Doped cluster ground state structure optimization method and system based on machine learning

Abstract

The invention relates to the technical field of machine learning, and discloses a doped cluster ground state structure optimization method and system based on machine learning, and the method comprises the steps: collecting and optimizing original doped cluster data as a doped cluster standard data set, and dividing the doped cluster standard data set into a training set, a verification set and a test set; constructing a doped cluster potential energy surface fitting machine learning network by adding an interatomic interaction attention mechanism in an original SchNet, and training by using a training set and a verification set; inputting the test set into the trained doping cluster potential energy surface fitting machine learning network, and obtaining and outputting a final doping cluster potential energy surface; and performing structure search and optimization on the final doped cluster potential energy surface by using an artificial bee colony algorithm, and continuously updating the structure until a ground state structure with the lowest energy is found. According to the method, the fitting accuracy of the potential energy surface is improved, the calculation benefit is improved, and good transportability is achieved.

Description

Technical Field

[0001] The present invention relates to the field of machine learning technology, and in particular to a method and system for optimizing the ground state structure of a doping cluster based on machine learning. Background Art

[0002] Clusters are multinuclear aggregates between atoms, molecules and macroscopic matter. They have a certain atomic composition and chemical structure and represent the primary ecology of condensed matter. The electronic structure and optical properties of clusters will change significantly with size. This change law can relate the macroscopic properties to the microscopic structure of matter, which is of great significance for a deep understanding of the law of material transformation. Clusters are the basic units for building new functional materials and an important way to prepare nanomaterials with unique properties. By assembling clusters with specific functions into ordered structures, high-performance optical, electrical or magnetic materials can be obtained. By changing the reaction conditions, ligand types and other factors in the cluster synthesis process, the porosity, adsorption performance, catalytic activity and other important properties of the material can be finely controlled to meet different application requirements. The microstructure of clusters is difficult to obtain by a single experimental method, and theoretical calculations must be carried out. The determination of the ground state structure of clusters requires a global search on the high-dimensional potential energy surface, which is one of the most basic and important topics in the field of cluster research. For a cluster containing N atoms, the number of local minima on the potential energy surface increases exponentially with the dimension 3N. As the size of clusters increases, their potential energy surface becomes extremely complex, so it becomes very difficult to find the ground state structure of clusters. How to develop an efficient global optimization model is an urgent problem to be solved in the field of clusters.

[0003] The commonly used methods to accurately describe the potential energy surface of larger systems are the ab initio method and density functional theory method in quantum mechanics calculations. However, this increases exponentially with the increase in the size of the system. Even with some approximations, a large amount of computing resources are still required when dealing with larger systems. With the development of artificial intelligence technology, machine learning is undoubtedly a very promising tool. Summary of the invention

[0004] The purpose of the present invention is to solve the problems in the prior art.

[0005] The technical solution adopted by the present invention to solve the technical problem is: to provide a method for optimizing the ground state structure of doping clusters based on machine learning, comprising the following steps:

[0006] Collecting original doping cluster data and optimizing them as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping clusters; and dividing the doping cluster standard data set into a training set, a verification set, and a test set;

[0007] By adding the interatomic interaction attention mechanism to the original SchNet, a doping cluster potential energy surface fitting machine learning network is constructed, and the training set and validation set are used for training; the test set is input into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface;

[0008] The artificial bee colony algorithm is used to perform structural search and optimization on the potential energy surface of the final doped cluster, and the structure is continuously updated until the lowest energy ground state structure is found.

[0009] Preferably, the original doping cluster data are collected using Gaussian 09, PBE0 hybrid density functional theory method and 3-21G basis set.

[0010] Preferably, the structural information of the doped cluster is the molecular formula and the three-dimensional Cartesian coordinates of the molecule.

[0011] Preferably, the doped clusters in the doped cluster standard data set are all boron atom clusters doped with one selenium atom; the doped clusters are divided into nine types according to the number of boron atoms, and the number of boron atoms ranges from 16 to 24.

