Information processing methods, information processing systems, and programs

Abstract

Expand the explanatory variables for materials with crystalline structures. [Solution] The information processing method includes acquiring spectral data showing the spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure using an X-ray diffraction measuring device 20, calculating multiple spectral feature quantities that show the characteristics of the spectrum from the spectral data, and extracting lattice angle feature quantities by applying arithmetic operations to them.

Description

[Technical Field]

[0001] This disclosure relates to a technique for analyzing the properties of materials having a crystalline structure. [Background technology]

[0002] The technology described in Patent Document 1 discloses a method for calculating characteristic quantities such as peak full width at half maximum, diffraction position, integrated intensity, and Lorentz component ratio by approximating spectral data with a pseudo-Voigt function. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] International Publication No. 2022 / 180776 [Overview of the project] [Problems that the invention aims to solve]

[0004] However, in the technology shown in Patent Document 1, none of the calculated features are related to lattice angles, so further improvements are needed to expand the explanatory variables for materials having a crystalline structure.

[0005] This disclosure is made to solve the above-mentioned problems and aims to provide a technology that can expand the explanatory variables of materials having a crystalline structure. [Means for solving the problem]

[0006] An information processing method in one aspect of the present disclosure is an information processing method performed by an information processing device, comprising: acquiring spectral data showing a spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure; and extracting lattice angle features relating to the lattice angle in the spectrum from the spectral data. [Effects of the Invention]

[0007] According to this disclosure, the explanatory variables for materials having a crystalline structure can be expanded. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram showing an example of the configuration of an information processing system in Embodiment 1 of the present disclosure. [Figure 2] This is a flowchart showing an example of the processing of the information processing system in Embodiment 1 of this disclosure. [Figure 3] This graph shows an example of spectral data. [Figure 4] This flowchart shows the details of the process for extracting spectral features, as shown in step S2 of Figure 2. [Figure 5] This graph shows an example of a Gaussian function and a Lorentz function. [Figure 6] This table summarizes the relationship between multiple materials and the peak positions of the 018 plane. [Figure 7] This flowchart shows the details of the process for extracting lattice angle features, as shown in step S3 of Figure 2. [Figure 8] This figure shows an example of how to calculate lattice angle features. [Figure 9] This diagram summarizes the difference obtained by subtracting the peak position of the 012 plane from the peak position of the 018 plane, for each of several materials. [Figure 10] This flowchart shows an example of the process used when training a predictive model. [Figure 11] This flowchart shows the details of the process in step S21. [Figure 12] This is a block diagram showing an example of the configuration of an information processing system in Embodiment 2 of this disclosure. [Figure 13] This flowchart shows an example of the processing of the information processing system in Embodiment 2 of this disclosure. [Figure 14] This figure shows the first example of a selection screen for selecting characteristic information and lattice angle features. [Figure 15] This is a diagram showing the first example map image. [Figure 16] It is a diagram showing the selection screen of the second example. [Figure 17] It is a diagram showing the map image of the second example. [Figure 18] It is a diagram showing the map image of the third example. [Figure 19] It is a table summarizing the results of the examples. [Figure 20] It is a scatter diagram showing the relationship between the difference between peak positions and the lattice angle. [Figure 21] It is a scatter diagram showing the relationship between the difference between peak intensities and the lattice angle. [Figure 22] It is a scatter diagram showing the relationship between the difference between peak positions and peak intensities. [Figure 23] It is a diagram showing the crystal structure of the hexagonal system. [Figure 24] It is a diagram showing the plane defined in the hexagonal system. [Figure 25] It is a diagram showing the change in symmetry due to the change in the lattice angle in a material of the hexagonal system.

Mode for Carrying Out the Invention

[0009] (Knowledge underlying the present disclosure) Research is underway to obtain spectral data of a material using an XRD acquisition device and analyze the characteristics of the material from the obtained spectral data. A plurality of diffraction peaks appear in the spectral data. Conventionally, feature quantities such as the position and intensity of the diffraction peak have been considered useful as explanatory variables for the characteristics of the material.

[0010] However, for many materials, plotting the relationship between material properties and the characteristics of a certain diffraction peak on a scatter plot did not yield sufficient correlation, indicating that the characteristics of the diffraction peak are insufficient as explanatory variables for material properties. Therefore, the inventors focused on lattice angle characteristics, such as the difference in positions between diffraction peaks, instead of the characteristics of a certain diffraction peak. They then plotted the relationship between lattice angle characteristics and material properties on a scatter plot. It was found that a high correlation was obtained between these two. In other words, it was found that lattice angle characteristics are useful as explanatory variables for material properties.

[0011] In the prior art described in Patent Document 1, as mentioned above, feature quantities such as the peak full width at half maximum, diffraction position, integrated intensity, and Lorentz component ratio are used. These feature quantities are all simply feature quantities that focus on a particular diffraction peak, and are not feature quantities related to the lattice angle. Therefore, the feature quantities shown in the prior art are insufficient as explanatory variables for materials.

[0012] This disclosure was made to address these issues.

[0013] (1) An information processing method in one aspect of the present disclosure is an information processing method performed by an information processing device, which includes: acquiring spectral data showing a spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure; and extracting lattice angle features relating to the lattice angle in the spectrum from the spectral data.

[0014] This configuration extracts lattice angle features related to lattice angles from spectral data. Unlike the features described in Patent Document 1, these lattice angle features correlate with characteristic information indicating material properties. Specifically, lattice angle features correlate with characteristic information indicating material properties in materials where material properties change with changes in lattice angle. Therefore, this configuration can expand the explanatory variables for materials with crystalline structures. Furthermore, this configuration allows for a simple investigation of the relationship between changes in the lattice angle of the crystal structure and material properties without complex analysis. Moreover, understanding the above relationship can lead to obtaining new design guidelines for materials. This configuration can also facilitate the development of materials optimized for specific applications. Furthermore, this configuration can improve the accuracy of predicting material properties by expanding information on crystal structures. As a result, this configuration simplifies characterization and reduces the effort required for experiments.

