A crystal experiment data analysis method and system

Abstract

The present invention relates to the technical field of crystal data processing, and in particular to a method and system for analyzing crystal experimental data. The method comprises the steps of: obtaining a preset neighboring segment of the data point with a data point in a crystal XRD data sequence as the center, obtaining the local symmetry of the data point by the intensity change in the neighboring segment of the data point; calculating the possible degree of the diffraction peak of the data point; obtaining the noise degree of the data point by the mean value of the intensity difference between the adjacent data point and the previous group of adjacent data points in the neighboring segment of the data point; calculating the polynomial order adjustment coefficient of the sliding window; weighting the preset polynomial order by the polynomial order adjustment coefficient of the sliding window to obtain the polynomial order of each sliding window; using the polynomial order of the sliding window in the S-G filtering algorithm to filter the XRD data sequence to realize crystal experimental data analysis, and effectively improve the accuracy of crystal experimental data analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of crystal data processing, and in particular to a crystal experimental data analysis method and system. Background Art

[0002] Crystal refers to a solid material in which atoms, ions or molecules are arranged in an orderly manner according to specific geometric rules. The orderly arrangement of atoms, ions or molecules inside it gives it unique physical and chemical properties. X-ray diffraction (XRD) of crystals can be used to identify and analyze crystal structures. Analyzing XRD patterns can help researchers determine the structure of unknown crystals, conduct material identification, property evaluation, and phase change research.

[0003] There are many studies on the collected crystal spectrum data in the prior art. For example, the patent application document with publication number CN116230123A discloses a spectrum data processing device and its simulated spectrum curve simulation system and method. The application obtains the first fitting spectrum curve of the specified crystal and the first energy peak in the first fitting spectrum curve, establishes the correlation between the first fitting spectrum curve and the original spectrum data, and obtains the crystal energy peak of the specified crystal based on the correlation, establishes an energy spectrum curve database, and then obtains the error between the simulated spectrum curve and the actual spectrum curve to ensure the reliability and accuracy of the simulated spectrum curve simulation system.

[0004] The above-mentioned prior art reduces the problem of inaccurate results of statistical analysis of energy spectrum curves by obtaining the error between the simulated energy spectrum curve and the actual energy spectrum curve. However, when collecting the XRD spectrum of the crystal, there may be noise interference in its surrounding environment, resulting in poor accuracy of the obtained crystal energy spectrum data, affecting the accuracy of subsequent analysis of the crystal energy spectrum data.

[0005] Based on this, how to accurately denoise crystal experimental data is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the technical problem of how to accurately achieve denoising of crystal experimental data, the present invention provides a crystal experimental data analysis method and system.

[0007] In a first aspect, the present invention provides a crystal experiment data analysis method, which adopts the following technical solution:

[0008] A crystal experiment data analysis method comprises the steps of:

[0009] Taking a data point in a crystal XRD data sequence as the center, a preset neighboring segment of the data point is obtained, and the local symmetry of the data point is obtained through the intensity change in the neighboring segment of the data point; the intensity mean of each data point in the neighboring segment is recorded as the first mean, and the absolute value mean of the intensity difference between adjacent data points in the neighboring segment is recorded as the second mean; the local symmetry of the data point is normalized by the product of the first mean and the second mean to obtain the possible degree of the diffraction peak of the data point; the noise degree of the data point is obtained through the mean of the intensity difference between the adjacent data point in the neighboring segment of the data point and the previous group of adjacent data points; ; For the The polynomial order adjustment coefficient of the sliding window, is a hyperparameter, is the sliding window length, is the sliding window ordinal number, , are respectively the possible degree of diffraction peak and the noise degree of the ith data point; the preset polynomial order is weighted by the polynomial order adjustment coefficient of the sliding window to obtain the polynomial order of each sliding window; the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to realize crystal experimental data analysis.

[0010] The present invention takes into account possible noise interference in the XRD data of the crystal, which affects the accuracy of the crystal data analysis. Therefore, the XRD data sequence of the crystal is filtered and denoised by the SG filtering algorithm, and then the data analysis is performed to improve the accuracy of the crystal data analysis. In this process, the present invention takes into account that the fixed polynomial order used in the SG filtering algorithm may cause noise fitting or crystal diffraction peak distortion; based on this, the present invention obtains the possible degree and noise degree of the diffraction peak of the data point by analyzing the intensity change characteristics in the neighboring fragments of the data point, and determines the polynomial order of the sliding window by the possible degree and noise degree of the diffraction peak of the data point in the sliding window, which can effectively improve the accuracy of the crystal XRD data denoising, thereby effectively improving the accuracy of the crystal experimental data analysis.