[0012] Preferably, the collecting of original doping cluster data and optimization thereof comprises:

[0013] Calculate the partial derivative of the system energy with respect to each atomic position as the energy gradient of the system;

[0014] Based on the energy gradient of the system, the atomic positions are gradually moved to update the atomic coordinates, so that the energy of the system is gradually reduced, the optimized cluster structure and energy are obtained, and the doped cluster data set is obtained.

[0015] Preferably, the process of inputting the test set into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface comprises the following steps:

[0016] Input the molecular formula of the cluster, the molecular three-dimensional Cartesian coordinates, and the energy value as a label;

[0017] The molecular formula is encoded according to the periodic table to obtain the atomic number, and the atomic number is embedded into a feature matrix;

[0018] The atomic distance is obtained by subtracting the absolute value of the atomic coordinates, and the atomic distance is expanded by radial basis function to expand the dimension;

[0019] The feature matrix and the atomic distances after the expanded dimension enter the interatomic interaction attention mechanism for feature extraction and parameter update, and the feature information containing the structure-energy correspondence network is obtained as the final doping cluster potential energy surface output.

[0020] Preferably, the feature matrix and the interatomic distances after the expanded dimension enter the interatomic interaction attention mechanism for feature extraction and parameter update, including the following steps:

[0021] Feature extraction: The feature matrix is ​​multiplied element by element with the atomic distance after the expanded dimension through a filter generation function module to output the node features.

[0022] Feature update: construct a learnable shared matrix to reconstruct the linear transformation of node features, and use a feedforward neural network to build a shared attention function to smoothly normalize the node features and update the node features;

[0023] Feature aggregation, finally, the updated node features and the extracted node features are aggregated through the residual network to obtain the feature information of the structure-energy correspondence network.

[0024] Preferably, the training is performed using a training set and a validation set, the optimization function adopted is a stochastic gradient descent function, and the mean absolute error, root mean square error, optimal convergence training cycle and optimal convergence loss value are used as evaluation indicators for the improved SchNet doped cluster potential energy surface fitting machine learning network.

[0025] Preferably, the use of an artificial bee colony algorithm to perform structure search and optimization on the potential energy surface of the final doping cluster comprises the following steps:

[0026] Initialization step: using artificial bee colony algorithm to search on the potential energy surface of the final doping cluster, initialize the bee colony and its position and calculate the fitness;

[0027] In the iterative step, the swarm position is updated through local search and global random search is performed during the exploration phase;

[0028] Calculation step, repeat the iteration step until the maximum number of iterations and fitness threshold are reached, record the best position in each iteration, use the final doped cluster potential energy surface to calculate the energy of each structure, and finally output the lowest energy ground state doped cluster structure and energy.

[0029] The present invention also provides a doping cluster ground state structure optimization system based on machine learning, comprising the following steps:

[0030] A data processing module collects original doping cluster data and optimizes it as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping cluster; and divides the doping cluster standard data set into a training set, a verification set and a test set;

[0031] A model training module, which constructs a doping cluster potential energy surface fitting machine learning network by improving SchNet and uses a training set and a validation set for training; the test set is input into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface; the improved SchNet refers to adding an interatomic interaction attention mechanism to the feature extraction and convolution filtering network modules of the original SchNet;

[0032] The structure search module uses the artificial bee colony algorithm to perform structure search and optimization on the potential energy surface of the final doping cluster, and continuously updates the structure until the ground state structure with the lowest energy is found.

[0033] The present invention has the following beneficial effects:

[0034] (1) Improve the accuracy of potential energy surface fitting. The machine learning model can adjust the framework according to the specific task, which is conducive to improving the accuracy of potential energy surface fitting, thereby ensuring that the searched ground state structure is more experimentally meaningful.

[0035] (2) Improve computational efficiency. Applying machine learning to the search for the ground state structure of doped clusters can reduce the computing resources occupied and reduce computing costs. The model only needs to be fully trained in the early stage to search for the ground state structure of doped clusters for a long time.

[0036] (3) Good scalability. Machine learning models can be fine-tuned or retrained by inputting new data sets, which can expand to higher accuracy and larger systems.

