Method, apparatus and storage medium for screening of laser glass compositions

Abstract

The application relates to the technical field of glass materials, and provides a laser glass component screening method and device, computer equipment and a storage medium. A plurality of component data of laser glass containing active ions is acquired, molecular dynamics simulation is performed on each component data, and structure property data of a local region of the active ions corresponding to each component data is calculated according to obtained glass structure data corresponding to each component data; laser performance data corresponding to each component data is acquired, and a structure property-laser performance data set is constructed according to the structure property data of the local region of the active ions corresponding to the same component data and the corresponding laser performance data; model training is performed by using the structure property-laser performance data set, a laser performance prediction model is obtained, and component data with excellent laser performance in a component space of the laser glass containing the active ions is screened from the component space. The laser performance prediction model constructed by the method has high releaseability.

Description

Technical Field

[0001] The present application relates to the technical field of glass materials, and in particular to a method, device, computer equipment, and storage medium for screening laser glass components. Background Art

[0002] Glass, one of the most important and influential materials in human history, is widely used in various fields, including daily life, safety protection, and national defense. Laser glass, a key laser gain material, is a core component in the construction of solid-state lasers and fiber lasers. With so many different types of laser glass, it is necessary to select laser glass components with superior performance to produce laser glass with excellent laser performance.

[0003] The existing screening method is to construct a mathematical model based on the relationship between the components and properties of laser glass to screen the components of laser glass, but this screening method has poor interpretability. Summary of the Invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment, storage medium and computer program product for screening laser glass components to address the above technical problems.

[0005] The present application provides a method for screening laser glass components, the method comprising:

[0006] Acquire multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain glass structure data corresponding to each component data;

[0007] According to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained;

[0008] Obtain the laser performance data corresponding to each component data, and construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data;

[0009] The structural property-laser performance dataset is used for model training to obtain a laser performance prediction model;

[0010] By using the laser performance prediction model, the component data with excellent laser performance in the component space of laser glass containing activated ions are screened out.

[0011] The present application provides a device for screening laser glass components, the device comprising:

[0012] Glass structure analysis module, used to obtain multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain the glass structure data corresponding to each component data;

[0013] The structural property analysis module obtains the structural property data of the activated ion localization corresponding to each component data based on the glass structure data corresponding to each component data;

[0014] A data set construction module is used to obtain the laser performance data corresponding to each component data, and to construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data;

[0015] A model training module is used to train the model using the structural property-laser performance dataset to obtain a laser performance prediction model;

[0016] The glass component screening module is used to use the laser performance prediction model to screen out component data with excellent laser performance from the component space of laser glass containing activated ions.

[0017] The present application provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0018] Acquire multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain glass structure data corresponding to each component data;

[0019] According to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained;

[0020] Obtain the laser performance data corresponding to each component data, and construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data;

[0021] The structural property-laser performance dataset is used for model training to obtain a laser performance prediction model;

[0022] By using the laser performance prediction model, the component data with excellent laser performance in the component space of laser glass containing activated ions are screened out.

[0023] The present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the following steps are implemented:

[0024] Acquire multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain glass structure data corresponding to each component data;

[0025] According to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained;

[0026] Obtain the laser performance data corresponding to each component data, and construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data;

[0027] The structural property-laser performance dataset is used for model training to obtain a laser performance prediction model;

[0028] By using the laser performance prediction model, the component data with excellent laser performance in the component space of laser glass containing activated ions are screened out.

