A computer-aided method and apparatus for measuring the particle size of lithium dihydrogen phosphate

Abstract

The present application relates to the technical field of using visual analysis materials, and particularly relates to a computer-aided lithium dihydrogen phosphate particle size determination method and device, the method comprising: obtaining the equivalent circle diameter of lithium dihydrogen phosphate of a current production batch; determining the equivalent circle diameter difference between a reference historical production batch associated with the current production batch under a measurement condition and a true condition, obtaining the correction degree of the current production batch in combination with the reference weight of the reference historical production batch on the current production batch; and correcting the equivalent circle diameter of lithium dihydrogen phosphate of the current production batch according to the correction degree, effectively compensating for systematic deviation in optical imaging of lithium dihydrogen phosphate, reducing the crystal particle size distortion degree in the lithium dihydrogen phosphate image, and improving the accuracy of the lithium dihydrogen phosphate particle size determination result.

Description

Technical Field

[0001] This invention relates to the field of visual analysis of materials, specifically to a computer-aided method and apparatus for determining the particle size of lithium dihydrogen phosphate. Background Technology

[0002] Lithium dihydrogen phosphate (LiH2PO4) is an inorganic compound, a colorless crystalline solid, soluble in water and alcohols. As a core precursor for lithium-ion battery cathode materials, its particle size distribution directly affects the rheological properties, coating uniformity, and cycle stability of the electrode slurry. Current methods for analyzing the particle size of LiH2PO4 using microscopic images involve capturing particle images with an optical fiber microscope to segment the particles and determine their size. However, the high transparency of LiH2PO4 causes scattering halos during imaging, blurring the edges of the particle regions and distorting the particle size in the images, resulting in low accuracy in particle size determination. Summary of the Invention

[0003] To address the technical problem of low accuracy in existing lithium dihydrogen phosphate particle size determination results, the present invention aims to provide a computer-aided method and apparatus for determining the particle size of lithium dihydrogen phosphate. The specific technical solution adopted is as follows: In a first aspect of the present invention, a computer-aided method for determining the particle size of lithium dihydrogen phosphate is provided, comprising: Obtain the equivalent circle diameter of the current production batch of lithium dihydrogen phosphate; The equivalent circle diameter difference between measured and actual conditions is determined for a reference historical production batch associated with the current production batch. Combined with the particle concentration and average roundness of the reference historical production batch, a reference weight is obtained for the current production batch. Based on the particle concentration of the reference historical production batch, the reference weight of the current production batch, and the similarity of the process characteristic parameters of the current production batch, the correction degree of the current production batch is determined. The similarity of the process characteristic parameters of the current production batch is the similarity of the process characteristic parameters of the historical production batch and the current production batch in quality inspection. The equivalent circle diameter of lithium dihydrogen phosphate in the current production batch is corrected according to the correction degree.

[0004] In an exemplary embodiment, before obtaining the equivalent circular diameter of the current production batch of lithium dihydrogen phosphate, the lithium dihydrogen phosphate particle size determination method further includes: Obtain microscopic images of lithium dihydrogen phosphate from the current production batch; The lithium dihydrogen phosphate microscopic image was segmented using a trained convolutional neural network model to obtain the particle regions of lithium dihydrogen phosphate. The step of obtaining the equivalent circular diameter of lithium dihydrogen phosphate in the current production batch includes: obtaining the equivalent circular diameter of the particle region of lithium dihydrogen phosphate in the current production batch.

[0005] In one exemplary embodiment, the process of obtaining the reference historical production batch includes: Determine the process characteristic parameters for the current production batch and several historical production batches respectively; Determine the similarity of process feature parameters between each historical production batch and the current production batch, and use historical production batches whose process feature parameter similarity is greater than a preset similarity threshold as reference historical production batches associated with the current production batch.

[0006] In an exemplary embodiment, the process of obtaining the reference weights includes: A negative correlation mapping is performed on the particle concentration, and the product of the negative correlation mapping result and the particle concentration of lithium dioxane in the reference historical production batch is used as the reference weight of the reference historical production batch corresponding to the current batch.

[0007] In an exemplary embodiment, the process of obtaining the degree of correction includes: The correction influence factor for each reference historical production batch is obtained from the difference in equivalent circle diameter and the reference weight of each reference historical production batch; the correction influence factor is positively correlated with the difference in equivalent circle diameter and negatively correlated with the reference weight. The correction degree of the current production batch is obtained based on the correction influence factors of each reference historical production batch and the similarity of process characteristic parameters with the current production batch.

[0008] In an exemplary embodiment, obtaining the correction degree of the current production batch based on the correction influence factors of each reference historical production batch and the similarity of process characteristic parameters with the current production batch includes: The correction degree performance index for each reference historical production batch is obtained by using the correction impact factor and process characteristic parameter similarity of each reference historical production batch; the correction degree performance index is positively correlated with both the correction impact factor and the process characteristic parameter similarity. The correction level of the current production batch is obtained by integrating the correction level performance indicators of each reference historical production batch.