[0015] (2) The information processing method described in (1) above may further include estimating characteristic information relating to the properties of the material based on the lattice angle feature quantity, and outputting the estimated characteristic information.

[0016] In this configuration, characteristic information is estimated using lattice angle features related to lattice angles. Thus, because this configuration estimates characteristic information using lattice angle features, which are specific and detailed features, the accuracy of characteristic information estimation is improved.

[0017] (3) In the information processing method described in (1) or (2) above, extracting the lattice angle feature may include extracting the lattice angle feature from the diffraction peaks in the spectrum.

[0018] With this configuration, lattice angle features are extracted using diffraction peaks, allowing for accurate calculation of lattice angle features.

[0019] (4) In the information processing method described in (3) above, the lattice angle feature may be calculated based on at least one of the position and intensity of each of the at least two diffraction peaks.

[0020] With this configuration, the lattice angle feature is calculated based on the position and at least one of the intensities of at least two diffraction peaks, so the lattice angle feature can be calculated with high accuracy.

[0021] (5) In the information processing method described in (4) above, the lattice angle feature may represent the difference or ratio of the positions of each of the at least two diffraction peaks.

[0022] According to this configuration, lattice angle features are extracted using the difference or ratio of the positions of at least two diffraction peaks. Therefore, this configuration allows for the extraction of specific, detailed, and highly accurate lattice angle features related to the crystal structure in materials having a crystalline structure.

[0023] (6) In the information processing method described in (3) above, extracting the lattice angle feature may include extracting the lattice angle feature from the diffraction peaks corresponding to a plane determined according to the type of crystal structure.

[0024] This configuration allows for the extraction of lattice angle features from diffraction peaks corresponding to planes defined according to the crystal structure.

[0025] (7) In the information processing method described in any of (1) to (6) above, the crystal structure has a crystal system with fixed lattice angles, and the lattice angle feature quantity may be a feature quantity that represents a change in symmetry caused by a change in the lattice angles in the crystal structure.

[0026] This configuration allows for the extraction of feature quantities that accurately capture changes in symmetry due to changes in the lattice angles of a crystalline structure in materials having a crystalline structure.

[0027] (8) In the information processing method described in any of (1) to (7) above, extracting the lattice angle feature may include calculating a plurality of spectral feature quantities that represent the characteristics of the spectrum from the spectral data, and extracting the lattice angle feature quantity by applying arithmetic operations to the plurality of spectral feature quantities.

[0028] With this configuration, lattice angle features are calculated by applying arithmetic operations to multiple spectral features that represent the characteristics of the spectrum, thus enabling accurate calculation of lattice angle features.

[0029] (9) In the information processing method described in (8) above, calculating the spectral features may include extracting multiple spectral features by approximating a model function to the spectral data.

[0030] With this configuration, multiple spectral features are calculated by approximating the spectral data with a model function, allowing for the accurate extraction of multiple spectral features.

[0031] (10) An information processing system in another aspect of the present disclosure includes an acquisition unit that acquires spectral data showing a spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure, and an extraction unit that extracts lattice angle features in the spectrum from the spectral data.

[0032] This configuration provides an information processing system that can achieve the same effects as the information processing method described in (1) above.

[0033] (11) The information processing system described in (10) above may further include a display control unit that generates an image representing one or more first maps showing the relationship between the difference in the peak positions of each of the at least two diffraction peaks included in the lattice angle feature quantity and the properties of the material, and displays the image on a display unit.

[0034] According to this configuration, one or more first maps showing the relationship between the difference in the positions of at least two diffraction peaks included in the lattice angle feature and the material properties have a correlation between the difference and the material properties. Therefore, the correlation between the difference and the material properties can be presented.

[0035] (12) In the information processing system described in (11) above, the image may include a plurality of second maps in which the positions of each of the at least two diffraction peaks are different, and each of the plurality of second maps may show the relationship between the diffraction peaks and the properties of the material.

[0036] This configuration presents multiple second maps, each with a different diffraction peak position. Therefore, users can compare one or more first maps with the second maps to gain a more detailed understanding of the material's properties. Furthermore, comparing the first and second maps facilitates the extraction of features most closely related to the material's properties, leading to a more detailed understanding of the material's characteristics.

[0037] (13) The information processing system described in (10) or (11) above further comprises an estimation unit that estimates characteristic information relating to the properties of the material by inputting the lattice angle feature into a prediction model, and the prediction model may be trained to take the lattice angle feature as input and output characteristic values ​​of the material.

[0038] With this configuration, the characteristic information of lattice angle features, whose characteristic information is unknown, can be estimated using a predictive model.

[0039] (14) A program in another aspect of the present disclosure causes a computer to perform the following actions: acquire spectral data showing a spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure; and extract lattice angle features relating to the lattice angles in the spectrum from the spectral data.

[0040] This configuration provides a program that can achieve the same effect as the information processing method described in (1) above.

[0041] It goes without saying that such computer programs can be distributed via computer-readable, non-temporary recording media such as CD-ROMs or via communication networks such as the Internet.

[0042] The embodiments described below are all specific examples of this disclosure. The numerical values, shapes, components, steps, and order of steps shown in the following embodiments are examples only and are not intended to limit this disclosure. Furthermore, among the components in the following embodiments, those not described in the independent claim representing the highest-level concept will be described as optional components. In addition, the contents of each embodiment can be combined.

[0043] (Embodiment 1) Figure 1 is a block diagram showing an example of the configuration of the information processing system 1 in Embodiment 1 of this disclosure. The information processing system 1 is a system for analyzing materials having a crystalline structure. The materials are, for example, substances having crystalline structures such as hexagonal, monoclinic, orthorhombic, tetragonal, trigonal, cubic, and triclinic systems.