[0011] According to a crystal experimental data analysis method provided by the present invention, the preset neighbor fragment of the data point is obtained with the data point in the crystal XRD data sequence as the center, and the method also includes: collecting the XRD data of the crystal through a preset diffraction angle value range and collection frequency, taking the intensity of the crystal X-ray diffraction obtained each time as a data point, and performing preprocessing to obtain the crystal XRD data sequence.

[0012] The present invention takes into account the possibility of data missing in the originally collected data, and therefore improves the overall quality of the data through preprocessing to facilitate subsequent data processing.

[0013] According to a crystal experimental data analysis method provided by the present invention, the local symmetry of the data point satisfies the relationship:

[0014] ;

[0015] is the local symmetry of the ith data point, is the radius of the nearest fragment, , , , Respectively , , , The strength of the data point, is the linear normalization function, To preserve positive value function.

[0016] The present invention provides an accurate calculation method for the local symmetry of a data point, and obtains the symmetry on both sides of the data point by analyzing the data change gap on both sides of the data point.

[0017] According to a crystal experiment data analysis method provided by the present invention, the noise level of the data point satisfies the relationship: ; is the noise level of the ith data point, , , Respectively , , The strength of the data point, is the radius of the nearest fragment.

[0018] The present invention provides an accurate method for calculating the noise level of a data point. By analyzing the overall intensity change in the neighboring segments of the data point, the noise level of the data point can be accurately obtained.

[0019] According to a crystal experimental data analysis method provided by the present invention, the preset polynomial order is weighted by the polynomial order adjustment coefficient of the sliding window, including: rounding the product of the polynomial order adjustment coefficient of the sliding window and the preset polynomial order to obtain the polynomial order of the sliding window.

[0020] According to a crystal experimental data analysis method provided by the present invention, the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence, including: performing polynomial fitting on the intensity of the data points in the sliding window based on the polynomial order, taking the polynomial value in the sliding window when a data point is located at the center of the sliding window as the filtering value of the data point, and obtaining a filtered XRD data sequence.

[0021] According to a crystal experimental data analysis method provided by the present invention, the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to achieve crystal experimental data analysis, including: extracting the diffraction peak of the crystal from the filtered XRD data sequence through the AMPD algorithm and matching it with the crystal standard diffraction peak to obtain the crystal experimental data analysis result.

[0022] The present invention performs crystal experimental data analysis based on the denoised XRD data sequence, which can effectively avoid the influence of noise data on the analysis results and improve the accuracy of the obtained crystal experimental data analysis results.

[0023] In a second aspect, the present invention provides a crystal experiment data analysis system, which adopts the following technical solution:

[0024] A crystal experiment data analysis system comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned crystal experiment data analysis method is implemented.

[0025] By adopting the above technical solution, the above-mentioned crystal experimental data analysis method is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0026] The present invention has the following technical effects:

[0027] Based on the above technical solution, when determining the crystal experimental data analysis results, the present invention uses the SG filtering algorithm to filter and denoise the XRD data sequence of the crystal and then performs data analysis, which can effectively improve the accuracy of crystal data analysis. In this process, the present invention takes into account that the fixed polynomial order used in the SG filtering algorithm may cause noise fitting or crystal diffraction peak distortion; based on this, the present invention obtains the possible degree and noise degree of the diffraction peak of the data point by analyzing the intensity change characteristics in the neighboring fragments of the data point, and determines the polynomial order of the sliding window by the possible degree and noise degree of the diffraction peak of the data point in the sliding window, which can effectively improve the accuracy of crystal XRD data denoising, thereby effectively improving the accuracy of crystal experimental data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0029] Figure 1 A schematic flow chart of a crystal experiment data analysis method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0031] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0032] Crystal refers to a solid material in which atoms, ions or molecules are arranged in an orderly manner according to specific geometric rules. The orderly arrangement of atoms, ions or molecules inside it gives it unique physical and chemical properties. The XRD data of crystals can be used to identify and analyze the crystal structure. When the sample is irradiated with X-rays with sufficient energy, the substances in the sample are excited and produce secondary fluorescent X-rays. The crystal surface reflection follows the Bragg law. Qualitative analysis of crystals can be performed by measuring the diffraction angle position, quantitative analysis of crystals can be performed by measuring the peak intensity of the spectrum line, and the size and shape of the grains can be detected by measuring the relationship between the intensity of the spectrum line and the angle.

