Lithium hydroxide particle stability detection method, device, medium and equipment

Abstract

The invention belongs to the technical field of cell material detection, and discloses a lithium hydroxide particle stability detection method, device, medium and system.The method comprises the steps that a lithium hydroxide particle sample to be detected is obtained, and the lithium hydroxide particle sample to be detected is pretreated; detecting the pretreated lithium hydroxide particle sample to be detected to obtain detection data of the lithium hydroxide particle sample to be detected; a lithium hydroxide metal particle stability detection model is constructed and trained; and inputting the detection data of the lithium hydroxide particle sample to be detected into the detection model to obtain the stability index of the lithium hydroxide particle sample. The lithium hydroxide particle stability detection precision can be improved.

Description

Technical Field

[0001] The present application belongs to the technical field of battery material detection, and specifically relates to a method, device, medium and equipment for detecting the stability of lithium hydroxide particles. Background Art

[0002] As an important chemical raw material and battery material, lithium hydroxide (LiOH) exhibits broad application prospects in fields such as new energy and aerospace. However, LiOH metal particles are susceptible to humidity, leading to aggregation, decomposition, or deterioration, seriously affecting their performance stability and efficiency. Therefore, it is necessary to develop an efficient and accurate detection method to monitor and improve the stability of LiOH particles. Summary of the Invention

[0003] The main purpose of this application is to provide a method, device, medium and equipment for detecting the stability of lithium hydroxide particles. This application can improve the accuracy of lithium hydroxide particle stability detection.

[0004] To achieve the above objectives, this application provides the following technical solutions:

[0005] A lithium hydroxide particle stability detection method comprises: obtaining a lithium hydroxide particle sample to be tested; pre-treating the lithium hydroxide particle sample to be tested; testing the pre-treated lithium hydroxide particle sample to be tested to obtain detection data of the lithium hydroxide particle sample to be tested; constructing and training a lithium hydroxide metal particle stability detection model; and inputting the detection data of the lithium hydroxide particle sample to be tested into the lithium hydroxide metal particle stability detection model to obtain a stability index of the lithium hydroxide metal particle sample.

[0006] Optionally, pre-treating the lithium hydroxide particle sample to be tested includes: cleaning the lithium hydroxide particle sample to be tested; and drying the cleaned lithium hydroxide particle sample to be tested.

[0007] Optionally, the testing of the pretreated lithium hydroxide particle sample to be tested includes: physical property testing, chemical property testing and electrochemical performance testing, wherein the physical property testing includes specific surface area testing, particle size testing and color testing; the chemical property testing includes thermal stability testing, moisture content testing and impurity content testing; the electrochemical performance testing includes cyclic voltammetry testing and constant current charge and discharge testing.

[0008] Optionally, the lithium hydroxide metal particle stability detection model includes: a feature extraction network for extracting features from the detection data of the lithium hydroxide particle sample to be tested; a feature fusion network for fusing the features extracted by the feature extraction network; and an output network for outputting the stability index of the lithium hydroxide particles to be tested.

[0009] Optionally, the feature extraction network introduces superimposed convolutional layers to perform layer-by-layer feature extraction on the detection data of the lithium hydroxide particle sample to be tested.

[0010] Optionally, the lithium hydroxide metal particle stability detection model is trained by the following steps: collecting a large number of lithium hydroxide particles and pre-processing them; testing the pre-treated lithium hydroxide particles, including performing an initial test to obtain initial test data, and performing a secondary test after simulation under different humidity conditions to obtain secondary test data; integrating the initial test data and the secondary test data to form a test data set, and dividing the test data set into a training set and a validation set in proportion; setting training parameters, and training the model based on the training set until the maximum number of training times is reached;

[0011] The model is verified based on the validation set, and accuracy, recall rate and F1 score are used as evaluation indicators. When the accuracy, recall rate and F1 score reach the preset values, the model verification is passed; otherwise, the training parameters are adjusted or the training set samples are expanded to retrain the model until the model verification is passed.