[0044] The information processing system 1 includes an XRD (X-ray diffractometer) measuring device 20 and an information processing device 100.

[0045] The XRD measuring device 20 is a device for measuring the spectrum of a material. The XRD measuring device 20 irradiates a material with a crystalline structure with X-rays and analyzes the diffraction pattern to obtain spectral data that shows the spectrum of the material.

[0046] The information processing device 100 is comprised of, for example, a computer. The information processing device 100 includes a processor 10, memory 40, and an operation unit 50.

[0047] The processor 10 includes an acquisition unit 11, an extraction unit 12, an estimation unit 13, an output unit 14, and a learning unit 15. The acquisition unit 11 to the learning unit 15 are realized by the processor 10 executing an information processing program stored in the memory 40. However, this is just an example, and the acquisition unit 11 to the learning unit 15 may be composed of dedicated hardware circuits.

[0048] The acquisition unit 11 acquires spectral data from the XRD measuring device 20 that shows the spectrum measured by the XRD measuring device 20.

[0049] The extraction unit 12 extracts one or more lattice angle features related to the lattice angle in the spectrum from the spectral data acquired by the acquisition unit 11.

[0050] In detail, the extraction unit 12 calculates multiple spectral features that represent the characteristics of the spectrum from the spectral data. The extraction unit 12 calculates a lattice angle feature by applying arithmetic operations to the multiple spectral features.

[0051] Multiple spectral features include, for example, the position (peak position) and intensity of each of the multiple diffraction peaks included in the spectrum. These spectral features may also include peak width and the Lorentz blending ratio.

[0052] The extraction unit 12 can calculate multiple spectral features by approximating the spectral data with a model function 43. The model function 43 is, for example, a symmetric pseudo-Voigt function.

[0053] Lattice angle features are, for example, the difference or ratio of the positions of multiple diffraction peaks.

[0054] The estimation unit 13 estimates characteristic information regarding the material properties based on a plurality of lattice angle features extracted by the extraction unit 12. The material properties used are, for example, those that are easily affected by the lattice angle. For example, the material properties may include battery properties, thermal conductivity, ionic conductivity, mechanical properties, and dielectric constant.

[0055] Battery characteristics include, for example, charge capacity, discharge capacity, and capacity retention rate. These characteristics can change due to the influence of crystal structure distortion (Jahn-Teller effect). Thermal conductivity can generally decrease with crystal structure distortion. Ionic conductivity can change due to the influence of crystal structure distortion on ion transport paths. Mechanical properties include, for example, Young's modulus and fracture strength. Mechanical properties can be affected by crystal structure distortion. Dielectric constant is sensitive to crystal structure distortion.

[0056] In detail, the estimation unit 13 estimates characteristic information regarding the material's properties by inputting multiple lattice angle features into the prediction model 42. The prediction model 42 is a model trained to take multiple lattice angle features as input and output characteristic information indicating the material's characteristic values. The prediction model 42 can employ, for example, a supervised learning machine learning model. Examples of the prediction model 42 include neural networks, linear regression, logistic regression, decision trees, random forests, and support vector machines.

[0057] The output unit 14 outputs the characteristic information extracted by the estimation unit 13. For example, the output unit 14 can output the characteristic information to the database 41.

[0058] The learning unit 15 trains the predictive model 42 using a dataset from the database 41 that stores lattice angle features and actual input characteristic information.

[0059] The memory 40 consists of a non-volatile, rewritable storage device such as a solid-state drive. The memory 40 stores the database 41, the prediction model 42, and the model function 43.

[0060] Database 41 stores spectral data and characteristic information in association. Prediction model 42 is a model that predicts characteristic information from lattice angle features. Model function 43 is a function used when extracting spectral features.

[0061] The control unit 50 is an input device such as a keyboard, mouse, or touch panel.

[0062] The information processing system 1 may consist of a system including a server and terminal devices connected to the server via a network. The network may be, for example, the Internet or a local area network. In this case, each block of the processor 10 may be provided by the server, or it may be distributed between the server and the terminals. The information processing system 1 may consist of a system including a standalone computer and an XRD measuring device 20. The information processing system 1 constitutes an example of an information provision system. The information processing system 1 may also consist of terminals connected to the server.

[0063] Figure 2 is a flowchart showing an example of the processing of the information processing system 1 in Embodiment 1 of this disclosure.

[0064] (Step S1) The acquisition unit 11 acquires spectral data from the XRD measuring device 20. Figure 3 shows an example of spectral data. In Figure 3, the vertical axis is intensity and the horizontal axis is angle 2θ. The spectral data defines the relationship between angle 2θ and X-ray intensity. Angle 2θ is the X-ray scattering angle. The spectral data has multiple diffraction peaks. The angle 2θ corresponding to the diffraction peak is called the peak position. In the example in Figure 3, the peak position of the 012 plane is 37.8 degrees and the peak position of the 018 plane is 64.2 degrees. In step S1, the acquisition unit 11 may acquire multiple spectral data. The acquisition unit 11 may perform a process to remove unnecessary spectra from the acquired spectral data. Unnecessary spectra are, for example, spectra originating from k-alpha2.

[0065] (Step S2) The extraction unit 12 extracts spectral features for each of the multiple diffraction peaks by approximating the spectral data with a model function 43. Details of the process in step S2 will be described later in Figure 4.

[0066] (Step S3) The extraction unit 12 extracts lattice angle features by applying arithmetic operations to the multiple spectral features extracted in step S2. Details of the process in step S3 will be described later in Figure 7.