[0029] The above-mentioned screening method, device, computer equipment and storage medium for laser glass components generate corresponding glass structure data based on the glass component data of the laser glass and molecular dynamics simulation, and obtain the corresponding structural property data of the local activated ions based on the laser glass structure data. Finally, the laser performance data corresponding to each component data are used to construct a structural property-laser performance data set, which fully takes into account the decisive role of the structural properties of the local activated ions in the laser glass on the laser performance. In addition, the structural property-laser performance data set is used for model training to obtain a laser performance prediction model; the laser performance prediction model is used to screen out component data with excellent laser performance in the component space from the component space of the laser glass containing the activated ions. Based on the constructed component-structure-performance relationship of the laser glass, the laser performance prediction model obtained by training is used to predict the laser performance of a large range of the component space of the laser glass, from which laser glass components with excellent laser performance are screened. Furthermore, the laser performance prediction model can also be used to infer the glass structure with the target laser performance based on the target laser performance. Therefore, the screening method for laser glass components of the present application is highly interpretable. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 1 is a schematic flow chart of a method for screening laser glass components in one embodiment;

[0031] Figure 2 A schematic diagram of the structure and working process of a laser performance prediction model in one embodiment;

[0032] Figure 3 is the mean square error corresponding to the intermediate laser performance prediction model with different numbers of intermediate layer nodes in one embodiment;

[0033] Figure 4 Result analysis diagram of the predicted value and experimental value of the intermediate laser performance prediction model corresponding to the minimum mean square error in one embodiment;

[0034] Figure 5 1 is a block diagram of a device for screening laser glass components in one embodiment;

[0035] Figure 6 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0037] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0038] In one embodiment, Figure 1 As shown, a method for screening laser glass components is provided, comprising the following steps:

[0039] Step S101 : acquiring multiple component data of laser glass containing activated ions, performing molecular dynamics simulation on each component data, and obtaining glass structure data corresponding to each component data.

[0040] Specifically, laser glass refers to multi-component oxide glass doped with laser ions. Laser ions include one or more rare earth ions, transition metal ions or main group metal elements. Rare earth ions include Nd 3+ 、Yb 3+ 、Er 3+ 、Tm 3+ and Ho 3+ The structure of laser glass generally includes network formers, network modifiers, and network intermediates. Network former oxides include SiO2, P2O5, B2O3, and GeO2; network intermediate oxides include ZnO, Al2O3, TiO2, PbO, and La2O3; and network modifier oxides include alkali metal and alkaline earth metal oxides. Laser glass composition data and corresponding laser performance data are collected from databases, literature, or experiments.

[0041] Molecular dynamics is a thermodynamic calculation method that uses the intrinsic dynamics of a system to calculate the coordinates and momenta of atoms and determine changes in the system structure. Because glass structure is a topologically disordered atomic network lacking the long-range order of a crystal structure, it is best described using statistics. Based on the data of multiple laser glass components, atomic-scale modeling and computer simulation using molecular dynamics methods for high-temperature quenching simulations can be performed to obtain the glass structure data corresponding to each component.

[0042] Step S102 : obtaining structural property data of the activated ion localization corresponding to each component data based on the glass structure data corresponding to each component data.

[0043] Specifically, molecular dynamics simulation provides the coordinate information of each atom in the glass structure of laser glass. Based on the glass structure data corresponding to each component data, the distribution function can be used to obtain the structural property data of the local activated ions corresponding to each component data.

[0044] Step S103: Obtain laser performance data corresponding to each component data, and construct a structural property-laser performance data set based on the structural property data of the activated ion localization corresponding to the same component data and the corresponding laser performance data.

[0045] Specifically, each component data of laser glass has corresponding measured laser performance data. Based on the structural property data of the local activated ions and the corresponding laser performance data of the same component data, the connection between the structural property data and the corresponding laser performance data is established one by one to construct a structural property-laser performance data set.

[0046] Step S104 , performing model training using the structural property-laser performance data set to obtain a laser performance prediction model.

[0047] Specifically, the difference in glass components between laser glasses results in different glass structures between laser glasses, and the different structural properties of the activated ion localization also correspond to different laser performance. Regression is a statistical analysis method that studies the dependence of the dependent variable on the independent variable, with the purpose of estimating or predicting the value of the dependent variable by the given value of the independent variable. Therefore, the functional relationship between the structural properties of the activated ion localization and the laser performance can be analyzed by establishing a multivariate linear regression model or other methods. The established model is trained using the structural property-laser performance data set to obtain a laser performance prediction model, so that the structural properties can be used as the independent variable to accurately estimate or predict the value of the laser performance as the independent variable. Exemplarily, the multivariate linear regression model can adopt a partial least squares regression algorithm model.