[0037] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments, but the present invention is not limited to the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A method step diagram of an embodiment of the present invention;

[0039] Figure 2 It is a schematic diagram of a flow chart of an embodiment of the present invention;

[0040] Figure 3 It is a machine learning network structure diagram of the doping cluster potential energy surface fitting according to an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of a process of performing ground state structure search using an artificial bee colony algorithm according to an embodiment of the present invention;

[0042] Figure 5 4 is a system structure diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0043] See also Figure 1 and Figure 2The method step diagram and flow chart of an embodiment of the present invention are shown, which include the following steps:

[0044] S101, collecting original doping cluster data and optimizing them as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping clusters; and dividing the doping cluster standard data set into a training set, a verification set, and a test set;

[0045] S102, constructing a doping cluster potential energy surface fitting machine learning network by adding an interatomic interaction attention mechanism to the original SchNet, and training it using a training set and a validation set; inputting the test set into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface;

[0046] S103, using the artificial bee colony algorithm to perform structure search and optimization on the potential energy surface of the final doping cluster, continuously updating the structure until the lowest energy ground state structure is found.

[0047] Specifically, the embodiment of the present invention uses Gaussian 09 to perform an unbiased search for cluster structures at the PBE0 / 3-21G computing level to construct a standard doped cluster data set. The SchNet machine learning model is improved, and the cluster is then passed into the improved SchNet machine learning model, and feature extraction is performed through the atomic interaction attention mechanism to obtain the potential energy surface of the corresponding cluster. The potential energy surface is deployed in the search process of the cluster ground state structure, and the ground state structure of the doped cluster is finally obtained. The following is mainly divided into four parts for detailed description.

[0048] (1) Cluster data set collection: Nine doped clusters with boron clusters as the main body were selected, random structure search was performed and initial energy calculation was performed by Gaussian 09 to establish the original doped cluster data set. Then the original clusters were optimized at the PBE0 / 3-21G calculation level, and the original doped cluster data set was expanded. For the construction of the standard doped cluster data set, the original doped cluster structure was generated by random search and density functional theory calculation was performed at the PBE0 / 3-21G level. All energy calculations and geometric optimizations of the doped clusters were performed using the Gaussian 09 program. The original doped cluster data was optimized by calculating the energy gradient of the system, that is, the partial derivative of the system energy with respect to each atomic position. This reflects the direction of the force and indicates how the atoms should move to reduce the total energy of the system. If all gradients are zero, it means that the system is at the minimum energy point; the atomic positions are gradually moved to update the atomic coordinates, so that the energy of the system is gradually reduced, and the optimized cluster structure and energy are obtained, and the standard doped cluster data set is obtained. By restricting the filters of the continuous filter convolution layer to be rotationally invariant, the feature prior is used to replace the data augmentation process to reduce the amount of calculation. During the model training process, the dataset is randomly divided into 8:1:1 ratios, 80% as the training set, 10% as the validation set, and the remaining 10% as a test set alone.

[0049] The embodiment of the present invention selects 9 representative doping clusters, randomly generates target doping cluster structures, and uses Gaussian 09 to calculate energy and optimization, and produces a standard doping cluster data set. The density functional theory method and basis set adopted are PBE0 / 3-21G, which takes both accuracy and calculation speed into consideration. This data set not only includes low-energy structures, but also includes high-energy structures compared to most open source data sets, so that the machine learning model can obtain a relatively complete structure-energy relationship and more accurately fit the overall picture of the potential energy surface of the doping cluster.

[0050] (2) Feature extraction algorithm: The traditional machine learning potential energy function aims to use machine learning methods to approximate the mapping between material structure and energy to simulate the atomic interaction potential energy model, effectively avoiding the errors caused by the use of empirical parameters in traditional empirical force fields and the problem of low portability. Although machine learning methods for constructing potential energy surfaces have made progress, there are still many problems, mainly including the following three aspects: processing high-dimensional data and feature selection strategies, the construction of molecular descriptors, and scarce high-quality cluster databases. In response to the above problems, the present invention proposes an atomic interaction attention mechanism, which is introduced into the feature extraction and convolutional filter network modules of the original SchNet to improve the feature capture of the atomic environment. By constructing an end-to-end machine learning model, the regression task of atomic feature vectors to energy is realized to save computing costs and improve the portability of the model, and reduce human intervention and debugging of descriptor hyperparameters.