[0067] (Step S4) The estimation unit 13 determines whether or not to perform an estimation process to estimate characteristic information. For example, if characteristic information corresponding to the spectral data acquired in step S1 is obtained, the estimation unit 13 determines to perform the estimation process. On the other hand, if characteristic information corresponding to the spectral data acquired in step S1 is not obtained, the estimation unit 13 does not perform the estimation process and terminates the process. If it is determined that the estimation process should be performed (YES in step S4), the process proceeds to step S5; if it is determined that the estimation process should not be performed (NO in step S4), the process terminates.

[0068] (Step S5) The estimation unit 13 estimates feature information by inputting the lattice angle features extracted in step S3 into the prediction model 42. If multiple spectral data are obtained in step S1, the estimation unit 13 estimates feature information for each of the spectral data.

[0069] Figure 4 is a flowchart detailing the process of extracting spectral features shown in step S2 of Figure 2.

[0070] (Step S41) The extraction unit 12 approximates the shape of the spectral data using a symmetric pseudo-Voigt function. That is, the extraction unit 12 adjusts the parameters of the symmetric pseudo-Voigt function to match the shape of the spectrum shown by the spectral data. The extraction unit 12 can perform the approximation using the least squares method. The symmetric pseudo-Voigt function is given by equation (1).

[0071]

number

[0072] In equation (1), the first term on the right-hand side is the Gaussian function, and the second term on the right-hand side is the Lorentz function. A represents the peak intensity, μ represents the peak position, σ represents the standard deviation of the Gaussian function, and α represents the Lorentz composite ratio. The peak intensity A, peak position μ, standard deviation σ, and Lorentz composite ratio α are parameters of the symmetric pseudo-Voigt function. The standard deviation σ is correlated with the peak width of the diffraction peak, and therefore represents the peak width.

[0073] Figure 5 shows graphs illustrating examples of Gaussian and Lorentz functions. Both Gaussian and Lorentz functions have a single peak and exhibit symmetry with respect to this peak. The Lorentz function is characterized by a slower change in its tail compared to the Gaussian function. The symmetric pseudo-Voigt function is a function obtained by combining the Gaussian and Lorentz functions using the Lorentz combination ratio α. As the Lorentz combination ratio α approaches 1, the influence of the Lorentz function increases, and the symmetric pseudo-Voigt function takes on a shape with a wider tail.

[0074] (Step S42) The extraction unit 12 extracts the peak intensity A, peak position μ, standard deviation σ, and Lorentz composite ratio α, which are parameters of a symmetric pseudo-Voigt function approximated by the spectral data, as spectral features.

[0075] (Step S43) The extraction unit 12 identifies the crystal plane for each of the multiple diffraction peaks. For example, the extraction unit 12 can calculate the interplanar spacing by applying Bragg's law to the peak positions and then identify the crystal plane based on the obtained interplanar spacing. Alternatively, the extraction unit 12 may use software to identify crystal planes from spectral data to identify the crystal plane for each diffraction peak.

[0076] In the example shown in Figure 3, the diffraction peak with a peak position of 37.8 degrees is labeled with the 012 plane, and the diffraction peak with a peak position of 64.2 degrees is labeled with the 018 plane.

[0077] Figure 6 is a table summarizing the relationship between multiple materials and the peak positions on the 018 plane. The estimation unit 13 labels each diffraction peak, so it is possible to summarize the peak positions for each material on a given crystal plane.

[0078] (Step S44) The extraction unit 12 assigns a label indicating the identified crystal plane to each spectral feature extracted in step S42 and stores it in the memory 40.

[0079] Figure 7 is a flowchart detailing the process of extracting the lattice angle features shown in step S3 of Figure 2.

[0080] (Step S51) The extraction unit 12 obtains spectral features from the memory 40.

[0081] (Step S52) The extraction unit 12 extracts lattice angle features from the acquired spectral features. Here, the extraction unit 12 extracts lattice angle features using peak positions as spectral features. For example, the extraction unit 12 extracts all combinations of any two spectral features from among multiple spectral features corresponding to one spectral data. Then, for each of the extracted combinations, the extraction unit 12 calculates the difference in peak positions and extracts the resulting difference as the lattice angle feature.

[0082] For example, suppose there are six spectral features corresponding to one spectral data point. From these six spectral features, there are 15 possible combinations of any two spectral features. Therefore, the extraction unit 12 extracts 15 combinations of spectral features. Then, the extraction unit 12 calculates the difference in peak positions for each of the 15 combinations of spectral features. The process of calculating the difference in peak positions is an example of arithmetic operation.

[0083] Figure 8 shows an example of lattice angle feature calculation. In the example in Figure 8, the difference between the peak position of the 012 plane and the peak position of the 018 plane is calculated. In this example, the peak position of the 012 plane was 37.8 degrees, and the peak position of the 018 plane was 64.2 degrees. Therefore, the extraction unit 12 calculates the difference between the peak positions of the 012 plane and the 018 plane as 26.4 degrees. The extraction unit 12 also calculates the difference in peak positions for combinations of diffraction peaks other than the combination of the 012 plane and the 018 plane. As a result, multiple lattice angle feature quantities corresponding to a single spectral data are calculated.

[0084] Furthermore, if multiple spectral data sets are to be processed, the extraction unit 12 only needs to calculate the lattice angle feature quantities corresponding to each of the multiple spectral data sets.

[0085] The arithmetic operation for calculating the lattice angle feature may be the ratio of the peak position of one diffraction peak to the peak position of the other diffraction peak at any two diffraction peaks. Alternatively, the arithmetic operation may be at least one of the difference, ratio, and squared difference between the peak positions of one diffraction peak to the peak position of the other diffraction peak at any two diffraction peaks.

[0086] In other words, the arithmetic operation is an arithmetic operation on identical spectral features at any two mutually distinct diffraction peaks. In this case, the extraction unit 12 only needs to apply this arithmetic operation to all combinations of two diffraction peaks included in the spectral data.