[0033] However, when collecting XRD data of a crystal, there may be noise interference in its surrounding environment, resulting in poor accuracy of the obtained crystal data, affecting the accuracy of subsequent crystal data analysis. The Savitzky-Golay (SG) filtering algorithm is a technique for smoothing data and extracting useful features from signals. It achieves data smoothing and denoising by fitting a polynomial in a sliding window.

[0034] Based on this, the embodiment of the present invention discloses a crystal experimental data analysis method, which denoises the crystal XRD data sequence through the SG filtering algorithm, and analyzes the denoised crystal XRD data sequence, which can effectively improve the accuracy of crystal data analysis. Figure 1 As shown, Figure 1 A flow chart of a crystal experiment data analysis method provided in an embodiment of the present invention, the method specifically comprises the following steps.

[0035] S1: Taking a data point in the crystal XRD data sequence as the center, a preset neighboring fragment of the data point is obtained.

[0036] It should be noted that the shape of the diffraction peak in the crystal XRD data sequence can directly reflect the crystal structure. In order to ensure the accuracy of subsequent crystal data analysis, when removing the noise of the crystal XRD data through SG filtering, it is necessary to retain the characteristics of the diffraction peak while denoising. However, the traditional SG filtering uses a fixed-size preset polynomial order for filtering. If the polynomial order is too small, the diffraction peak after denoising will be distorted; when the polynomial order is too large, the noise will be overfitted, resulting in low accuracy in subsequent identification of the crystal structure based on the diffraction peak.

[0037] Based on this, when removing crystal XRD data noise based on SG filtering, the embodiment of the present invention can first obtain the position of the crystal diffraction peak, and adjust the polynomial order in the SG filtering according to the possibility that each data point is a diffraction peak.

[0038] By way of example, in an embodiment of the present invention, a preset neighboring fragment of a data point in a crystal XRD data sequence is obtained with the data point in the data sequence as the center, which also includes: collecting the XRD data of the crystal through a preset diffraction angle value range and collection frequency, taking the intensity of the crystal X-ray diffraction obtained each time as a data point, and performing preprocessing to obtain the crystal XRD data sequence.

[0039] The diffraction angle range can be set to , the angle sampling frequency can be set to , that is, 200 data points are collected at equal intervals when the diffraction angle sweeps over 1°; the diffraction angle value range and the sampling frequency can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0040] Among them, XRD data of the crystal can be collected by an X-ray diffractometer.

[0041] For example, the preprocessing method may be missing data interpolation, data format conversion, etc., which may be specifically set according to actual needs.

[0042] For example, in an embodiment of the present invention, a crystal XRD data spectrum can be constructed with the diffraction angle corresponding to the data point in the crystal XRD data sequence as the horizontal coordinate and the intensity corresponding to the data point as the vertical coordinate to obtain the diffraction peak in the crystal XRD data sequence.

[0043] It should be further explained that the intensity of the data points on both sides of the crystal diffraction peak is first gradually increased on the rising edge, and then gradually decreased on the falling edge. The rising edge and the falling edge are highly symmetrical, and the data points closer to the peak are more likely to be data points on the diffraction peak. However, due to the noise interference in the crystal XRD data, the rising and falling edges of the diffraction peak will have waveform changes that do not conform to the characteristics of the diffraction peak. If the position of the diffraction peak is determined directly based on the intensity performance of the rising and falling edges, the diffraction peak position may be misidentified.

[0044] Based on this, the embodiment of the present invention can analyze the intensity changes in the neighboring fragments on both sides of each data point in the crystal XRD data sequence, obtain the degree of symmetry on both sides of each data point, and determine the possible degree of diffraction peak of the data point through the degree of symmetry of each data point.

[0045] For example, the radius of the neighbor segment of a data point may be preset to 5; the radius may be specifically set according to actual needs.