[0051] The attention mechanism is a computational model that makes the model more efficient and flexible when processing sequence data. The core idea of ​​the attention mechanism is to let the model focus on the most relevant part of the current task when processing input data, thereby improving the efficiency and performance of the model. The attention mechanism greatly improves the performance of the model when processing sequence data by adaptively focusing on the important information of the input sequence. The atomic interaction attention mechanism is constructed as follows: first, the characteristic relationship between atoms and bonds is constructed by fusing node information with the high-dimensional extended dynamic filtering network of SchNet, while avoiding the computational overhead caused by dense matrix multiplication; then the importance of different atomic interactions is modeled using the graph attention method; finally, through the idea of ​​the residual network, the information of nodes and edges, nodes and nodes are aggregated together at the same time, and the feature information is updated for output. In view of the problem that the weight of doping atoms in the machine learning model is more obvious, and the filter convolution module used does not pay enough attention to the doping atom features, the embodiment of the present invention proposes an atomic interaction attention mechanism and adds it to the feature extraction of SchNet. It aggregates the information of nodes and edges, nodes and nodes in the molecular graph at the same time, so that the machine learning model can better learn atomic features and atomic environments. The interatomic interaction attention mechanism can help the network better focus on the feature representation of doped atoms, thereby improving the accuracy of potential energy surface fitting.

[0052] (3) Parameter update. The embodiment of the present invention uses the SMU algorithm to smooth the activation function to reduce the overfitting caused by high-energy abnormal structures. When training machine learning models, problems such as gradient vanishing and overfitting are often encountered. Especially when dealing with high-energy abnormal structures, high-energy abnormal structures usually cause the model to overfit the training data, thereby affecting the generalization ability of the model. The activation function is responsible for introducing nonlinearity and helping the model learn complex mapping relationships. It plays an important role in the parameter update process. The SMU algorithm smoothes the activation function through maximum unit approximation to obtain continuity and differentiability, thereby optimizing the gradient update process and maintaining the continuity and smoothness of the potential energy surface.

[0053] The SMU algorithm is an activation function used in machine learning, which can approximate the maximum value operation from the smooth approximation of the absolute value function, thereby providing a smoother gradient update. The activation function after the SMU algorithm is smoothed is smooth and differentiable in the entire domain, which makes it easier to calculate the gradient during back propagation and optimization, thereby providing a more stable parameter update process and accelerating convergence; compared with the traditional activation function, the activation function after the SMU algorithm is smoothed can still maintain a non-zero gradient when the input value is large or small, which helps to alleviate the gradient disappearance problem; due to its smoothness, the SMU has a certain ability to suppress noise in the input data, and compared with the traditional activation function, the robustness of the model is significantly improved. The embodiment of the present invention uses the SMU algorithm idea to smooth the activation process of the interaction between nodes, specifically constructing a learnable shared matrix to reconstruct the linear transformation of the input node features, and using a feedforward neural network to construct a shareable attention function, and smooth normalization to update the node features. Finally, the two features obtained through the idea of ​​the residual network are aggregated to output feature information.

[0054] For details, see Figure 3 As shown, the embodiment of the present invention obtains a doping cluster potential energy surface fitting machine learning network by improving SchNet, and its internal execution process is:

[0055] Input the molecular formula (i.e. the atoms contained in the cluster), molecular coordinates (three-dimensional Cartesian coordinates) and energy values ​​as labels; the atomic number encodes the molecular formula in the order of the periodic table, such as CO2: [6, 8, 8]; the interatomic distance is obtained by subtracting the absolute value of the atomic coordinates; the atomic number is embedded through features to obtain the feature matrix; the interatomic distance is expanded through radial basis function to expand the dimension. If the interatomic distance is 4 and the sampling interval is 0.5, it is expanded to [0, 0.5, 1, …, 3.5, 4] and Gaussian transformation is performed so that the closer the neighboring atoms are to the central atom, the greater the weight they occupy, and finally [e 0 , e- 0.5 , e -1 , …,e -3.5, e -4 ]; The feature matrix obtained by feature embedding of the atomic number enters the inter-atomic interaction attention mechanism. In subsequent convolution filtering and other calculations, it is element-by-element multiplied with the atomic distance after the expansion dimension to obtain edge features.