[0012] In addition, the present application also provides a lithium hydroxide particle stability detection device, which includes: an acquisition module for acquiring a lithium hydroxide particle sample to be tested; a pretreatment module for preprocessing the lithium hydroxide particle sample to be tested; a detection module for detecting the pretreated lithium hydroxide particle sample to be tested to obtain detection data of the lithium hydroxide particle sample to be tested; a model construction and training module for constructing and training a lithium hydroxide metal particle stability detection model; and a stability detection module for inputting the detection data of the lithium hydroxide particle sample to be tested into the detection model to obtain a stability index of the lithium hydroxide metal particle sample.

[0013] Optionally, the pretreatment module includes: a cleaning submodule for cleaning the lithium hydroxide particle sample to be tested; and a drying submodule for drying the cleaned lithium hydroxide particle sample to be tested.

[0014] In addition, the present application also provides a storage medium comprising instructions, which, when executed on a computer, enables the computer to execute the method as described in any of the preceding items.

[0015] In addition, the present application also provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the processor implements the method described in any of the preceding items when executing the program.

[0016] The beneficial effects achieved by the present application are as follows: by constructing a lithium hydroxide metal particle stability detection model and deeply processing the detection data, the present application can extract features that are critical to stability assessment, and organically combine features of different dimensions to achieve a comprehensive and integrated evaluation of lithium hydroxide particle detection data, thereby being able to accurately judge the stability of lithium hydroxide particles. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic flow chart of a method for detecting the stability of lithium hydroxide particles provided in one embodiment of the present application;

[0018] Figure 2 This is a schematic structural diagram of a lithium hydroxide particle stability detection model provided in another embodiment of the present application;

[0019] Figure 3 yes Figure 2 Schematic diagram of the structure of the global feature extraction module in the shown model;

[0020] Figure 4 This is a schematic structural diagram of a lithium hydroxide particle stability detection device provided by another embodiment of the present application;

[0021] Figure 5 is a schematic structural diagram of a storage medium proposed in another embodiment of the present application;

[0022] Figure 6 It is a structural diagram of an electronic device proposed in another embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] It should be noted that all directional indications in the embodiments of the present application (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0025] In this application, unless otherwise specified or limited, the terms "connection" and "fixation" should be understood in a broad sense. For example, "fixation" can mean fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two elements or interaction between two elements, unless otherwise specified. For those skilled in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0026] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme in which A and B are satisfied at the same time. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement it. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0027] Figure 1 FIG. 1 is a flow chart of a method for detecting the stability of lithium hydroxide particles proposed in an exemplary embodiment of the present application. Figure 1 As shown, the method includes:

[0028] S100: obtaining a lithium hydroxide particle sample to be tested;

[0029] S200: pre-treating the lithium hydroxide particle sample to be tested;

[0030] S300: Detecting the pretreated lithium hydroxide particle sample to be tested to obtain detection data of the lithium hydroxide particle sample to be tested;

[0031] S400: Build and train a lithium hydroxide metal particle stability detection model;

[0032] S500: Inputting the detection data of the lithium hydroxide particle sample to be tested into the lithium hydroxide metal particle stability detection model to obtain a stability index of the lithium hydroxide metal particle sample.

[0033] This embodiment constructs a lithium hydroxide metal particle stability detection model and performs in-depth processing on the detection data. It can extract features that are critical to stability assessment and organically combine features of different dimensions, thereby achieving a comprehensive and integrated evaluation of lithium hydroxide particle detection data, and thus accurately judging the stability of lithium hydroxide particles.

[0034] In another exemplary embodiment, in step S200, the lithium hydroxide particle sample to be tested is pretreated, including:

[0035] S201: Cleaning the lithium hydroxide particle sample to be tested;

[0036] In this step, the lithium hydroxide particle sample to be tested is first placed in an appropriate solvent (such as anhydrous ethanol). Next, the container containing the particle sample and solvent is placed in an ultrasonic cleaning machine. The ultrasonic power and treatment time are set appropriately, and the machine is started. Under the action of ultrasound, the solvent deeply cleans and disperses the particles. After cleaning, the particle sample and solvent are separated by filtration or centrifugation.