[0046] Specifically, when obtaining the neighboring fragments of a data point, the current data point can be taken as the center and data points with a collection interval not greater than the neighboring fragment radius can be obtained within the neighboring fragment radius of the current data point as the neighboring fragments of the current data point.

[0047] It can be understood that the acquisition interval between adjacent data points is 1, and the obtained neighbor segment of the data point includes the data point itself. If the radius of the neighbor segment of the data point is 5, the length of the neighbor segment of the data point is 11.

[0048] After obtaining the neighboring fragments of each data point in the crystal XRD data sequence based on the above steps, the intensity changes of other data points in the neighboring fragments of the data point can be analyzed through the following steps to obtain the possible degree of the diffraction peak of the data point.

[0049] S2: The local symmetry of the data point is obtained through the intensity change in the neighboring segments of the data point; the intensity mean of each data point in the neighboring segments is recorded as the first mean, and the absolute value mean of the intensity difference between adjacent data points in the neighboring segments is recorded as the second mean; the local symmetry of the data point is normalized by the product of the first mean and the second mean to obtain the possible degree of diffraction peak of the data point.

[0050] It should be noted that the diffraction peak appears as a waveform that increases first and then decreases. The smaller the difference between the increase on the left and the decrease on the right, the more likely the data point is to be near the peak. The diffraction peak waveform is slender, and the closer the data point is to the peak, the more likely it is to be a data point on the diffraction peak.

[0051] Based on this, an embodiment of the present invention obtains data points on the left and right sides of a data point to construct a neighboring segment of the data point, obtains the local symmetry of the data point by analyzing the data intensity changes in the neighboring segment, and based on this, obtains the possible degree of diffraction peak of each data point.

[0052] For example, in an embodiment of the present invention, the local symmetry of a data point is determined, and specific reference may be made to the following relationship:

[0053] ;

[0054] is the local symmetry of the ith data point, is the radius of the nearest fragment, For the The strength of the data point, For the The strength of the data point, For the The strength of the data point, For the The strength of the data point, is the linear normalization function, To preserve positive functions, is the absolute value symbol.

[0055] In the above formula, when the variable in the retain positive function is greater than 0, the retain positive function outputs the original value normally; when the variable in the retain positive function is not greater than 0, the retain positive function outputs 0.

[0056] It can be understood that the left side of the diffraction peak in the crystal XRD data is an increase, and the right side is a decrease. The data points closer to the diffraction peak position are more consistent with the intensity changes on both sides of the diffraction peak. or If it is greater than 0, it means that it conforms to the intensity variation on both sides of the diffraction peak. Such data points can contribute to the local symmetry of the data points and need to be retained. or If it is not greater than 0, it means that it does not conform to the intensity change on both sides of the diffraction peak. Such data points will destroy the local symmetry of the data points, so they need to be set to 0 to reduce the impact on the local symmetry of the data points.

[0057] It indicates the difference in intensity change between the data points at symmetrical positions on both sides of the i-th data point. The larger the value, the lower the degree of symmetry on both sides of the i-th data point.

[0058] After obtaining the local symmetry of each data point in the crystal XRD data sequence based on the above formula, the possible degree of its diffraction peak can be obtained based on the local symmetry of the data point.

[0059] For example, in an embodiment of the present invention, the possible degree of the diffraction peak of a data point is determined, and specifically, the following relationship can be referred to:

[0060] ;

[0061] is the possible degree of diffraction peak of the ith data point, is the radius of the nearest fragment, For the The strength of the data point, is the local symmetry of the ith data point, For the The strength of the data point, is the linear normalization function, is the absolute value symbol.

[0062] In the above formula, represents the mean intensity of each data point in the neighboring fragments of the ith data point, that is, the first mean value of the ith data point, It represents the absolute value mean of the intensity difference between adjacent data points in the neighborhood of the ith data point, that is, the second mean of the ith data point. When the first mean of the ith data point is higher and the second mean of the ith data point weighted by the local symmetry is higher, it means that the intensity at the ith data point is relatively high, and the difference between the increase on the left side and the decrease on the right side of the ith data point is smaller, and the overall change of the surrounding data points is larger, that is, the ith data point is The higher the probability that a data point is located in the diffraction peak region and near the peak value of the diffraction peak, the higher the probability of the diffraction peak of the corresponding i-th data point.