[0087] However, this is just one example, and the arithmetic operation may be performed on heterogeneous spectral features at any two mutually distinct diffraction peaks. For example, the arithmetic operation may be the sum of the peak position on the 018 plane and the peak width (standard deviation) on the 003 plane. In this case, the extraction unit 12 only needs to perform this arithmetic operation for all combinations of two diffraction peaks included in the spectral data.

[0088] Alternatively, the arithmetic operation may be an arithmetic operation between different spectral features at any single diffraction peak. For example, the arithmetic operation may be the product of the peak position of the 018 plane and the Lorentz blending ratio of the 018 plane. In this case, the extraction unit 12 only needs to perform this arithmetic operation for all diffraction peaks included in the spectral data.

[0089] (Step S53) The extraction unit 12 assigns labels to the lattice angle features calculated in step S52. These labels identify which diffraction peak each lattice angle feature relates to. For example, a lattice angle feature representing the difference between the peak position of the 012 plane and the peak position of the 018 plane is labeled as "012 plane peak position - 018 plane peak position".

[0090] Figure 9 is a diagram summarizing the difference between the peak position of the 012 plane and the peak position of the 018 plane for multiple materials. Since the extraction unit 12 assigns a label to each of the lattice angle features, the extraction unit 12 can summarize the difference between the peak position of the 012 plane and the peak position of the 018 plane for multiple materials.

[0091] Next, we will explain how to train the prediction model 42. Figure 10 is a flowchart showing an example of the process for training the prediction model 42. This process is performed, for example, when the number of datasets stored in the database 41 exceeds a predetermined number. The predetermined number is, for example, the number of datasets that can be used to train the prediction model 42.

[0092] (Step S21) The acquisition unit 11 executes the database construction process for the database 41. Details of this process will be described later in Figure 11.

[0093] (Step S22) The extraction unit 12 reads spectral data from the database 41 and extracts multiple spectral features by applying the method described in Figure 4 to the read spectral data. In other words, the extraction unit 12 extracts multiple spectral features by approximating the model function 43 to the spectral data.

[0094] (Step S23) The extraction unit 12 calculates the lattice angle feature by applying the method described in Figure 7 to the multiple spectral features extracted in step S22. In other words, the extraction unit 12 extracts the lattice angle feature by applying the above-described arithmetic operations to the multiple spectral features extracted in step S22.

[0095] (Step S24) The learning unit 15 trains the prediction model 42 on a dataset in which the lattice angle features extracted in step S23 are used as explanatory variables and the characteristic information corresponding to the lattice angle features is used as the target variable. The learning unit 15 only needs to train the prediction model 42 using all the datasets stored in the database 41. If a new dataset is added to the database 41, the learning unit 15 may update the prediction model 42 using the newly added dataset.

[0096] Figure 11 is a flowchart detailing the process in step S21.

[0097] (Step S31) The acquisition unit 11 acquires spectral data from the XRD measuring device 20. Here, the acquisition unit 11 may acquire multiple spectral data corresponding to multiple materials.

[0098] (Step S32) The acquisition unit 11 acquires characteristic information corresponding to the spectral data acquired in step S31. If multiple spectral data are acquired in step S31, characteristic information corresponding to each of the multiple spectral data should be acquired. Any method can be used to acquire the characteristic information. For example, the characteristic information may be input by the user operating the operation unit 50.

[0099] (Step S33) The acquisition unit 11 generates a dataset by associating the spectral data acquired in step S31 with the characteristic information acquired in step S32, and adds the generated dataset to the database 41. If multiple spectral data sets are acquired in step S31, multiple datasets are stored in the database 41. The acquisition unit 11 may also store the lattice angle features extracted in step S23 in the database 41 in association with the spectral data.

[0100] As described above, according to the information processing system 1 of Embodiment 1, lattice angle features are extracted from spectral data. Unlike the features shown in Patent Document 1, lattice angle features are correlated with characteristic information that indicates the properties of the material. Therefore, this configuration can expand the explanatory variables for materials having a crystalline structure.

[0101] (Embodiment 2) Embodiment 2 outputs a map image showing the relationship between the lattice angle features extracted in Embodiment 1 and characteristic information. In this embodiment, the same reference numerals are used for components that are the same as in Embodiment 1, and their descriptions are omitted.

[0102] Figure 12 is a block diagram showing an example of the configuration of an information processing system 1A in Embodiment 2 of the present disclosure. The information processing system 1A further includes a display unit 30 in addition to the information processing device 100. The output unit 14 also includes a display control unit 16.

[0103] The display unit 30 is composed of a display device such as a liquid crystal panel. The display unit 30 displays various images under the control of the display control unit 16.

[0104] The display control unit 16 generates a map image representing one or more first maps that show the relationship between the difference in the peak positions of each of the multiple diffraction peaks included in the lattice angle feature quantity and the material properties. The display control unit 16 displays the generated map image on the display unit 30.

[0105] Figure 13 is a flowchart showing an example of the processing of the information processing system 1A in Embodiment 2 of this disclosure.

[0106] (Step S1301) The control unit 50 receives instructions from the user to select the characteristic information they wish to display from among multiple characteristic information items. These multiple characteristic information items include, for example, battery characteristics, thermal conductivity, ionic conductivity, mechanical properties, and dielectric constant. The user selects one characteristic information item from these that they wish to display.

[0107] (Step S1302) The operation unit 50 receives an instruction to select one or more lattice angle features to be displayed from among a plurality of lattice angle features. The plurality of lattice angle features include, for example, the difference in peak positions for all combinations of two diffraction peaks.

[0108] Figure 14 shows a selection screen 1500 of a first example for selecting characteristic information and lattice angle features. The selection screen 1500 includes a characteristic selection field 1501 and a lattice angle feature selection field 1504.