[0037] S202: Drying the cleaned lithium hydroxide particle sample to be tested.

[0038] In this step, after the particle sample is cleaned, it needs to be placed in a drying oven and dried at an appropriate temperature. It should be noted that the drying temperature should not be too high to avoid damaging the particle structure or inducing chemical reactions. Generally, the drying temperature is set at around 100°C and the drying time is about 1-2 hours.

[0039] In addition, vacuum drying technology can also be used to dry the samples. Compared with using a drying oven, vacuum drying can quickly remove moisture and solvent residues in the particles at a lower temperature, avoiding the impact of high temperature on the particle structure and performance. At the same time, vacuum drying can also effectively prevent the particle samples from oxidation or secondary contamination during the drying process.

[0040] S203: performing surface treatment on the dried lithium hydroxide particle sample to be tested;

[0041] In this step, the dried lithium hydroxide particle sample to be tested can be surface treated using low-temperature plasma. Low-temperature plasma can remove organic pollutants, oxides, and other impurities from the particle surface by bombarding the particle surface with high-energy particles without changing the properties of the particle itself. It may also change the particle surface microstructure and increase its surface activity.

[0042] S204: Screening the surface-treated particle sample using a magnetic field.

[0043] In this step, since the lithium hydroxide particle sample may contain magnetic impurities, it is necessary to separate the magnetic impurities from the particles through the action of a magnetic field to improve the purity of the particle sample and thereby improve the accuracy of the test results.

[0044] In another exemplary embodiment, the pretreated lithium hydroxide particle sample to be tested is tested, including: physical property testing, chemical property testing and electrochemical performance testing, wherein:

[0045] Physical property testing includes:

[0046] Specific surface area detection: Use a specific surface area analyzer (such as a BET specific surface area analyzer) to detect the initial specific surface area of ​​the sample particles.

[0047] Particle size detection: Use a laser particle size analyzer to analyze the particle size distribution of sample particles to understand the uniformity and size range of sample particles.

[0048] Color detection: Use a colorimeter to detect the color of the sample particles. The color change can indicate oxidation, contamination or other chemical changes of the sample particles. For example, if the color of the sample particles changes from white to yellow or brown, it may indicate that the particles have undergone oxidation reaction.

[0049] Chemical property testing includes:

[0050] Thermal stability testing: Use a thermogravimetric analyzer or differential scanning calorimeter to test the stability of sample particles under high temperature conditions, observe the mass change or thermal effect change of the sample particles, and evaluate whether the sample particles decompose or release gas.

[0051] Moisture content detection: Use a Karl Fischer moisture analyzer or an infrared moisture analyzer to detect the moisture content in the sample particles.

[0052] Impurity content detection: Use analytical techniques such as inductively coupled plasma mass spectrometry (ICP-MS), atomic absorption spectroscopy (AAS) or X-ray fluorescence spectroscopy (XRF) to detect the impurity components and content in the sample particles, so as to analyze the impurities such as metal elements, non-metallic elements and compounds in the sample particles, and then evaluate the purity, electrochemical properties and stability of the sample particles.

[0053] Electrochemical performance testing includes:

[0054] Cyclic voltammetry: Cyclic voltammetry is performed on sample particles using an electrochemical workstation. Specifically, the sample particles are formed into electrodes and placed in a solution containing a supporting electrolyte. A cyclic potential sweep is applied over a certain potential range while the current-voltage curve is recorded. Parameters such as the curve shape, peak potential, and peak current are analyzed to assess the electrochemical activity and redox reversibility of the sample particles.

[0055] Constant current charge and discharge testing: The sample particles were also subjected to constant current charge and discharge testing using an electrochemical workstation. Specifically, after the sample particles were formed into electrodes, charge and discharge cycles were performed under constant current conditions, while parameters such as charge and discharge capacity, voltage change, and cycle stability were recorded. By comparing indicators such as capacity retention and voltage drop at different cycle times, the electrochemical performance and cycle stability of the sample particles were evaluated.