[0048] Step S105 , using the laser performance prediction model, screening out component data with excellent laser performance from the component space of the laser glass containing the activated ions.

[0049] Specifically, the component space refers to the multi-component oxides that make up the laser glass. The types of these oxides are determined, while the concentrations of each component oxide are variable. The component space contains multiple component data. Referring to steps S101 to S103, the structural properties of the localized activated ions corresponding to the multiple components within the component space are analyzed through a traversal process. Laser performance prediction models are then used to predict the laser performance of the multiple component data within the component space, screening out component data with superior laser performance within the component space.

[0050] In the above-mentioned method for screening laser glass components, the corresponding glass structure data is generated based on the glass component data of the laser glass and molecular dynamics simulation, and the corresponding structural property data of the local activated ions are obtained based on the laser glass structure data. Finally, the laser performance data corresponding to each component data are used to construct a structural property-laser performance data set, which fully takes into account the decisive role of the structural properties of the local activated ions in the laser glass on the laser performance. Moreover, the structural property-laser performance data set is used for model training to obtain a laser performance prediction model; the laser performance prediction model is used to screen out component data with excellent laser performance in the component space from the component space of the laser glass containing the activated ions. Based on the constructed component-structure-performance relationship of the laser glass, the laser performance prediction model obtained by training is used to predict the laser performance of a large range of the component space of the laser glass, and laser glass components with excellent laser performance are screened out. Furthermore, the laser performance prediction model can also be used to infer the glass structure with the target laser performance based on the target laser performance. Therefore, the screening method for laser glass components of the present application is highly interpretable.

[0051] In one embodiment, the structural properties of the activation ion localization corresponding to each component data include: the glass matrix ion distribution in the radial direction of the activation ions and the activation ion distribution in the radial direction of the activation ions.

[0052] Specifically, glass matrix ions refer to other types of ions in the glass structure except for the active ions.

[0053] In one embodiment, according to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained, including:

[0054] Glass structure data and expressions corresponding to each component data Obtain the radial glass matrix ion distribution of the activated ions corresponding to each component data;

[0055] Where r is the radius of the region with the central activation ion as the reference point; dn ReX (r) is the number of glass matrix ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0056] Specifically, the expression for the radial distribution of glass matrix ions within the activation ion region represents the probability of occurrence of glass matrix ions within the activation ion localization region. The peaks in the distribution function curve calculated using this expression contain information about the structural geometry of the glass matrix ions within the activation ion localization region. The different distribution function curves for glass matrix ions of different compositions indicate different structural properties within the activation ion localization region. These local structural variations influence the laser performance of the laser glass.

[0057] In this embodiment, based on the glass structure data and related expressions corresponding to each component data, the radial glass matrix ion distribution characteristics of the activated ions corresponding to each component data are accurately analyzed to improve the performance of the laser performance prediction model, thereby improving the accuracy of the laser glass component screening method of this application.

[0058] In one embodiment, according to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained, including:

[0059] Glass structure data and expressions corresponding to each component data Obtain the radial distribution of activated ions corresponding to each component data;

[0060] Where r is the radius of the region with the central activation ion as the reference point; dn ReRe (r) is the number of activated ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0061] Specifically, the expression for the radial distribution of activated ions represents the probability of occurrence of localized activated ions. The peaks in the distribution function curve calculated using this expression contain information about the structural geometry of the activated ions in the localized activated ion cluster. Different distribution function curves for different components indicate different structural properties of the localized activated ions. These localized structural changes affect the laser performance of laser glass.

[0062] In this embodiment, based on the glass structure data and related expressions corresponding to each component data, the radial activation ion distribution characteristics of the activation ions corresponding to each component data are accurately analyzed to improve the performance of the laser performance prediction model, thereby improving the accuracy of the laser glass component screening method of this application.