[0109] The characteristic selection field 1501 allows the user to input the characteristic information they wish to display from among multiple characteristic information fields. For example, when the mouse clicks while the mouse pointer (not shown in the figure) is positioned in the characteristic selection field 1501, a display list (not shown in the figure) showing multiple characteristic information fields is displayed. The user selects the characteristic information they wish to display from among the multiple characteristic information fields displayed in the display list. In the example in Figure 14, charging capacity, which is an example of a battery characteristic, is selected.

[0110] The lattice angle feature selection field 1504 is a field for selecting one or more lattice angle features to display from among multiple lattice angle features. The lattice angle feature selection field 1504 displays a list of multiple lattice angle features 1502. The lattice angle feature selection field 1504 displays a list of multiple lattice angle features 1502 that show the difference in peak positions for all possible combinations of any two assumed diffraction peaks. The lattice angle feature selection field 1504 includes multiple checkboxes 1503 corresponding to multiple lattice angle features 1502.

[0111] In the example in Figure 14, the lattice angle feature selection field 1504 displays a list of lattice angle features 1502, such as the difference between the peak position of the 018 plane and the peak position of the 012 plane, and the difference between the peak position of the 018 plane and the peak position of the 003 plane.

[0112] The user inputs an instruction using the operation unit 50 to select one or more lattice angle features 1502 that they wish to display from among several lattice angle features 1502. Specifically, the user inputs an instruction to select lattice angle features 1502 by checking the checkboxes 1503 corresponding to the one or more lattice angle features 1502 that they wish to display.

[0113] (Step S1303) The display control unit 16 generates a map image that includes a first map showing the correspondence between the characteristic information selected in step S1301 and the grid angle feature quantity 1502 selected in step S1302.

[0114] (Step S1304) The display control unit 16 displays the map image generated in step S1303 on the display unit 30. For example, the display control unit 16 displays the map image 1600 when the operation to select the display button (not shown) is input on the selection screen 1500.

[0115] Figure 15 shows the map image 1600 of the first example. Map image 1600 contains three first maps 1601. This is because three grid angle features 1502 were selected on the selection screen 1500. In other words, if n grid angle features 1502 are selected on the selection screen 1500, map image 1600 will display n first maps 1601.

[0116] The vertical axis of all three first maps 1601 represents charging capacity. This is because charging capacity was selected as characteristic information on the selection screen 1500.

[0117] The horizontal axis of the leftmost first map 1601 represents the difference between the peak position of the 018 face and the peak position of the 012 face; the horizontal axis of the center first map 1601 represents the difference between the peak position of the 018 face and the peak position of the 003 face; and the horizontal axis of the rightmost first map 1601 represents the difference between the peak position of the 018 face and the peak position of the 104 face. This is because these three grid angle features 1502 were selected on the selection screen 1500.

[0118] Each point plotted in Map 1601 corresponds to a single material. In other words, Map 1601 is a scatter plot showing the correspondence between charge capacity and selected lattice angle features for multiple materials. The correlation coefficient r is displayed for each of the three Map 1601s. The correlation coefficient r is a value that shows the correlation between each of the three lattice angle features and charge capacity. The closer the correlation coefficient r is to 1, the higher the correlation between the lattice angle feature and charge capacity. In this example, the correlation coefficient r for all three Map 1601s is greater than 0.5, confirming that each lattice angle feature is correlated with charge capacity.

[0119] Figure 16 shows the selection screen 1700 of the second example. The selection screen 1700 includes a characteristic selection field 1701, an order selection field 1702, a checkbox 1703, and a grid angle feature selection field 1704. The characteristic selection field 1701 is the same as the characteristic selection field 1501. The grid angle feature selection field 1704 displays a list of grid angle features. In detail, the grid angle feature selection field 1704 includes a display selection field, a grid angle feature name field, a correlation coefficient field, and a field for the absolute value of the correlation coefficient. The display selection field is for the user to select the grid angle features they wish to display. The grid angle feature name field displays the name of each grid angle feature. The correlation coefficient field displays the correlation coefficient of each grid angle feature. The absolute value of the correlation coefficient field displays the absolute value of the correlation coefficient.

[0120] The order selection field 1702 sorts the lattice angle features listed in the lattice angle feature selection field 1704 according to the absolute value of the correlation coefficient r. In the order selection field 1702, you can select either descending or ascending order. In this example, descending order is selected in the order selection field 1702. Therefore, the lattice angle feature selection field 1704 displays the lattice angle features in descending order of the absolute value of the correlation coefficient r. If ascending order is selected in the order selection field 1702, the lattice angle features will be displayed in ascending order of the absolute value of the correlation coefficient r.

[0121] The user selects the lattice angle features they wish to display from the lattice angle features displayed in the lattice angle feature selection field 1704 by checking the display selection field.

[0122] Figure 17 shows the map image 1800 of the second example. The display control unit 16 displays the map image 1800 when, for example, the operation of selecting the display button (not shown) is input on the selection screen 1700. The map image 1800 displays two first maps 1801. The first map 1801 on the left is a scatter plot showing the relationship between charge capacity and the difference between the peak position of the 018 plane and the peak position of the 012 plane. The first map 1801 on the right is a scatter plot showing the relationship between charge capacity and the difference between the peak position of the 018 plane and the peak position of the 104 plane. Each point in the first map 1801 corresponds to one material. These two first maps 1801 are displayed because, on the selection screen 1700, the difference between the peak position of the 018 plane and the peak position of the 012 plane, and the difference between the peak position of the 018 plane and the peak position of the 104 plane were selected as lattice angle features.

[0123] The correlation coefficient r is displayed for each of the two first maps 1801. Both correlation coefficients r are close to 0.6, confirming a correlation between charge capacity and the selected lattice angle features.