[0056] In another exemplary embodiment, the lithium hydroxide particle stability detection model includes: a feature extraction network, a feature fusion network and an output layer, wherein:

[0057] like Figure 2 As shown, the feature extraction network includes an input layer, an embedding layer, a first convolutional layer, a first mixed pooling layer, a second convolutional layer, a second mixed pooling layer, a third convolutional layer, a third mixed pooling layer, a flattening layer, and a global feature extraction module, which are connected in sequence. The first convolutional layer and the second convolutional layer, as well as the first convolutional layer and the third convolutional layer, use residual connections, i.e., the input of the first convolutional layer is simultaneously fused with the output of the second convolutional layer and the third convolutional layer. In addition, a nonlinear activation function, such as a ReLU function, is provided after the first to third convolutional layers.

[0058] In the feature extraction network, the input layer receives multiple-dimensional data on lithium hydroxide particles, including specific surface area, particle size, color, thermal stability, moisture content, impurity content, and cyclic voltammetry and constant current charge-discharge tests. The embedding layer maps high-dimensional input data into low-dimensional, dense vectors, thereby reducing the model's computational workload and extracting more effective features. The first, second, and third convolutional layers perform layer-by-layer feature extraction on the data. The first convolutional layer extracts basic features, such as surface characteristics of lithium hydroxide sample particles to detect whether the particle surface has significant unevenness or pores. The second convolutional layer further extracts higher-level features based on the basic features extracted by the first convolutional layer, such as detecting changes in particle mass or thermal effects during heating to assess particle stability. The third convolutional layer extracts more complex features based on the first and second convolutional layers, such as detecting the electrochemical activity of the particles and evaluating their electrochemical performance through parameters such as the shape of the cyclic voltammetry curve, peak potential, and peak current. Another example is to evaluate the charge and discharge performance of the particles under constant current conditions, including capacity retention, voltage drop, etc., to determine the cycle stability of the particles. Through the layer-by-layer feature extraction of the three convolutional layers, the model can gradually establish an in-depth understanding of the stability of lithium hydroxide particles based on the basic features. This multi-level feature extraction helps the model to more comprehensively evaluate the stability changes of particles under different humidity conditions, thereby improving the accuracy and reliability of model detection. In addition, the input of the first convolutional layer is directly fused with the output of the second and third convolutional layers, which enables the model to learn higher-level features while retaining the original input information, thereby helping the model to more accurately extract and understand the stability characteristics of lithium hydroxide particles. Furthermore, the first, second, and third convolutional layers all use residual connections, which increases the flexibility of the model and allows the model to bypass certain layers when necessary, thereby helping the model to better adapt to different types of input data. In this embodiment, this flexibility helps the model to more accurately evaluate the stability changes of lithium hydroxide particles under different humidity conditions.

[0059] The present embodiment is provided with a hybrid pooling layer after the first convolutional layer, the second convolutional layer and the third convolutional layer, and the hybrid pooling layer helps the model to gradually extract and compress the multidimensional characteristics of lithium hydroxide particles, while reducing computational complexity, thereby improving the generalization ability and robustness of the model. It should be noted that the reason why the present application adopts a hybrid pooling layer instead of a traditional single pooling layer (such as maximum pooling or average pooling) is that: the first hybrid pooling layer combines the advantages of maximum pooling and average pooling, can retain significant features and reduce noise, so that the model can better handle the changes in input data, improve the robustness to changes in input data; the second is that the hybrid pooling layer combines the advantages of the two pooling methods, helps the model to reduce overfitting; the third is that through the hybrid pooling layer, the model can better handle the changes in the data, such as the characteristic changes of lithium hydroxide particles under different humidity conditions, helps to improve the performance of the model on unknown data, and enhances the generalization ability of the model.