[0063] In one embodiment, a model is trained using a structural property-laser performance data set to obtain a laser performance prediction model, including: using the structural property data of the activated ion localization as input variables and the laser performance data as output variables to train the model to obtain a laser performance prediction model.

[0064] Specifically, the model framework consists of an input layer, an intermediate layer, and an output layer. The input variables of the input layer are the structural property data of the laser glass, the intermediate layer is the set algorithm structure, and the output variables of the output layer are the laser performance data of the laser glass.

[0065] In one embodiment, the structural property data of the local activated ion is used as the input variable and the laser performance data is used as the output variable to perform model training to obtain a laser performance prediction model, including: setting different numbers of intermediate layer nodes for the initial laser performance prediction model to obtain multiple intermediate laser performance prediction models; using the structural property data of the local activated ion as the input variable and the laser performance data as the output variable to obtain the mean square error corresponding to each intermediate laser performance prediction model; and obtaining a laser performance prediction model based on the intermediate laser performance prediction model corresponding to the minimum mean square error.

[0066] Specifically, training involves fine-tuning the model's structural parameters through methods such as ten-fold cross-validation. The structural properties-laser performance dataset is divided into a test dataset and a training dataset. For example, the test dataset accounts for 20% and the training dataset accounts for 80%. Ten-fold cross-validation involves splitting the training dataset into ten parts: nine for training the model and one for testing. The results are averaged after ten iterations.

[0067] By setting different numbers of intermediate-layer nodes for the initial laser performance prediction model, multiple intermediate laser performance prediction models were obtained, each with a different mean square error (MSE). The MSE refers to the difference between the predicted value and the actual experimental value, and is used to characterize the performance of the laser performance prediction model. A larger MSE indicates a higher performance, while a smaller MSE indicates a lower performance. Based on the intermediate laser performance prediction model that minimizes the MSE, the optimal number of intermediate-layer nodes was determined, and the laser performance prediction model was then obtained.

[0068] In this embodiment, by setting different numbers of intermediate layer nodes for the laser performance prediction model, the number of nodes corresponding to the laser performance prediction model with the best performance can be adjusted and confirmed, thereby improving the performance of the laser performance prediction model and thereby improving the accuracy of the laser glass component screening method of this application.

[0069] In one embodiment, a laser performance prediction model is obtained based on the intermediate laser performance prediction model corresponding to the minimum mean square error, including: when the mean square error of the intermediate laser performance prediction model corresponding to the minimum mean square error is less than a standard threshold, the intermediate laser performance prediction model corresponding to the minimum mean square error is used as the laser performance prediction model.

[0070] Specifically, the performance of the laser performance prediction model is evaluated using the test data set in the structural property-laser performance data set. When the mean square error of the intermediate laser performance prediction model corresponding to the minimum mean square error is less than the standard threshold, it is considered that the performance of the intermediate laser performance prediction model corresponding to the minimum mean square error meets the requirements and can be used as a laser performance prediction model for screening laser glass components.

[0071] In this embodiment, by using the intermediate laser performance prediction model with the smallest mean square error and less than the standard threshold as the laser performance prediction model, the performance of the laser performance prediction model can be further ensured to be excellent, and the laser performance of the laser glass can be accurately predicted based on the structural properties of the local activated ions, thereby improving the accuracy of the laser glass component screening method of this application.

[0072] In one embodiment, the laser performance data includes at least one of the following: emission peak position, peak stimulated emission cross section, radiation lifetime, fluorescence effective linewidth, and fluorescence half-width.

[0073] In order to better understand the above method, an application example of the screening method of laser glass components of the present application is described in detail below.

[0074] First, obtain Nd from the INTERGLAD glass database 3+ Composition data of doped phosphate and silicate laser glasses and Nd 3+ Ionic 4 F 3 / 2 → 4 I 11 / 2 Energy level transition peak stimulated emission cross section data, a total of 145 groups. 3+ Multiple component data of ion laser glass are obtained, and molecular dynamics simulation is performed on each component data to obtain the corresponding glass structure data of each component data.

