 9201 and / or cache memory 9202, and may further include read-only memory (ROM) 9203.

[0107] The storage unit 920 may also include a program / utility 9204 having a set (at least one) program module 9205, such program module 9205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0108] Bus 930 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0109] Electronic device 900 can also communicate with one or more external devices 940 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 900, and / or with any device that enables electronic device 900 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 950. Furthermore, electronic device 900 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 960. As shown, network adapter 960 communicates with other modules of electronic device 900 via bus 930. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0110] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0111] This disclosure also provides a non-transitory computer-readable storage medium, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is able to perform the methods in any of the above method embodiments.

[0112] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0113] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0114] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0115] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for particle size distribution of materials, characterized in that, include: Obtain a first particle size distribution dataset, wherein the first particle size distribution dataset includes at least one set of particle size distribution data; The particle size distribution data in the first particle size distribution dataset is input into the first generation model, and a second particle size distribution dataset is generated based on the first generation model. The particle size distribution data in the second particle size distribution dataset is not included in the first particle size distribution dataset. From the first particle size distribution dataset and the second particle size distribution dataset, a target particle size distribution data is determined so that the material properties obtained after N sinterings based on the target particle size distribution data meet a preset condition, where N is an integer greater than or equal to 1.

2. The method according to claim 1, characterized in that, The particle size distribution data in the first particle size distribution dataset and the second particle size dataset include the following data: particle size ratio, sintering process parameters, and particle size distribution information after N sintering cycles.

3. The method according to claim 2, characterized in that, The process of obtaining the first particle size distribution dataset includes: Obtain at least one set of sintering conditions, the sintering conditions including particle size distribution and sintering process parameters; For each set of sintering conditions, the sintering conditions are input into the second generation model, and the particle size distribution information after N sinterings is predicted based on the second generation model. Based on the sintering conditions of each group and the particle size distribution information after N sinterings under each group of sintering conditions, the first particle size distribution dataset is generated.

4. The method according to claim 2, characterized in that, Determining the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset includes: For the particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, a preset performance prediction model is used to predict the material properties corresponding to the particle size distribution data. The material properties refer to the material properties obtained after N sintering based on the particle size distribution data. Based on the material properties corresponding to each particle size distribution data in the first particle size distribution dataset and the second particle size distribution dataset, the particle size distribution data whose material properties are closest to the target material properties is determined as the target particle size distribution data.

5. The method according to claim 2, characterized in that, Determining the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset includes: With the target material properties as the optimization objective, a Bayesian optimization algorithm is used to determine the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset.

6. A material particle size distribution device, characterized in that, include: The acquisition module is configured to acquire a first particle size distribution dataset, wherein the first particle size distribution dataset includes at least one set of particle size distribution data; The generation module is configured to input the particle size distribution data in the first particle size distribution dataset into the first generation model, and generate a second particle size distribution dataset based on the first generation model. The particle size distribution data contained in the second particle size distribution dataset is not included in the first particle size distribution dataset. The determining module is configured to determine target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset, such that the material properties obtained after N sinterings based on the target particle size distribution data meet preset conditions, where N is an integer greater than or equal to 1.

7. The apparatus according to claim 6, characterized in that, The particle size distribution data in the first particle size distribution dataset and the second particle size dataset include the following data: particle size ratio, sintering process parameters, and particle size distribution information after N sintering cycles.

8. The apparatus according to claim 7, characterized in that, The determining module is configured to: use the target material properties as the optimization objective and employ a Bayesian optimization algorithm to determine the target particle size distribution data from the first particle size distribution dataset and the second particle size distribution dataset.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured as follows: Implement the method according to any one of claims 1 to 5.

10. A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method of any one of claims 1 to 5.

Images