[0009] In an exemplary embodiment, the difference in equivalent circle diameter is the increase in the first average equivalent circle diameter relative to the second average equivalent circle diameter of a historical production batch; the first average equivalent circle diameter is the average equivalent circle diameter of lithium dihydrogen phosphate under measurement conditions, and the second average equivalent circle diameter is the average equivalent circle diameter of lithium dihydrogen phosphate under actual conditions.

[0010] In an exemplary embodiment, the step of correcting the equivalent circular diameter of the current production batch of lithium dihydrogen phosphate according to the correction degree includes: Based on the stated correction level, a correction coefficient for the current production batch of lithium dihydrogen phosphate is obtained; the correction coefficient is inversely correlated with the stated correction level. Multiplying the equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch by the correction coefficient yields the corrected equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch.

[0011] In an exemplary embodiment, after correcting the equivalent circular diameter of the current production batch of lithium dihydrogen phosphate according to the correction degree, the lithium dihydrogen phosphate particle size determination method further includes: Based on the corrected equivalent circle diameter of each particle region of the current production batch of lithium dihydrogen phosphate, determine the number of particle regions within each preset particle size range. By performing curve fitting on the number of particle regions within each preset particle size range, the particle size distribution curve of lithium dihydrogen phosphate in the current production batch is obtained.

[0012] In a second aspect of the present invention, a computer-aided lithium dihydrogen phosphate particle size determination device is provided, comprising: a memory and a processor; the memory is connected to the processor; the memory is used to store program instructions; the processor is used to implement the above-described computer-aided lithium dihydrogen phosphate particle size determination method when the program instructions are executed.

[0013] The present invention has the following beneficial effects: by introducing the measurement data of lithium dihydrogen phosphate from a reference historical production batch associated with the current production batch, and based on the difference in the equivalent circle diameter between the reference historical production batch and the actual situation, as well as the reference weight of the reference historical production batch for the current production batch, the correction degree of the current production batch can be obtained. This enables reliable correction of the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch, effectively compensating for systematic deviations in the optical imaging of lithium dihydrogen phosphate, reducing the degree of crystal grain size distortion in the microscopic image of lithium dihydrogen phosphate, and improving the accuracy of the particle size measurement results of lithium dihydrogen phosphate. Attached Figure Description

[0014] Figure 1This is a flowchart of a computer-aided method for determining the particle size of lithium dihydrogen phosphate according to an embodiment of the present invention; Figure 2 This is a flowchart of the process for obtaining reference historical production batches provided in one embodiment of the present invention; Figure 3 This is a flowchart illustrating the process of obtaining reference weights according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the process of obtaining the correction level according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating the specific implementation of step S3 provided in one embodiment of the present invention; Figure 6 This is a flowchart of the steps included in a computer-aided method for determining the particle size of lithium dihydrogen phosphate provided in one embodiment of the present invention; Figure 7 This is a schematic diagram of the particle size distribution of lithium dihydrogen phosphate before particle size correction, provided in one embodiment of the present invention; Figure 8 This is a schematic diagram of the particle size distribution of lithium dihydrogen phosphate after particle size correction, provided in one embodiment of the present invention. Detailed Implementation

[0015] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. All data and information collected in this application have been obtained with full consent.

[0017] This embodiment provides a computer-aided method for determining the particle size of lithium dihydrogen phosphate. The purpose is to improve the accuracy of lithium dihydrogen phosphate particle size determination by using AI (Artificial Intelligence) technology to assist in the particle size detection of lithium dihydrogen phosphate.

[0018] In a specific application scenario, lithium dihydrogen phosphate (LiH2PO4) is produced in batches. For any given batch, a high-speed microscopic imaging module is placed inside a closed dispersion chamber. Within this chamber, LiH2PO4 is dispersed into a continuous stream of particles. The high-speed microscopic imaging module then captures images of the LiH2PO4, resulting in microscopic images. For any given batch, either a single microscopic image can be captured, or a continuous video stream of microscopic images can be obtained. It should be understood that the captured microscopic images of LiH2PO4 must meet high resolution requirements. Furthermore, noise reduction and image enhancement processing can be applied to the microscopic images of LiH2PO4 to further improve image quality.

[0019] Based on time, each production batch is divided into the current production batch and several historical production batches. This embodiment is used to determine the particle size of lithium dihydrogen phosphate in the current production batch.

[0020] For any production batch, taking the current production batch as an example, the lithium dihydrogen phosphate microscopic image of the current production batch is first segmented to obtain the various particle regions of the lithium dihydrogen phosphate crystal particles, thereby obtaining the particle parameters of the particle regions. In an exemplary embodiment, a trained CNN (Convolutional Neural Networks) model is used to perform pixel-level semantic segmentation processing on the initial lithium dihydrogen phosphate image, outputting a binary segmentation mask, thereby using the binary segmentation mask to obtain the various particle regions of the lithium dihydrogen phosphate crystal particles in the lithium dihydrogen phosphate microscopic image.