[0075] Secondly, according to the glass structure data corresponding to each component data, the structural property data of the local activated ions corresponding to each component data are obtained. The structural properties of the local activated ions corresponding to each component data include: Nd 3+ Radial glass matrix ion distribution and Nd 3+ Radial distribution of activated ions, these two distributions are considered as Nd 3+ Local structural properties.

[0076] Glass structure data and expressions corresponding to each component data Get the corresponding Nd for each component data 3+ Radial distribution of glass matrix ions; where r is the radius of the region with the central activation ion as the reference point; dn ReX (r) is the number of glass matrix ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0077] Glass structure data and expressions corresponding to each component data Get the corresponding Nd for each component data 3+ Radial distribution of activated ions; where r is the radius of the region with the central activated ion as the reference point; dn ReRe (r) is the number of activated ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0078] Next, obtain the laser performance data corresponding to each component data, and calculate the Nd 3+ Local structural property data and the corresponding Nd 3+ Ionic 4 F 3 / 2 → 4 I 11 / 2 Energy level transition peak stimulated emission cross section data is used to construct a structural property-laser performance data set.

[0079] like Figure 2 The structure and working process diagram of the laser performance prediction model shown in the figure is taken as an example. 3+ The local structural property data is used as input variables. 3+ Local structural property data include Nd 3+ Radial glass matrix ion distribution and Nd 3+ Radial distribution of activated ions, Nd 3+ Ionic 4 F 3 / 2 → 4 I 11 / 2 The energy level transition peak stimulated emission cross section data is used as the output variable, and the middle layer is the partial least squares regression algorithm structure. The model structure parameters are tuned through methods such as ten-fold cross validation, the model is trained, and the predicted value and experimental value of each data point are statistically analyzed.

[0080] It should be understood that, except for taking the peak stimulated emission cross section in the laser performance data as the output variable, the working process of the model with the emission peak position, radiation lifetime, fluorescence effective linewidth and fluorescence half-width as output variables is the same as the working process in the above embodiment.

[0081] Among them, Figure 3 The mean square error of the intermediate laser performance prediction model with different numbers of intermediate layer nodes is shown in the figure. It can be seen that when the number of intermediate layer nodes is 8, the mean square error of the intermediate laser performance prediction model is the smallest. Figure 4 From the result analysis diagram of the predicted values ​​and experimental values ​​of the intermediate laser performance prediction model corresponding to the minimum mean square error shown, it can be seen that the mean square error between the predicted values ​​and the experimental values ​​of each data point is 0.15. It is considered that the mean square error is less than the standard threshold, that is, the model performance meets the requirements. Based on the intermediate laser performance prediction model corresponding to the minimum mean square error, the laser performance prediction model is obtained.

[0082] Finally, the laser performance prediction model was used to 3+ The component data with excellent laser performance were screened out from the component space of K2O-La2O3-B2O3-P2O5-Nd2O3. The components with higher peak stimulated emission areas were finally screened out as 0.15K2O-0.3La2O3-0.1B2O3-0.695P2O5-0.025Nd2O3 and 0.1K2O-0.3La2O3-0.15B2O3-0.695P2O5-0.025Nd2O3. The calculated peak stimulated emission areas were 4.67×10 -20 cm -2 and 4.63×10 - 20 cm -2 .

[0083] Furthermore, laser glass can be prepared by screening glass components with excellent laser performance and adopting high-temperature melting-annealing method, sol-gel method, MCVD method or suspension furnace melting method.

[0084] Specifically, two pieces of laser glass were prepared using a traditional high-temperature melting-annealing method. Each component of the glass was accurately weighed and ground uniformly in a mortar. The mixture was then poured into a crucible and melted at 1300°C for 30 minutes. The mixture was then poured into a copper mold and quenched to form glass. Finally, the quenched glass was quickly transferred to an annealing furnace for annealing.