[0124] Note that map images 1600 and 1800 may include multiple secondary maps. Each of these secondary maps has a different peak position for diffraction peaks and is a scatter plot showing the relationship between peak position and characteristic information.

[0125] Figure 18 shows a map image 1900 of the third example. Map image 1900 includes two second maps 1902 and one first map 1901. The second map 1902 is a scatter plot showing the relationship between peak position and charge capacity. In the example in Figure 18, the leftmost second map 1902 shows the relationship between the O18 plane peak position and charge capacity, and the central second map 1902 shows the relationship between the O12 plane peak position and charge capacity. The first map 1901 shows the relationship between the difference between the O18 plane peak position and the O12 plane peak position and charge capacity. In both the first map 1901 and the second map 1902, one point corresponds to one material.

[0126] The correlation coefficient r is displayed for both of the second maps 1902 and the first map 1901. The correlation coefficients for both second maps 1902 are both less than 0.4, confirming that the correlation between peak position and charge capacity is low. On the other hand, the correlation coefficient r for the first map 1901 is 0.59, which is significantly larger than the correlation coefficient r for the second map 1902. Therefore, it can be confirmed that the difference in peak positions has a higher correlation with charge capacity than the peak positions themselves. Thus, it can be confirmed that lattice angle features are useful parameters as explanatory variables for materials with crystalline structures.

[0127] When the method of displaying the map image 1900 is adopted, the selection screen 1500 or selection screen 1700 may have a field for selecting a peak position. The display control unit 16 may then display the second map 1902 corresponding to the peak position selected by the user on the selection screen 1500 or selection screen 1700 on the map image 1900.

[0128] Thus, according to Embodiment 2, a scatter plot showing the relationship between the lattice angle feature of interest to the user and the characteristic information can be presented. Furthermore, the user can compare the first map and the second map to gain a more detailed understanding of the material's properties.

[0129] (Examples) Next, we will describe an example in which the correlation between multiple arithmetic operations used to calculate lattice angle features and lattice angles was investigated using XRD simulation. Figure 19 is a table summarizing the results of the example. In this example, the following arithmetic operations were used: an arithmetic operation to calculate the difference between peak positions, an arithmetic operation to calculate the difference between peak intensities, and an arithmetic operation to calculate the difference between peak position and peak intensity.

[0130] Figure 20 shows the difference between peak positions and the grid angle α. hex This is a scatter plot showing the relationship between the peak positions of plane 201 and plane 202, and the grid angle α. Specifically, Figure 20 shows the difference between the peak positions of plane 201 and plane 202, and the relationship between the grid angle α. hex This is a scatter plot showing the relationship between peak intensities and the grid angle α. Figure 21 shows the difference between peak intensities and the grid angle α. hexThis is a scatter plot showing the relationship between the peak intensity of the 105 plane and the peak intensity of the 113 plane, and the lattice angle α. Specifically, Figure 21 shows the difference between the peak intensity of the 105 plane and the peak intensity of the 113 plane, and the lattice angle α. hex This is a scatter plot showing the relationship between the peak position and the difference between the peak intensity and the grid angle α. Specifically, Figure 22 shows the difference between the peak position of the 105 plane and the peak intensity of the 108 plane, and the grid angle α. hex This is a scatter plot showing the relationship. Figure 23 is a diagram showing the hexagonal crystal structure.

[0131] As shown in Figures 19 to 22, correlation coefficients of 0.9 or higher were obtained for all arithmetic operations. Therefore, it was confirmed that lattice angle features have a high correlation with lattice angles.

[0132] This disclosure may be modified in the following ways:

[0133] (Variation 1) In Embodiment 1, the extraction unit 12 extracted lattice angle features for all combinations of any two peak positions appearing in the spectral data, but this is just one example. The extraction unit 12 may also extract lattice angle features from diffraction peaks corresponding to planes determined according to the type of crystal structure. This is expected to allow for the acquisition of features that are more closely related to material properties.

[0134] For example, in some cases, the diffraction peaks to be of interest are known in advance for each crystal structure. In this case, the extraction unit 12 calculates the spectral features of at least two diffraction peaks to be of interest according to the crystal structure. Then, the extraction unit 12 can extract the lattice angle features for all combinations of any two diffraction peaks from the at least two diffraction peaks calculated. The display control unit 16 can then display the calculated lattice angle features in a list on the selection screens 1500 and 1700. In this modified example, the extraction unit 12 may have the user input the crystal structure using the operation unit 50.

[0135] FIG. 24 is a diagram showing planes defined in the hexagonal crystal system. The planes defined in the hexagonal crystal system include, for example, the 012 plane and the 018 plane. Therefore, when the material is of the hexagonal crystal system, the extraction unit 12 sets the 012 plane and the 018 plane, and extracts the spectral feature amounts of the 012 plane and the 018 plane respectively. Then, the extraction unit 12 may calculate the lattice angle feature amount by applying an arithmetic operation to the spectral feature amount of the 012 plane and the spectral feature amount of the 018 plane.

[0136] (Modified Example 2) The crystal structure of the material may be a crystal system with a fixed lattice angle. In this case, the lattice angle feature amount can represent the change in symmetry caused by the change in the lattice angle in the crystal structure. The crystal systems with a fixed lattice angle are, for example, the hexagonal crystal system and the monoclinic crystal system. FIG. 25 is a diagram showing the change in symmetry due to the change in the lattice angle in a material of the hexagonal crystal system.