[0021] The CNN model is trained using a general training method. In an exemplary embodiment, the training method for the CNN model is as follows: A U-Net architecture is used to construct the CNN model. A training sample set is determined, consisting of multiple training samples, each a labeled lithium dihydrogen phosphate microscopic image. Pixel-level region annotations are performed on the particle regions in the lithium dihydrogen phosphate microscopic image to form corresponding labels. The training sample set is used as input to the CNN model, and the CNN model outputs a binary segmentation mask of the label-corresponding regions in the lithium dihydrogen phosphate microscopic image, resulting in the trained CNN model. The CNN model is a commonly used existing neural network model, and the process of training the CNN model based on labeled sample images is a conventional technique and will not be elaborated further.

[0022] In the particle size determination of lithium dihydrogen phosphate (LDP) particles, the optical properties of LDP particles in terms of refractive index and transparency, as well as image distortion caused by particle concentration, lead to deviations in parameters such as particle size and shape extracted from LDP microscopic images from their true values. Specifically, since LDP is a transparent crystal, it is prone to scattering in the visible light band. Therefore, when high-speed microscopic imaging modules capture LDP microscopic images, the scattering halos of LDP crystal particles are misjudged by the segmentation algorithm as being relatively larger. At the same time, for non-spherical particles, anisotropic scattering leads to blurred edges, affecting the accuracy of parameters such as the roundness and aspect ratio of the crystal particles. In addition, in the particle flow, high particle concentration areas exacerbate multiple scattering, causing overlap and shadows in LDP microscopic images, further distorting the particle size distribution. Therefore, this embodiment analyzes each particle region in the LDP microscopic image to obtain relevant parameters of the particle region, so as to facilitate subsequent particle size distribution analysis and perform light scattering distortion correction.

[0023] like Figure 1 As shown, the computer-aided method for determining the particle size of lithium dihydrogen phosphate provided in this embodiment includes the following steps: Step S1: Obtain the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch; Step S2: Determine the difference in equivalent circle diameter between the measured and actual conditions for the reference historical production batches associated with the current production batch, and obtain the correction degree for the current production batch by combining the reference weight of the reference historical production batches with respect to the current production batch. Step S3: Correct the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch according to the correction degree.

[0024] The following detailed explanation of each step, in conjunction with the accompanying drawings, is provided.

[0025] Step S1: Obtain the equivalent circle diameter and particle concentration of lithium dihydrogen phosphate in the current production batch.

[0026] Taking the current production batch as an example, obtain the equivalent circular diameter of lithium dihydrogen phosphate in the current production batch (using the equivalent circular diameter as the particle size), that is, obtain the equivalent circular diameter of each particle region of lithium dihydrogen phosphate in the current production batch. In addition to the equivalent circular diameter, also obtain the roundness of each particle region.

[0027] For any given particle region, the area of ​​that particle region is obtained. That is, the number of pixels contained in the particle region. This is the equivalent circle diameter of the particle region. Let π be the mathematical constant. Obtain the perimeter of the particle region. ,Will The roundness of the particle region is defined as a value ranging from 0 to 1. The closer the roundness is to 1, the closer the shape of the particle region is to a circle. In addition, this embodiment can also obtain the minimum bounding rectangle of the particle region and use the ratio of the long side to the short side of the minimum bounding rectangle as the aspect ratio of the particle region.

[0028] The diameter of the calibration circle is used to directly reflect the particle size of the particle region, the roundness is used to indicate the degree of influence of scattering on the non-spherical particle region, and the aspect ratio is used to indicate the degree of anisotropic scattering effect. By obtaining multi-dimensional particle parameters, data support is provided for subsequent particle size measurement result correction, avoiding the limitations of using a single parameter in subsequent measurement result correction.

[0029] The process of obtaining particle concentration is existing technology. In this embodiment, particle concentration can be understood as the number of lithium dihydrogen phosphate particle regions within a unit area. In an exemplary embodiment, the microscopic image of the lithium dihydrogen phosphate from the reference historical production batch is divided into several unit areas. The number of particle regions within each unit area is obtained, and then the average number of particle regions within each unit area is calculated as the particle concentration of the lithium dihydrogen phosphate from the reference historical production batch. The size of the unit area is set according to actual needs and is usually not set too large.

[0030] Step S2: Determine the difference in equivalent circle diameter between the measured and actual conditions for the reference historical production batches associated with the current production batch. Combine the particle concentration and average roundness of the reference historical production batches to obtain the reference weight for the current production batch. Based on the particle concentration of the reference historical production batches, the reference weight of the current production batch, and the similarity of the process characteristic parameters of the current production batch, determine the correction degree for the current production batch.