[0085] In this embodiment, glass structure data is generated based on the glass component data of the laser glass and molecular dynamics simulation, and the corresponding structural property data of the local activated ions are obtained based on the laser glass structure data. Finally, the laser performance data corresponding to each component data are used to construct a structural property-laser performance data set, fully considering the decisive role of the structural properties of the local activated ions in the laser glass on the laser performance. In addition, the structural property-laser performance data set is used for model training to obtain a laser performance prediction model; the laser performance prediction model is used to screen component data with excellent laser performance in the component space of the laser glass containing the activated ions. Based on the constructed component-structure-performance relationship of the laser glass, the laser performance prediction model obtained through training is used to predict laser performance for a wide range of the component space of the laser glass, from which laser glass components with excellent laser performance are screened. Furthermore, the laser performance prediction model can also be used to infer glass structures with the target laser performance based on the target laser performance. Therefore, the screening method of laser glass components of this application is highly interpretable.

[0086] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0087] In one embodiment, Figure 5 As shown, a screening device for laser glass components is provided, comprising:

[0088] The glass structure analysis module 501 is used to obtain multiple component data of the laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain glass structure data corresponding to each component data;

[0089] Structural property analysis module 502, for obtaining structural property data of activated ion localization corresponding to each component data based on glass structural data corresponding to each component data;

[0090] A data set construction module 503 is used to obtain laser performance data corresponding to each component data, and to construct a structure property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data;

[0091] A model training module 504 is used to perform model training using the structural property-laser performance data set to obtain a laser performance prediction model;

[0092] The glass component screening module 505 is used to screen component data with excellent laser performance from the component space of the laser glass containing activated ions using the laser performance prediction model.

[0093] In one embodiment, the structural properties of the activation ion localization corresponding to each component data include: the glass matrix ion distribution in the radial direction of the activation ions and the activation ion distribution in the radial direction of the activation ions.

[0094] In one embodiment, the activated ion localized structural property analysis module 502 is further configured to analyze the glass structure data and expressions corresponding to each component data. The radial distribution of glass matrix ions corresponding to the activation ions of each component data is obtained; where r is the radius of the region with the central activation ion as the reference point; dn ReX (r) is the number of glass matrix ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0095] In one embodiment, the structural property analysis module 502 is further configured to analyze the glass structure data and expressions corresponding to each component data. The radial distribution of activated ions corresponding to each component data is obtained; where r is the radius of the region with the central activated ion as the reference point; dn ReRe (r) is the number of activated ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

[0096] In one embodiment, the model training module 504 is further configured to perform model training using the structural property data of the local activated ion as input variables and the laser performance data as output variables to obtain a laser performance prediction model.

[0097] In one embodiment, the model training module 504 is also used to set different numbers of intermediate layer nodes for the initial laser performance prediction model to obtain multiple intermediate laser performance prediction models; using the structural property data of the activated ion localization as input variables and the laser performance data as output variables to obtain the mean square error corresponding to each intermediate laser performance prediction model; and obtaining the laser performance prediction model based on the intermediate laser performance prediction model corresponding to the minimum mean square error.

[0098] In one embodiment, the model training module 504 is further configured to use the intermediate laser performance prediction model corresponding to the minimum mean square error as the laser performance prediction model when the mean square error of the intermediate laser performance prediction model corresponding to the minimum mean square error is less than a standard threshold.

[0099] In one embodiment, the laser performance data includes at least one of the following: emission peak position, peak stimulated emission cross section, radiation lifetime, fluorescence effective linewidth, and fluorescence half-width.

[0100] The specific definitions of the laser glass component screening device can be found in the definitions of the laser glass component screening method described above and will not be repeated here. Each module of the aforementioned laser glass component screening device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of these modules can be embedded in or independent of a processor within a computer device in hardware form, or stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0101] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store screening data of laser glass components. The network interface of the computer device is used to communicate with an external terminal via a network connection. The computer device also includes an input and output interface, which is a connection circuit for exchanging information between the processor and the external device. They are connected to the processor via a bus, referred to as an I / O interface. When the computer program is executed by the processor, a method for screening laser glass components is implemented.