[0060] The flattening layer is located after the third hybrid pooling layer, and is used to convert the features extracted by multiple convolutional layers and hybrid pooling layers into one-dimensional vectors, thereby providing necessary data preparation for subsequent feature fusion and model output.

[0061] In addition, the innovation of this embodiment is to introduce a global feature extraction module, such as Figure 3As shown, the global feature extraction module uses a multi-head attention mechanism to extract key features from the input data. The module includes a 1×1 convolutional layer (Conv) that maps the input data to a new representation space to better capture interdependencies between features. A batch normalization layer (BN) is connected after the 1×1 convolutional layer (Conv) to normalize the output of the 1×1 convolutional layer (Conv), thereby accelerating model training. The output of the 1×1 convolutional layer (Conv) is divided into a query vector, a key vector, and a numeric vector for calculating attention weights. To enhance the model's ability to model local features, the module adds local attention heads before and after the softmax function. (Local attention heads introduce a type of local attention to enhance the model's ability to model local features. A linear mapping between multiple attention heads is introduced before and after the softmax operation to increase information exchange between multiple attention mechanisms. The local attention head consists of a convolutional layer, a Transformer encoder layer, a GRU gating unit, and a multi-head attention mechanism.) This linear mapping increases information exchange between the multiple attention heads, thereby better capturing local features and their interactions. After passing through the global and local attention mechanisms, the numerical vector is multiplied by the obtained attention weights to obtain a weighted feature representation. This weighted feature representation is processed through a 1×1 convolutional layer (Conv) and a batch normalization layer (BN) to obtain the final feature representation. This feature representation passes through the Transformer encoder layer and is superimposed with the input of the global feature extraction module. It then passes through a fully connected module consisting of a normalization layer, two fully connected layers, and an activation function (GeLU) to obtain the final global feature map. It should be noted that the final global feature map needs to be fused with the result of superimposing the output of the Transformer encoder layer and the input of the global feature extraction module. This fusion ensures that the model considers high-level abstract features when evaluating stability while retaining the detailed information and temporal characteristics of the original data, thereby enhancing information complementarity. It also enables the model to more comprehensively evaluate the stability of the particles. Even if information in certain feature dimensions is incomplete or noisy, it can be compensated by information in other feature dimensions, thereby improving the robustness of the model. By fusing the global feature map with the raw input and encoder layer output, the model can integrate multiple pieces of information for stability assessment, taking into account not only the physical, chemical, and electrochemical properties of the particles, but also how these properties change under varying humidity conditions. This comprehensive assessment helps the model more accurately identify key factors influencing stability, such as changes in the particle surface microstructure, thermal stability, and electrochemical performance, thereby providing stronger support for quality control of battery materials.

[0062] In summary, in the global feature extraction module, the multi-head attention mechanism processes multiple "heads" in parallel, with each head independently focusing on different parts of the input data or different feature dimensions. This means that the model can simultaneously capture characteristic information about lithium hydroxide particles from multiple perspectives, such as specific surface area, particle size, color, and thermal stability, thereby more comprehensively assessing particle stability. Furthermore, through the parallel processing of multiple heads, the model is able to learn the complex interactions and dependencies between these features. For example, the correlation between specific surface area and particle size may vary under different humidity conditions, while color changes may be closely related to the degree of oxidation of the particles. These complex interactions are effectively modeled by the multi-head attention mechanism, thereby improving the accuracy and robustness of the model's predictions. Furthermore, the stability of lithium hydroxide particles can vary significantly under different humidity conditions. By processing multiple feature dimensions in parallel, the multi-head attention mechanism can better capture these changes and learn the changing patterns of particle stability under different humidity conditions, thereby helping the model to demonstrate better generalization when dealing with unknown humidity conditions or new samples.

[0063] At the same time, the introduction of local attention heads further enhances the model's ability to model local features. By linearly mapping between multiple attention heads, local attention heads can capture and emphasize the importance of local features, enabling the model to more accurately identify key factors affecting stability when evaluating lithium hydroxide particle stability.