[0031] First, the historical production batches associated with the current production batch are determined from all historical production batches. These historical production batches are defined as reference historical production batches associated with the current production batch, i.e., the reference historical production batch of the current production batch. In an exemplary embodiment, this embodiment obtains the reference historical production batch based on the process characteristic parameters of the production batch, such as... Figure 2 As shown, the following is a specific process for obtaining reference to historical production batches: Step S21: Determine the process characteristic parameters of the current production batch and several historical production batches respectively.

[0032] For the current production batch and any production batch from previous historical production batches, the process characteristic parameters of that production batch are determined. Process characteristic parameters are features related to the production process. In an exemplary embodiment, these parameters include dispersion chamber airflow velocity, dispersion chamber temperature, dispersion chamber humidity, and particle concentration. A process characteristic parameter sequence is generated based on these parameters. This yields the process characteristic parameter sequence for the current production batch and the process characteristic parameter sequences for each previous historical production batch. It should be understood that this embodiment is not limited to the specific composition of the process characteristic parameters. The process characteristic parameters can be flexibly selected according to actual needs, such as including only dispersion chamber airflow velocity, dispersion chamber temperature, and dispersion chamber humidity, or adding other process parameters in addition to these parameters.

[0033] Step S22: Determine the similarity of process feature parameters between each historical production batch and the current production batch, and use historical production batches with process feature parameter similarity greater than a preset similarity threshold as reference historical production batches associated with the current production batch.

[0034] Obtain the similarity of process feature parameters between each historical production batch and the current production batch, that is, the similarity of the process feature parameter sequences of each historical production batch and the process feature parameter sequences of the current production batch. The specific implementation method of process feature parameter similarity is set according to actual needs. This embodiment takes cosine similarity as an example, that is, calculate the cosine similarity between the process feature parameter sequences of each historical production batch and the process feature parameter sequences of the current production batch. For ease of subsequent comparison, the cosine similarity is normalized, and the normalized cosine similarity is used as the process feature parameter similarity between each historical production batch and the current production batch. The larger the value of cosine similarity, the higher the similarity of the process feature parameters between the corresponding historical production batch and the current production batch, and the more related the corresponding historical production batch and the current production batch are. It should be understood that the normalization method is set according to actual needs. Since the value range of cosine similarity in principle is (-1, 1), the normalization method here can be: (cosine similarity + 1) / 2.

[0035] This embodiment presets a similarity threshold, which is used to compare the similarity of process feature parameters between each historical production batch and the current production batch to determine the larger similarity of process feature parameters. The numerical range of this preset similarity threshold is (0, 1), and the specific value is set according to actual needs. As an example, a value of 0.8 may be used.

[0036] The similarity of process feature parameters between each historical production batch and the current production batch is compared with a preset similarity threshold. Process feature parameter similarities exceeding the preset threshold are identified, and the historical production batches corresponding to these similarities are designated as reference historical production batches associated with the current production batch. It should be understood that there may be only one or multiple reference historical production batches associated with the current production batch.

[0037] For any reference historical production batch, the equivalent circle diameter of the reference historical production batch under measurement conditions is obtained. Specifically, the equivalent circle diameter of each particle region measured by image segmentation during the particle size measurement of lithium dihydrogen phosphate in the reference historical production batch is obtained. Then, the average value of the equivalent circle diameter of each particle region of lithium dihydrogen phosphate obtained under measurement conditions in the reference historical production batch is calculated as the first average equivalent circle diameter of lithium dihydrogen phosphate in the reference historical production batch.

[0038] The equivalent circular diameter of lithium dihydrogen phosphate (LDH) in a real-world context is obtained from the historical production batch. Specifically, the true equivalent circular diameter of each particle region in the historical production batch is obtained, and then the average of these true equivalent circular diameters is calculated as the second average equivalent circular diameter of the lithium dihydrogen phosphate in the historical production batch. It should be understood that this embodiment can employ high-precision techniques with extremely high accuracy and minimal or no influence from light scattering to obtain the true equivalent circular diameter of each particle region of lithium dihydrogen phosphate. In one exemplary embodiment, scanning electron microscopy can be used to perform electron beam imaging of lithium dihydrogen phosphate, achieving extremely high resolution (down to the nanometer level), virtually eliminating blurring caused by light scattering, and resulting in extremely sharp image edges. The obtained equivalent circular diameter of each particle region of lithium dihydrogen phosphate is its true equivalent circular diameter and serves as a standard. Alternatively, several samples of lithium dihydrogen phosphate from the historical production batch can be placed in a laboratory, and the equivalent circular diameter of each particle region in the samples can be detected using high-precision instruments in the laboratory to obtain the true equivalent circular diameter of the historical production batch.

[0039] Because lithium dihydrogen phosphate (LiHP) always exhibits a certain degree of scattering effect when imaged using a high-speed microscopic imaging module, the average equivalent circle diameter obtained under measurement conditions for this reference historical production batch is slightly larger than the true average equivalent circle diameter, and the degree of difference is related to the extent of the scattering effect. Accordingly, the first average equivalent circle diameter of this reference historical production batch is slightly larger than the second average equivalent circle diameter. Therefore, the difference in equivalent circle diameter between the measured and actual conditions for this reference historical production batch is obtained. This difference represents the increase in the first average equivalent circle diameter relative to the second average equivalent circle diameter. The greater the increase, the more the measured average equivalent circle diameter deviates from the true average equivalent circle diameter, indicating a greater influence of the scattering effect on the particle size deviation.