[0102] Those skilled in the art will understand that Figure 6The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0103] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0104] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0105] In one embodiment, a computer program product is provided, on which a computer program is stored. The computer program is used by a processor to execute the steps in the above-mentioned various method embodiments.

[0106] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0107] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0108] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] The above embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for screening laser glass components, characterized in that: The method comprises: Acquire multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain glass structure data corresponding to each component data; According to the glass structure data corresponding to each component data, the structural property data of the activated ion localization corresponding to each component data is obtained; Obtain the laser performance data corresponding to each component data, and construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data; Using the structural property-laser performance data set to perform model training, a laser performance prediction model is obtained; The laser performance prediction model is used to screen out component data with excellent laser performance from the component space of the laser glass containing the activated ions.

2. The method according to claim 1, characterized in that The structural properties of the activation ion localization corresponding to each component data include: the glass matrix ion distribution in the radial direction of the activation ion and the activation ion distribution in the radial direction of the activation ion.

3. The method according to claim 2, characterized in that The method of obtaining the structural property data of the activated ion localization corresponding to each component data according to the glass structure data corresponding to each component data includes: Glass structure data and expressions corresponding to each component data Obtain the radial glass matrix ion distribution of the activated ions corresponding to each component data; Where r is the radius of the region with the central activation ion as the reference point; dn ReX (r) is the number of glass matrix ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

4. The method according to claim 2, characterized in that The method of obtaining the structural property data of the activated ion localization corresponding to each component data according to the glass structure data corresponding to each component data includes: Glass structure data and expressions corresponding to each component data Obtain the radial distribution of activated ions corresponding to each component data; Where r is the radius of the region with the central activation ion as the reference point; dn ReRe (r) is the number of activated ions in the region from r to r+dr; ρ Re is the number density of activated ions in the glass structure.

5. The method according to claim 1, wherein The method of using the structural property-laser performance data set to perform model training to obtain a laser performance prediction model includes: The structural property data of the local activated ion region is used as input variables, and the laser performance data is used as output variables to perform model training to obtain a laser performance prediction model.

6. The method according to claim 5, characterized in that The structural property data of the local activated ion region is used as input variables, and the laser performance data is used as output variables to perform model training to obtain a laser performance prediction model, including: Setting different numbers of intermediate layer nodes for the initial laser performance prediction model to obtain multiple intermediate laser performance prediction models; Using the structural property data of the local activated ion as input variables and the laser performance data as output variables, obtaining the mean square error corresponding to each intermediate laser performance prediction model; Based on the intermediate laser performance prediction model corresponding to the minimum mean square error, a laser performance prediction model is obtained.

7. The method according to claim 6, characterized in that Based on the intermediate laser performance prediction model corresponding to the minimum mean square error, the laser performance prediction model is obtained, including: When the mean square error of the intermediate laser performance prediction model corresponding to the minimum mean square error is less than a standard threshold, the intermediate laser performance prediction model corresponding to the minimum mean square error is used as the laser performance prediction model.

8. The method according to claim 1, characterized in that The laser performance data includes at least one of the following: emission peak position, peak stimulated emission cross section, radiation lifetime, fluorescence effective line width and fluorescence half-maximum width.

9. A device for screening laser glass components, characterized in that: The device comprises: Glass structure analysis module, used to obtain multiple component data of laser glass containing activated ions, perform molecular dynamics simulation on each component data, and obtain the glass structure data corresponding to each component data; The structural property analysis module obtains the structural property data of the activated ion localization corresponding to each component data based on the glass structure data corresponding to each component data; A data set construction module is used to obtain the laser performance data corresponding to each component data, and to construct a structural property-laser performance data set based on the structural property data of the activated ion localization and the corresponding laser performance data corresponding to the same component data; A model training module, configured to perform model training using the structural property-laser performance data set to obtain a laser performance prediction model; The glass component screening module is used to use the laser performance prediction model to screen out component data with excellent laser performance from the component space of the laser glass containing the activated ions.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Images