[0063] After obtaining the possible degree of diffraction peak of each data point in the crystal XRD data sequence based on the above steps, the polynomial order of the sliding window can be adjusted by the possible degree of diffraction peak of the data point to achieve accurate denoising of the crystal XRD data, that is, continue to perform the following steps.

[0064] S3: The noise level of the data point is obtained by taking the mean of the intensity differences between the adjacent data points in the neighborhood segments of the data point and the previous group of adjacent data points.

[0065] It should be noted that a low polynomial order will lead to distortion of the diffraction peak, and a high polynomial order will lead to fitting noise. The noise in the crystal XRD data sequence comes from the inherent noise of the instrument and the environmental noise. Compared with the signal composed of the diffraction peak, this type of noise signal is more stable. Therefore, the faster the change of the diffraction signal, the higher the proportion of the noisy signal. In order to avoid the distortion of the diffraction peak as much as possible without fitting the noise, a higher polynomial order can be set during denoising.

[0066] Based on this, the embodiment of the present invention can analyze the amplitude changes before and after the data point in the neighborhood segment of the data point to obtain the noise level of each data point.

[0067] For example, in the embodiment of the present invention, the noise level of a data point is determined, and specifically, the following relationship can be referred to:

[0068] ;

[0069] is the noise level of the ith data point, For the The strength of the data point, For the The strength of the data point, For the The strength of the data point, is the radius of the nearest fragment, is the linear normalization function, is the absolute value symbol.

[0070] In the above formula, It represents the intensity difference between a group of adjacent data points in the i-th data point and the previous group of adjacent data points. When the intensity difference between all adjacent data points in the neighboring segment of the i-th data point and the previous group of adjacent data points is small as a whole, it means that the intensity change in the neighboring segment of the i-th data point is relatively smooth, the possibility that the i-th data point is noise data is smaller, and the corresponding noise degree is smaller.

[0071] Take an example to illustrate the adjacent data points in the nearest neighbor segment and the previous set of adjacent data points: if i=4, is 2, then the neighboring segment of the current data point contains 5 data points, then the 1st data point and the 2nd data point, the 2nd data point and the 3rd data point, the 3rd data point and the 4th data point, the 4th data point and the 5th data point in the neighboring segment are adjacent data points, and there are 4 groups of adjacent data points in total. Among them, the previous group of adjacent data points of the 4th data point and the 5th data point is the 3rd data point and the 4th data point, the previous group of adjacent data points of the 3rd data point and the 4th data point is the 2nd data point and the 3rd data point, and the previous group of adjacent data points of the 2nd data point and the 3rd data point is the 1st data point and the 2nd data point. There are 4 such combinations in total. indivual.

[0072] After obtaining the noise level of each data point based on the above steps, continue to perform the following steps.

[0073] S4: calculating the polynomial order adjustment coefficient of each sliding window in the XRD data sequence; weighting the preset polynomial order by the polynomial order adjustment coefficient of the sliding window to obtain the polynomial order of each sliding window.

[0074] Among them, the size of the sliding window can be set to 51, and the sliding step of the sliding window can be set to 1; they can be set specifically according to actual needs.

[0075] It should be noted that the diffraction peak probability and noise degree of the data point can be obtained based on the above steps. When filtering by SG, for areas with lower noise level and higher diffraction peak probability, the polynomial order in the sliding window needs to be increased to ensure that the diffraction peak is not distorted while not fitting the noise.

[0076] Based on this, the embodiment of the present invention determines the polynomial order adjustment coefficient of the sliding window by the possible degree of diffraction peak and noise degree of each data point in the sliding window, so as to accurately obtain the polynomial order of the data points in the sliding window when denoising.

[0077] For example, in the embodiment of the present invention, the polynomial order adjustment coefficient of the sliding window is determined, and the specific relationship can be referred to as follows:

[0078] ;

[0079] For the The polynomial order adjustment coefficient of the sliding window, is a hyperparameter, is the sliding window length, is the sliding window ordinal number, is the possible degree of diffraction peak of the ith data point, is the noise level of the ith data point.

[0080] Among them, the hyperparameter can be set to 2; the hyperparameter is used to control the value range of the polynomial order adjustment coefficient. The specific settings can be made according to actual needs.