[0040] Then, the reference weight of the reference historical production batch relative to the current production batch is determined. The reference weight reflects the degree to which the measurement of the equivalent circle diameter is affected by scattering and overlap between lithium dihydrogen phosphate crystal particles in the reference historical production batch. A larger reference weight indicates a higher degree of influence, and consequently, a lower confidence level in the first average equivalent circle diameter measured from the reference historical production batch. In an exemplary embodiment, such as... Figure 3 As shown, the following is a specific process for obtaining the reference weights: Step S23: Obtain the particle concentration and average sphericity of lithium dihydrogen phosphate from a reference historical production batch.

[0041] Obtain the particle concentration of lithium dihydrogen phosphate from the reference historical production batch. During high-speed microscopic imaging, high particle concentration of lithium dihydrogen phosphate particles easily induces multiple scattering and particle overlap, resulting in blurred particle outlines and difficulty in boundary segmentation in the microscopic image. Consequently, the lower the reference weight, the more inversely correlated the reference weight is with the particle concentration.

[0042] The roundness of each particle region in the reference historical production batch of lithium dihydrogen phosphate is obtained, and then the average roundness of each particle region is calculated as the average roundness of the lithium dihydrogen phosphate in the reference historical production batch. Non-spherical lithium dihydrogen phosphate particles suffer from edge distortion due to anisotropic scattering, and roundness directly characterizes the regularity of shape. The smaller the roundness, the more severe the scattering distortion, the lower the measurement confidence, and correspondingly, the lower the reference weight. Therefore, the reference weight is positively correlated with the average roundness.

[0043] Step S24: Obtain the reference weight based on particle concentration and average roundness.

[0044] The reference weight of the reference historical production batch for the current production batch is obtained based on the particle concentration and average sphericity of the lithium dihydrogen phosphate in the reference historical production batch.

[0045] A negative correlation mapping is performed on the particle concentration, and the product of the negative correlation mapping result and the particle concentration of lithium dioxane in the reference historical production batch is used as the reference weight of the reference historical production batch corresponding to the current batch.

[0046] Based on the logical analysis above, the following is a specific method for calculating the reference weight: ; in, Indicates the number of the current production batch. Reference weights for each historical production batch; Indicates the first The particle concentration of lithium dihydrogen phosphate from a reference historical production batch. This represents an exponential function with the natural constant as its base. This represents the influence coefficient of particle concentration, which is set according to the actual situation, such as an empirical value of 5. Indicates the first The average roundness of lithium dihydrogen phosphate from a historical production batch.

[0047] The reference weights of each historical production batch in the current production batch are obtained through the above method.

[0048] Then, based on the differences in equivalent circle diameters and reference weights of each historical production batch, the correction level for the current production batch is obtained. In an exemplary embodiment, such as... Figure 4 As shown, the following is a specific process for obtaining the degree of correction: Step S25: Obtain the correction influence factor for each reference historical production batch based on the difference in equivalent circle diameter and reference weight of each reference historical production batch.

[0049] For any reference historical production batch, the greater the difference in the equivalent circle diameter of the reference historical production batch, the greater the correction range for the equivalent circle diameter of the current production batch of lithium dihydrogen phosphate. In other words, the greater the correction degree of the current production batch, the greater the correction influence factor of the reference historical production batch. The correction influence factor is positively correlated with the difference in the equivalent circle diameter.

[0050] The greater the reference weight of the historical production batch, the lower the degree of scattering halo produced by lithium dihydrogen phosphate particles in the microscopic image of the historical production batch, the lower the degree of particle edge blurring, the lower the availability of the correction influence factor of the historical production batch, the smaller the correction range for the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch, that is, the smaller the correction degree of the current production batch, the smaller the correction influence factor of the historical production batch. The correction influence factor is inversely correlated with the reference weight.

[0051] Based on the difference in equivalent circle diameter of the reference historical production batch and the reference weight of the reference historical production batch, the correction impact factor of the reference historical production batch is obtained. Based on the above logic, a specific calculation method for the correction impact factor is given below: ; in, Indicates the first Corrected impact factor for a reference historical production batch. Indicates the first The second average equivalent circle diameter of lithium dihydrogen phosphate from a reference historical production batch; Indicates the first The first average equivalent circle diameter of lithium dihydrogen phosphate from a reference historical production batch; Represents the absolute value symbol.

[0052] pass Characterizing the first The difference in equivalent circle diameter among several historical production batches is such that the first average equivalent circle diameter is greater than the second average equivalent circle diameter. The numerical range of is (0,1). The greater the increase in the diameter of the first average equivalent circle relative to the diameter of the second average equivalent circle, the better. The smaller the value, the better. The larger the value, the better.