[0137] In FIG. 25, the left figure shows the crystal structure of the hexagonal crystal system, and the right figure shows the crystal structure of the monoclinic crystal system. In the hexagonal crystal system, the lattice angle β between the a mon axis and the c mon axis is 110.5 degrees. In the monoclinic crystal system, the lattice angle β between the a mon axis and the c mon axis is 110.6 degrees. The hexagonal crystal system has a higher symmetry than the monoclinic crystal system. Suppose that the lattice angle β between the a mon axis and the c mon axis obtained from the spectral data of the material of the hexagonal crystal system is 110.6 degrees. In this case, it can be seen that this material has changed from the hexagonal crystal system to the monoclinic crystal system, and the symmetry has changed accordingly. mon axis and the c mon axis is 110.6 degrees. In this case, it can be seen that this material has changed from the hexagonal crystal system to the monoclinic crystal system, and the symmetry has changed accordingly. mon axis is 110.6 degrees. In this case, it can be seen that this material has changed from the hexagonal crystal system to the monoclinic crystal system, and the symmetry has changed accordingly

[0138] An example of the lattice angle feature amount, the difference between the peak positions corresponds to the lattice angle. Therefore, when the difference between the peak positions is adopted as the lattice angle feature amount and the crystal structure of the material is known, the change in symmetry can be confirmed from the lattice angle feature amount.

[0139] (Modified Example 3) The extraction unit 12 may extract features that have a high correlation with characteristic information from among multiple lattice angle features. The database 41 stores multiple lattice angle features and characteristic information or estimated characteristic information in association for each of the multiple materials. As a result, a multiple regression analysis model is obtained that estimates characteristic information from multiple lattice angle features.

[0140] For example, suppose database 41 stores for each of several materials the battery capacity, the difference D1 between the peak position of plane 018 and the peak position of plane 012, the difference D2 between the peak position of plane 018 and the peak position of plane 003, the difference D3 between the peak position of plane 018 and the peak position of plane 104, the difference D4 between the peak position of plane 012 and the peak position of plane 003, and the difference D5 between the peak position of plane 012 and the peak position of plane 104.

[0141] In this case, the extraction unit 12 performs multiple regression analysis and calculates a multiple regression analysis model in which the differences D1 to D5 are explanatory variables and the battery capacity is the dependent variable. In the multiple regression analysis model, the extraction unit 12 extracts the lattice angle features corresponding to the difference D whose partial regression coefficient is greater than the reference value from among the differences D1 to D5 as highly correlated lattice angle features. The display control unit 16 may display the first maps 1601 and 1801, which show the relationship between the extracted highly correlated lattice angle features and the battery capacity, on the map images 1600 and 1800. Alternatively, the display control unit 16 may display the extracted highly correlated lattice angle features in a list on the selection screens 1500 and 1700. Then, the display control unit 16 may display the first maps 1601 and 1801 related to the lattice angle features selected by the user on the selection screens 1500 and 1700 on the map images 1600 and 1800.

[0142] This allows us to present users with lattice angle features that are highly correlated with the properties of the material they are interested in. [Industrial applicability]

[0143] This disclosure is useful in techniques for analyzing materials with crystalline structures because it expands the explanatory variables for materials with crystalline structures. [Explanation of Symbols]

[0144] 1: Information Processing System 10: Processor 11: Acquisition part 12:Extraction part 13: Estimation part 14: Output section 15: Learning Department 16: Display Control Unit 20:XRD measurement device 30:Display section 40: Memory 41: Database 42: Predictive Models 43: Model function 50:Operation unit 100: Information Processing Device

Claims

1. An information processing method performed by an information processing device, To obtain spectral data showing the spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure, This includes extracting lattice angle features related to the lattice angle in the spectrum from the spectral data, Information processing methods.

2. Based on the aforementioned lattice angle features, characteristic information regarding the properties of the material is estimated. The further includes outputting the estimated characteristic information, The information processing method according to claim 1.

3. Extracting the aforementioned lattice angle features means This includes extracting the lattice angle feature from the diffraction peaks in the spectrum, The information processing method according to claim 1.

4. The aforementioned lattice angle feature is calculated based on at least one of the position and intensity of each of at least two diffraction peaks. The information processing method according to claim 3.

5. The lattice angle feature quantity represents the difference or ratio of the positions of each of the at least two diffraction peaks. The information processing method according to claim 4.

6. Extracting the aforementioned lattice angle features means This includes extracting the lattice angle feature quantity from the diffraction peaks corresponding to a plane determined according to the type of crystal structure, The information processing method according to claim 3.

7. The aforementioned crystal structure has a crystal system with fixed lattice angles. The aforementioned lattice angle feature is a feature that represents the change in symmetry caused by the change in the lattice angle in the crystal structure. The information processing method according to claim 1.

8. Extracting the aforementioned lattice angle features means The process involves calculating multiple spectral feature quantities that represent the characteristics of the spectrum from the spectral data, This includes extracting the lattice angle feature by applying arithmetic operations to the plurality of spectral feature quantities, The information processing method according to claim 1.

9. Calculating the spectral features includes extracting multiple spectral features by approximating a model function to the spectral data. The information processing method according to claim 8.

10. An acquisition unit that acquires spectral data showing the spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure, The system includes an extraction unit that extracts lattice angle features relating to the lattice angle in the spectrum from the spectral data. Information processing system.

11. The system further includes a display control unit that generates an image representing one or more first maps showing the relationship between the difference in the peak positions of at least two diffraction peaks included in the lattice angle feature quantity and the properties of the material, and displays the image on a display unit. The information processing system according to claim 10.

12. The image includes a plurality of second maps in which the positions of each of the at least two diffraction peaks are different. The aforementioned multiple second maps each show the relationship between diffraction peaks and the properties of the material. The information processing system according to claim 11.

13. The system further includes an estimation unit that estimates characteristic information regarding the properties of the material by inputting the aforementioned lattice angle features into a prediction model. The prediction model is trained to take the lattice angle feature as input and output the characteristic information. The information processing system according to claim 10 or 11.

14. To obtain spectral data showing the spectrum obtained by X-ray diffraction measurement of a material having a crystalline structure, The computer is instructed to extract lattice angle features related to the lattice angle in the spectrum from the spectral data. program.

Images