[0081] In the above formula, the greater the probability of the diffraction peak of the data point in the sliding window and the smaller the noise level, the greater the polynomial order adjustment coefficient of the sliding window.

[0082] By way of example, in an embodiment of the present invention, the preset polynomial order is weighted by the polynomial order adjustment coefficient of the sliding window, including: rounding off the product of the polynomial order adjustment coefficient of the sliding window and the preset polynomial order to obtain the polynomial order of the sliding window.

[0083] Among them, the preset polynomial order can be set to 5, which can be set according to actual needs.

[0084] After the polynomial order of each sliding window in the XRD data sequence is obtained based on the above steps, denoising can be performed based on the polynomial order of the sliding window.

[0085] S5: The polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to achieve crystal experimental data analysis.

[0086] By way of example, in an embodiment of the present invention, filtering an XRD data sequence using the polynomial order of a sliding window in an SG filtering algorithm includes: performing polynomial fitting on the intensities of data points in the sliding window based on the polynomial order, using the polynomial value in the sliding window when a data point is located at the center of the sliding window as the filtering value of the data point, to obtain a filtered XRD data sequence.

[0087] Among them, for the data points at the endpoints of the XRD data sequence that cannot be located at the center of the sliding window, they can be extended by mirroring.

[0088] The specific steps of using the polynomial order of the sliding window to perform filtering and denoising in the SG filtering algorithm can be implemented by the existing technology, and the embodiments of the present invention will not be described in detail here.

[0089] After accurately achieving denoising of the crystal XRD data sequence based on the above steps, crystal data analysis can be performed based on the denoised XRD data sequence.

[0090] By way of example, in an embodiment of the present invention, the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to achieve crystal experimental data analysis, including: extracting the diffraction peak of the crystal from the filtered XRD data sequence through the AMPD algorithm and matching it with the crystal standard diffraction peak to obtain the crystal experimental data analysis results.

[0091] Among them, the diffraction peak of the crystal extracted from the filtered XRD data sequence can be matched with the standard diffraction peak of the crystal through cosine similarity, correlation coefficient, residual, etc., which can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0092] Specifically, when the diffraction peak of the crystal is extracted from the filtered XRD data sequence by the AMPD algorithm, a multi-scale profile matrix of the filtered XRD data sequence can be constructed by the AMPD algorithm, and a cumulative profile vector can be obtained according to each scale profile matrix; an empirical threshold is determined by the maximum and minimum values ​​in the cumulative profile vector, and data points whose cumulative profile vectors are less than the threshold are selected as peak values. With the peak value as the center, peak fitting is performed by a function under different diffraction peak angle ranges, and the angle range is adjusted according to the residual between the fitted peak and the XRD data, thereby realizing diffraction peak extraction.

[0093] The empirical threshold can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0094] For example, the peak fitting may be implemented by a Gaussian function, a Lorentzian function or a pseudo-Voigt function, which may be specifically set according to actual needs.

[0095] The specific implementation method of extracting the diffraction peak of the crystal from the filtered XRD data sequence by using the AMPD algorithm can be realized by the existing technology, and the embodiment of the present invention will not be described in detail here.

[0096] It can be seen that in the embodiment of the present invention, when determining the analysis result of the crystal test data, a preset neighboring segment of the data point can be obtained with the data point in the crystal XRD data sequence as the center, and the local symmetry of the data point can be obtained through the intensity change in the neighboring segment of the data point; the intensity mean of each data point in the neighboring segment is recorded as the first mean, and the absolute value mean of the intensity difference between adjacent data points in the neighboring segment is recorded as the second mean; the local symmetry of the data point is normalized by the product of the first mean and the second mean to obtain the possible degree of the diffraction peak of the data point; the noise degree of the data point is obtained through the mean of the intensity difference between the adjacent data point in the neighboring segment of the data point and the previous group of adjacent data points; ; For the The polynomial order adjustment coefficient of the sliding window, is a hyperparameter, is the sliding window length, is the sliding window ordinal number, , are respectively the possible degree of diffraction peak and the noise degree of the ith data point; the preset polynomial order is weighted by the polynomial order adjustment coefficient of the sliding window to obtain the polynomial order of each sliding window; the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to realize crystal experimental data analysis.