[0053] The correction influence factor reflects the degree to which the particle size distortion of lithium dihydrogen phosphate in the reference historical production batch is affected by the scattering intensity. The larger the value of the correction influence factor, the greater the degree to which the particle size distortion of the corresponding reference historical production batch is affected by the scattering intensity. Since the reference historical production batch and the current production batch are quite similar in production process, the larger the correction influence factor of the reference historical production batch, the greater the degree of correction for the equivalent circle diameter of the current production batch.

[0054] Step S26: Based on the correction influence factors of each reference historical production batch and the similarity of process characteristic parameters with the current production batch, obtain the correction degree of the current production batch.

[0055] For any reference historical production batch, the higher the similarity of the process characteristic parameters between the reference historical production batch and the current production batch, the greater the reliability of the correction influence factor of the reference historical production batch. Based on the correction influence factors of each reference historical production batch and the similarity of the process characteristic parameters between each reference historical production batch and the current production batch, the correction degree of the current production batch is obtained. In an exemplary embodiment, the correction degree performance index of each reference historical production batch is obtained from the correction influence factors and process characteristic parameter similarity. The higher the similarity of the process characteristic parameters, the higher the correction degree performance index of each reference historical production batch; the higher the correction influence factor, the higher the correction degree performance index of each reference historical production batch. The correction degree performance index is positively correlated with both the correction influence factor and the process characteristic parameter similarity. Then, the correction degree performance index of each reference historical production batch is merged to obtain the correction degree of the current production batch. Based on the above logic, a specific calculation method for the correction degree of the current production batch is given below: ; in, Indicates the degree of calibration for the current production batch; Indicates the first Similarity of process characteristic parameters between a reference historical production batch and the current production batch. This indicates the number of historical production batches referenced for the current production batch.

[0056] The correction level of the current production batch describes the degree of scattering exhibited by the lithium dihydrogen phosphate in the closed dispersion chamber. The higher the degree of scattering, the larger the particle size of lithium dihydrogen phosphate in the microscopic image is compared with the actual value. That is, the greater the degree of distortion of the particle size measurement results caused by scattering of lithium dihydrogen phosphate crystal particles, the greater the need for light scattering distortion correction to correct the particle size measurement results.

[0057] Step S3: Correct the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch according to the correction degree.

[0058] After obtaining the calibration level of the current production batch, in order to solve the problem that the particle size measurement results of lithium dihydrogen phosphate deviate from the actual value due to light scattering distortion in the dispersion chamber, this step corrects the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch according to the calibration level, thereby improving the accuracy of the particle size measurement results of lithium dihydrogen phosphate.

[0059] In one exemplary embodiment, such as Figure 5 As shown, the following is a specific implementation process for step S3: Step S31: Based on the degree of correction, obtain the correction coefficient for the current production batch of lithium dihydrogen phosphate.

[0060] Due to the high transparency of lithium dihydrogen phosphate (LDP), it produces scattering halos during visible light imaging, causing the measured particle size to typically exceed the actual particle size. Therefore, to compensate for this error, the equivalent circular diameter of each particle region of LDP is corrected. Since a higher correction level for the current production batch results in a larger particle size of LDP in the microscopic image compared to the actual value, leading to greater distortion in the measurement results, the reduction in the equivalent circular diameter of each particle region of LDP is more significant. Therefore, a correction coefficient for the current production batch of LDP is obtained based on the current correction level; a higher correction level results in a smaller correction coefficient, and the correction coefficient is inversely correlated with the correction level. In an exemplary embodiment, a specific calculation method for the correction coefficient is given below: ; in, This represents the correction factor for the current production batch of lithium dihydrogen phosphate.

[0061] Because lithium dihydrogen phosphate is a transparent crystal, incident light is scattered on the particle surface during high-speed microscopy, forming a scattering halo. This causes the particle edges to become blurred and the outline to expand in the microscopic image, resulting in a measured particle size that is larger than the true value. Therefore, to reduce the impact of scattering, this embodiment... As a correction factor, where, Attenuation factor characterizing scattering effect The smaller the value, the less the scattering effect occurs, and the closer the measured equivalent circle diameter is to the true value; conversely, when... The larger the value, the greater the attenuation factor (greater than 1). The smaller the value, the better, to compensate for the expansion caused by scattering and avoid the problem of overestimation of particle size caused by scattering.

[0062] Step S32: Multiply the equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch by the correction coefficient to obtain the corrected equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch.

[0063] For any particle region of lithium dihydrogen phosphate in the current production batch, the equivalent circle diameter of the particle region is multiplied by the correction coefficient to obtain the corrected equivalent circle diameter of the particle region. This process is repeated for each particle region of lithium dihydrogen phosphate in the current production batch to obtain the corrected equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch.