[0056] The embodiment of the present invention aims at the overfitting problem caused by high-energy abnormal data in doping clusters to the machine learning model, and proposes to use the maximum performance smoothing unit to smooth the gradient update process during the back propagation process. By smoothing the absolute value function, the smoothing formula of the corresponding activation function is fitted, so that the gradient can be smoothly transmitted during back propagation, reducing unnecessary local minima and discontinuities. It effectively avoids the overfitting caused by high-energy abnormal data and improves the stability of numerical calculation and the reliability of the model.

[0057] (4) Ground state structure search: Ground state structure search is the ultimate goal of cluster potential energy surface fitting. This requires not only the selection of appropriate methods such as classical molecular dynamics methods and / or density functional theory to calculate the energy, but also the selection of appropriate optimization algorithms, taking into account both global optimization search and local optimization. For the initial structure to be optimized, initial guesses are usually generated through literature, known structures or simple configurations, or a large number of initial structures are randomly generated for low-computational level pre-optimization search and higher-precision secondary optimization search, which usually consumes a lot of time and computing resources. The ground state structure search process adopts a scheme that combines the two initial structure acquisition methods of initial guessing and random generation to ensure a more accurate ground state structure and set reasonable energy and force convergence standards. The embodiment of the present invention uses an artificial bee colony algorithm to perform optimization search on the potential energy surface of the doped cluster of the improved SchNet, replacing the traditional low-computational level pre-optimization search and higher-precision secondary optimization search, which greatly saves computing resources and saves computing time; then the results of the optimization search are statistically analyzed, and the configuration with the lowest energy is selected as the final result. Finally, the lowest energy ground state doped cluster structure and energy are output, and the stability of the obtained ground state structure is verified by vibration analysis and other methods to complete the ground state structure search of the doped cluster. For the specific process, see Figure 4 As shown, it includes: initializing the swarm and its position, and calculating the fitness; then updating the swarm position through local search, and performing global random search in the exploration phase; recording the best position in each iteration until the maximum number of iterations and the fitness threshold are reached, and using the improved SchNet's doped cluster potential energy surface to calculate the energy of each structure, reducing the computational burden, and finally outputting the lowest energy ground state doped cluster structure and energy.

[0058] According to the task characteristics of doped clusters with more complex potential energy surfaces, the embodiment of the present invention uses an artificial bee colony algorithm to replace traditional annealing simulation and genetic algorithm. The artificial bee colony algorithm evaluates the fitness of each bee according to the potential energy value, which is used to select employed bees and scout bees. Employed bees randomly select other bees for information exchange based on their current position and fitness. If the fitness of the new position is higher, the bee position is updated. In order to prevent the algorithm from falling into a local optimum, employed bees are converted into scout bees when they iterate to a threshold. When looking for the lowest energy structure of a cluster, the artificial bee colony algorithm has the advantage of being able to perform global search and local search at the same time, providing a more effective means of searching for ground state structures.

[0059] For details, see Figure 5 FIG. 1 is a system structure diagram of an embodiment of the present invention, including:

[0060] The data processing module 501 collects and optimizes the original doping cluster data as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping cluster; and divides the doping cluster standard data set into a training set, a verification set, and a test set;

[0061] Model training module 502, constructing a doping cluster potential energy surface fitting machine learning network by adding an interatomic interaction attention mechanism to the original SchNet, and training using a training set and a validation set; inputting the test set into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface;

[0062] The structure search module 503 uses an artificial bee colony algorithm to perform structure search and optimization on the potential energy surface of the final doping cluster, and continuously updates the structure until the ground state structure with the lowest energy is found.

[0063] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the ground state structure of doping clusters based on machine learning, characterized in that: The following steps are involved: Collecting original doping cluster data and optimizing them as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping clusters; and dividing the doping cluster standard data set into a training set, a verification set, and a test set; By adding the interatomic interaction attention mechanism to the original SchNet, a doping cluster potential energy surface fitting machine learning network is constructed, and the training set and validation set are used for training; the test set is input into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface; The artificial bee colony algorithm is used to perform structural search and optimization on the potential energy surface of the final doped cluster, and the structure is continuously updated until the lowest energy ground state structure is found.

2. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 1, characterized in that: The original doping cluster data are collected using Gaussian 09, PBE0 hybrid density functional theory method and 3-21G basis set.

3. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 1, characterized in that: The structural information of the doping cluster is the molecular formula and the three-dimensional Cartesian coordinates of the molecule.

4. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 1, characterized in that: The doped clusters in the doped cluster standard data set are all boron atom clusters doped with one selenium atom; the doped clusters are divided into nine types according to the number of boron atoms, and the number of boron atoms ranges from 16 to 24.

5. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 3, characterized in that: The original doping cluster data is collected and optimized, and the optimization process includes: Calculate the partial derivative of the system energy with respect to each atomic position as the energy gradient of the system; Based on the energy gradient of the system, the atomic positions are gradually moved to update the atomic coordinates, so that the energy of the system is gradually reduced, the optimized cluster structure and energy are obtained, and the doped cluster data set is obtained.

6. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 1, characterized in that: The process of inputting the test set into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface includes the following steps: Input the molecular formula of the cluster, the molecular three-dimensional Cartesian coordinates, and the energy value as a label; The molecular formula is encoded according to the periodic table to obtain the atomic number, and the atomic number is embedded into a feature matrix; The atomic distance is obtained by subtracting the absolute value of the atomic coordinates, and the atomic distance is expanded by radial basis function to expand the dimension; The feature matrix and the atomic distances after the expanded dimension enter the interatomic interaction attention mechanism for feature extraction and parameter update, and the feature information containing the structure-energy correspondence network is obtained as the final doping cluster potential energy surface output.

7. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 6, characterized in that: The feature matrix and the interatomic distance after the expansion dimension enter the interatomic interaction attention mechanism for feature extraction and parameter update, including the following steps: Feature extraction: The feature matrix is ​​multiplied element by element with the atomic distance after the expanded dimension through a filter generation function module to output the node features. Feature update: construct a learnable shared matrix to reconstruct the linear transformation of node features, and use a feedforward neural network to build a shared attention function to smoothly normalize the node features and update the node features; Feature aggregation, finally, the updated node features and the extracted node features are aggregated through the residual network to obtain the feature information of the structure-energy correspondence network.

8. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 6, characterized in that: The training is performed using a training set and a validation set, the optimization function adopted is a stochastic gradient descent function, and the mean absolute error, root mean square error, optimal convergence training cycle and optimal convergence loss value are used as evaluation indicators of the improved SchNet doped cluster potential energy surface fitting machine learning network.

9. The method for optimizing the ground state structure of doping clusters based on machine learning according to claim 6, characterized in that: The method of using an artificial bee colony algorithm to search and optimize the structure on the potential energy surface of the final doping cluster comprises the following steps: Initialization step: using artificial bee colony algorithm to search on the potential energy surface of the final doping cluster, initialize the bee colony and its position and calculate the fitness; In the iterative step, the swarm position is updated by local search and global random search is performed in the exploration phase; Calculation step, repeat the iteration step until the maximum number of iterations and fitness threshold are reached, record the best position in each iteration, use the final doped cluster potential energy surface to calculate the energy of each structure, and finally output the lowest energy ground state doped cluster structure and energy.

10. A doping cluster ground state structure optimization system based on machine learning, comprising the following steps: A data processing module collects original doping cluster data and optimizes it as a doping cluster standard data set, wherein the original doping cluster data includes structural information and energy information of the doping cluster; and divides the doping cluster standard data set into a training set, a verification set and a test set; A model training module, which constructs a doping cluster potential energy surface fitting machine learning network by improving SchNet and uses a training set and a validation set for training; the test set is input into the trained doping cluster potential energy surface fitting machine learning network to obtain and output the final doping cluster potential energy surface; the improved SchNet refers to adding an interatomic interaction attention mechanism to the feature extraction and convolution filtering network modules of the original SchNet; The structure search module uses the artificial bee colony algorithm to perform structure search and optimization on the potential energy surface of the final doping cluster, and continuously updates the structure until the ground state structure with the lowest energy is found.

Images