[0097] In this way, the embodiment of the present invention performs data analysis after filtering and denoising the XRD data sequence of the crystal through the SG filtering algorithm, which can effectively improve the accuracy of crystal data analysis. In this process, the embodiment of the present invention takes into account that the fixed polynomial order used in the SG filtering algorithm may cause noise fitting or crystal diffraction peak distortion; based on this, the embodiment of the present invention obtains the possible degree and noise degree of the diffraction peak of the data point by analyzing the intensity change characteristics in the neighboring fragments of the data point, and determines the polynomial order of the sliding window through the possible degree and noise degree of the diffraction peak of the data point in the sliding window, which can effectively improve the accuracy of crystal XRD data denoising, thereby effectively improving the accuracy of crystal experimental data analysis.

[0098] An embodiment of the present invention further discloses a crystal experiment data analysis system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a crystal experiment data analysis method provided by the present invention is implemented.

[0099] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0100] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM, a dynamic random access memory DRAM, a static random access memory SRAM, an enhanced dynamic random access memory EDRAM, a high bandwidth memory HBM, a hybrid memory cube HMC, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0101] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0102] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A crystal experiment data analysis method, characterized in that: include: Taking a data point in the crystal XRD data sequence as the center, a preset neighboring segment of the data point is obtained, and the local symmetry of the data point is obtained through the intensity change in the neighboring segment of the data point; The intensity mean of each data point in the neighboring segment is recorded as the first mean, and the absolute value mean of the intensity difference between adjacent data points in the neighboring segment is recorded as the second mean; the local symmetry of the data point is normalized by the product of the first mean and the second mean to obtain the possible degree of the diffraction peak of the data point; The noise level of the data point is obtained by taking the mean of the intensity differences between the adjacent data points in the neighboring segments of the data point and the previous group of adjacent data points; ； For the The polynomial order adjustment coefficient of the sliding window, is a hyperparameter, is the sliding window length, is the sliding window ordinal number, , are the possible degree of diffraction peak and the degree of noise of the ith data point respectively; The preset polynomial order is weighted by the polynomial order adjustment coefficient of the sliding window to obtain the polynomial order of each sliding window; the polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to realize crystal experimental data analysis; The local symmetry of the data points satisfies the relationship: ; is the local symmetry of the ith data point, is the radius of the nearest fragment, , , , Respectively , , , The strength of the data points, is the linear normalization function, To preserve positive value function.

2. A crystal experiment data analysis method according to claim 1, characterized in that: The method of obtaining a preset neighboring fragment of a data point centered on a data point in the crystal XRD data sequence also includes: The XRD data of the crystal is collected by presetting the diffraction angle value range and collection frequency, and the intensity of the crystal X-ray diffraction obtained each time is taken as a data point, and preprocessing is performed to obtain the XRD data sequence of the crystal.

3. A crystal experiment data analysis method according to claim 2, characterized in that: The noise level of the data point satisfies the relationship: ； is the noise level of the ith data point, , , Respectively , , The strength of the data points, is the radius of the nearest fragment.

4. A crystal experiment data analysis method according to claim 1, characterized in that: The step of weighting the preset polynomial order by the polynomial order adjustment coefficient of the sliding window includes: The product of the polynomial order adjustment coefficient of the sliding window and the preset polynomial order is rounded to an integer to obtain the polynomial order of the sliding window.

5. A crystal experiment data analysis method according to claim 1, characterized in that: The method of filtering the XRD data sequence using the polynomial order of the sliding window in the SG filtering algorithm includes: The intensity of the data points in the sliding window is fitted with a polynomial based on the polynomial order, and the polynomial value in the sliding window when a data point is located at the center of the sliding window is used as the filtering value of the data point to obtain a filtered XRD data sequence.

6. A crystal experiment data analysis method according to claim 5, characterized in that: The polynomial order of the sliding window is used in the SG filtering algorithm to filter the XRD data sequence to achieve crystal experimental data analysis, including: The AMPD algorithm is used to extract the diffraction peaks of the crystal from the filtered XRD data sequence and match them with the standard diffraction peaks of the crystal to obtain the crystal experimental data analysis results.

7. A crystal experiment data analysis system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a crystal experiment data analysis method according to any one of claims 1 to 6 is implemented.

Images