[0064] To gain a more accurate and comprehensive understanding of the particle size distribution of lithium dihydrogen phosphate crystals in the current production batch, this embodiment, after obtaining the corrected equivalent circle diameters of each particle region in the current production batch, statistically analyzes the corrected equivalent circle diameters of each particle region to generate particle size statistics for the current production batch, thus reflecting the particle size distribution. Accordingly, as... Figure 6 As shown, this embodiment also includes the following steps: Step S4: Determine the number of particle regions within each preset particle size range based on the corrected equivalent circle diameter of each particle region of the current production batch of lithium dihydrogen phosphate.

[0065] The corrected equivalent circle diameter of each particle region of the lithium dihydrogen phosphate in the current production batch is used as the corrected particle size of each particle region. This embodiment sets several preset particle size intervals, which can encompass the corrected particle sizes of each particle region of the lithium dihydrogen phosphate in the current production batch, facilitating data statistics. In an exemplary embodiment, the maximum and minimum corrected particle sizes of each particle region of the lithium dihydrogen phosphate in the current production batch are obtained. These maximum and minimum corrected particle sizes constitute an overall particle size interval, which is then divided into several intervals as preset particle size intervals. The number of preset particle size intervals is inversely related to the width of the preset particle size interval; the wider the interval, the fewer the number of preset particle size intervals. It should be understood that the number of preset particle size intervals is set according to actual needs. The principle of this embodiment is to ensure that each preset particle size interval contains a particle region.

[0066] The corrected particle size of each particle region in the current production batch of lithium dihydrogen phosphate is compared with each preset particle size interval to determine the preset particle size interval in which the corrected particle size of each particle region falls. The number of particle regions in each preset particle size interval is counted. The total number of particle regions in the current production batch of lithium dihydrogen phosphate is obtained, and the ratio of the number of particle regions in each preset particle size interval to the total number of particle regions in the current production batch of lithium dihydrogen phosphate is calculated as the proportion of particle regions in each preset particle size interval. The proportion of particle regions in each preset particle size interval is treated as discrete distribution data.

[0067] Step S5: Perform curve fitting on the number of particle regions within each preset particle size range to obtain the particle size distribution curve of lithium dihydrogen phosphate in the current production batch.

[0068] A histogram is constructed based on the proportion of particle regions within each preset particle size range. A curve is then fitted to the proportion of particle regions within each preset particle size range in the histogram to obtain the particle size distribution curve for the current production batch of lithium dihydrogen phosphate. Specifically, Gaussian kernel smoothing is used to fit the discrete distribution data to a continuous curve.

[0069] The above steps of correcting the equivalent circle diameter correct the light scattering effect on the lithium dihydrogen phosphate particle size measurement results, i.e., light scattering correction, eliminate the influence of image distortion on the particle size distribution curve of the current production batch of lithium dihydrogen phosphate, and accurately reflect the actual size distribution of lithium dihydrogen phosphate crystal particles in the dispersion chamber. For example, without correction, the peak value of the particle size distribution curve shifts to a larger particle size due to the influence of scattered light halo, such as... Figure 7 As shown, after correction, the equivalent circle diameter will return to its true position, as... Figure 8 As shown. Figure 7 and Figure 8 In the graph, the horizontal axis represents the equivalent circle diameter (the unit can be micrometers), and the vertical axis represents the percentage of particle regions. The particle size distribution curve, generated by correcting for the particle size of lithium dihydrogen phosphate crystals determined using visual methods, accurately reflects the particle size distribution of lithium dihydrogen phosphate crystals in the current production batch, thus providing data guidance for adjusting the lithium dihydrogen phosphate production process based on the particle size curve.

[0070] Furthermore, this embodiment can also generate a particle size determination report for the current production batch based on the particle size distribution curve of the current production batch. For example, it determines the percentage of particles with a corrected particle size within the median range based on the particle size distribution curve of the current production batch. If the percentage of particles with a particle size within the median range is greater than or equal to a preset threshold, it indicates that the quality of the lithium dihydrogen phosphate in the current production batch is qualified; otherwise, the quality of the lithium dihydrogen phosphate in the current production batch is unqualified. Finally, the particle size distribution curve of the current production batch, the percentage of particles with a particle size within the median range, and the determination result of whether the quality of the lithium dihydrogen phosphate in the current production batch is qualified constitute the particle size determination report for the current production batch. In addition, with computer assistance, this invention can also support unattended operation, providing real-time decision-making basis for the quality control of lithium dihydrogen phosphate production.

[0071] This embodiment also provides a computer-aided lithium dihydrogen phosphate particle size determination device, including: a memory and a processor; the memory is connected to the processor, and the memory is used to store program instructions; the processor is used to implement the steps in the above-described computer-aided lithium dihydrogen phosphate particle size determination method embodiment when the program instructions are executed.

[0072] In one exemplary embodiment, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the embodiment of the computer-aided lithium dihydrogen phosphate particle size determination method.

[0073] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A computer-aided method for determining the particle size of lithium dihydrogen phosphate, characterized in that, include: Obtain the equivalent circle diameter and particle concentration of lithium dihydrogen phosphate in the current production batch; Determine the difference in equivalent circle diameter between measured and actual conditions for a reference historical production batch associated with the current production batch, and combine the particle concentration and average roundness of the reference historical production batch to obtain the reference weight for the current production batch. The correction degree of the current production batch is determined based on the difference in equivalent circle diameter of the reference historical production batches, the reference weight of the current production batch, and the similarity of the process characteristic parameters of the current production batch; the similarity of the process characteristic parameters of the current production batch is the similarity of the process characteristic parameters of the historical production batches and the current production batch in quality inspection. The equivalent circle diameter of lithium dihydrogen phosphate in the current production batch is corrected according to the correction degree. The process of obtaining the reference weight includes: performing a negative correlation mapping on the particle concentration, and using the product of the negative correlation mapping result and the average roundness of lithium dioxane in the reference historical production batch as the reference weight of the reference historical production batch corresponding to the current batch. The process of obtaining the correction degree includes: obtaining the correction influence factor of each reference historical production batch from the difference in equivalent circle diameter and the reference weight; the correction influence factor is positively correlated with the difference in equivalent circle diameter and negatively correlated with the reference weight; and obtaining the correction degree of the current production batch based on the correction influence factor of each reference historical production batch and the similarity of the process characteristic parameters with the current production batch. Specifically, the correction degree of the current production batch is obtained based on the correction influence factors of each reference historical production batch and the similarity of process characteristic parameters with the current production batch. This includes: obtaining a correction degree performance index for each reference historical production batch from the correction influence factors and the similarity of process characteristic parameters; the correction degree performance index is positively correlated with both the correction influence factors and the similarity of process characteristic parameters; and the correction degree performance index of each reference historical production batch is integrated to obtain the correction degree of the current production batch.

2. The computer-aided method for determining the particle size of lithium dihydrogen phosphate as described in claim 1, characterized in that, Before obtaining the equivalent circular diameter of the lithium dihydrogen phosphate in the current production batch, the lithium dihydrogen phosphate particle size determination method further includes: Obtain microscopic images of lithium dihydrogen phosphate from the current production batch; The lithium dihydrogen phosphate microscopic image was segmented using a trained convolutional neural network model to obtain the particle regions of lithium dihydrogen phosphate. The step of obtaining the equivalent circular diameter of lithium dihydrogen phosphate in the current production batch includes: obtaining the equivalent circular diameter of the particle region of lithium dihydrogen phosphate in the current production batch.

3. The computer-aided method for determining the particle size of lithium dihydrogen phosphate as described in claim 1, characterized in that, The process of obtaining the reference historical production batches includes: Determine the process characteristic parameters for the current production batch and several historical production batches respectively; Determine the similarity of process feature parameters between each historical production batch and the current production batch, and use historical production batches whose process feature parameter similarity is greater than a preset similarity threshold as reference historical production batches associated with the current production batch.

4. The computer-aided method for determining the particle size of lithium dihydrogen phosphate as described in claim 1, characterized in that, The difference in equivalent circle diameter refers to the increase in the first average equivalent circle diameter relative to the second average equivalent circle diameter, based on historical production batches; the first average equivalent circle diameter is the average equivalent circle diameter of lithium dihydrogen phosphate under measurement conditions, and the second average equivalent circle diameter is the average equivalent circle diameter of lithium dihydrogen phosphate under actual conditions.

5. The computer-aided method for determining the particle size of lithium dihydrogen phosphate as described in claim 2, characterized in that, The step of correcting the equivalent circle diameter of lithium dihydrogen phosphate in the current production batch according to the correction degree includes: Based on the stated correction level, a correction coefficient for the current production batch of lithium dihydrogen phosphate is obtained; the correction coefficient is inversely correlated with the stated correction level. Multiplying the equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch by the correction coefficient yields the corrected equivalent circle diameter of each particle region of lithium dihydrogen phosphate in the current production batch.

6. The computer-aided method for determining the particle size of lithium dihydrogen phosphate as described in claim 5, characterized in that, After correcting the equivalent circular diameter of the current production batch of lithium dihydrogen phosphate according to the correction degree, the lithium dihydrogen phosphate particle size determination method further includes: Based on the corrected equivalent circle diameter of each particle region of the current production batch of lithium dihydrogen phosphate, determine the number of particle regions within each preset particle size range. By performing curve fitting on the number of particle regions within each preset particle size range, the particle size distribution curve of lithium dihydrogen phosphate in the current production batch is obtained.

7. A computer-aided lithium dihydrogen phosphate particle size determination device, characterized in that it comprises: Memory and processor; The memory is connected to the processor; The memory is used to store program instructions; The processor is configured to implement the computer-aided lithium dihydrogen phosphate particle size determination method according to any one of claims 1-6 when the program instructions are executed.

Images