[0064] Finally, these feature representations processed by the global feature extraction module are fed into the feature fusion network and output layer for further feature fusion and stability assessment. By combining global and local features, the entire model achieves a highly accurate assessment of the stability of lithium hydroxide particles.

[0065] The feature fusion network includes a Bi-LSTM network, a multi-head attention mechanism, a first fully connected layer, an activation function ReLU, a second fully connected layer, an activation function ReLU and a dropout layer connected in sequence.

[0066] In the feature fusion network, considering that the data in the stability detection of lithium hydroxide particles has time series characteristics (for example, changes under different humidity conditions), the bidirectional long short-term memory network (Bi-LSTM) is used to capture information about these changes from both forward and backward directions, thereby helping the model to better understand the time series characteristics.

[0067] The multi-head attention mechanism allows the model to focus on multiple different information sources at the same time, such as specific surface area detection, particle size detection, color detection, etc. Through the multi-head attention mechanism, the model can better capture the relationship between these features and learn more complex patterns from them. Among them, the relationship between the features includes, for example, the relationship between color detection and stability (color changes can reflect chemical changes in particles, such as the degree of oxidation reaction. The model can learn the relationship between color change and stability. For example, a color change from white to light yellow may indicate that the particles have undergone oxidation reactions, thereby affecting their stability. By learning the relationship between color change and stability, the model can more accurately evaluate the stability of particles under different humidity conditions). Another example includes the relationship between thermal stability detection and impurity content detection (the presence of impurities may affect the thermal stability of particles. For example, the presence of certain impurities may cause particles to begin to decompose at lower temperatures. The model can learn this relationship and how this relationship changes with changes in humidity. By learning the relationship between thermal stability and impurity content, the model can more accurately evaluate the thermal stability of particles under different humidity conditions). Such patterns include, for example, multi-factor interactions (the model can learn the interactions between different test results, such as how the interaction between specific surface area and particle size tests affects particle stability) and particle aggregation tendencies (the model can learn the aggregation tendencies of particles under different humidity conditions, that is, whether the particles tend to aggregate together). Through the multi-head attention mechanism, the model can capture the relationships between the above features and learn more complex patterns, thereby more accurately evaluating the stability of lithium hydroxide particles under different humidity conditions.

[0068] The combination of the fully connected layer and the activation function enables the model to further extract and combine features and learn higher-level abstract features, such as the correlation between specific surface area and particle size, that is, how the specific surface area changes with changes in particle size under different humidity conditions (for example, in an environment with higher humidity, particles with larger particle sizes may show greater changes in specific surface area, which indicates that more chemical reactions have occurred on the particle surface). Another example includes the relationship between moisture content and electrochemical performance, that is, how moisture content affects the results of cyclic voltammetry tests and constant current charge and discharge tests (for example, as humidity increases, increased moisture content may lead to a decrease in electrochemical performance, which is manifested as a decrease in peak current in cyclic voltammetry tests or a decrease in capacity in constant current charge and discharge tests). Through the combination of fully connected layers and activation functions, the model is able to learn higher-level abstract features that can better describe the changes in the characteristics of lithium hydroxide particles under different humidity conditions, thereby providing strong support for stability evaluation.

[0069] In addition, the Dropout layer in the feature fusion network is used to randomly discard a portion of neurons to reduce the risk of model overfitting.

[0070] The output layer uses a sigmoid activation function to output a stability index.

[0071] In another exemplary embodiment, in step S400, the lithium hydroxide metal particle stability detection model is trained by the following steps:

[0072] S401: Collect a large amount of lithium hydroxide particles and perform pretreatment, the pretreatment process is as described above;

[0073] S402: Testing the pretreated lithium hydroxide particles, specifically including performing a primary test to obtain primary test data and performing a secondary test to obtain secondary test data after simulating the pretreated lithium hydroxide particles under different humidity conditions;

[0074] In this step, the pretreated lithium hydroxide particles are simulated under different humidity conditions and then subjected to secondary testing to obtain secondary test data, which specifically includes the following steps:

[0075] First, the samples need to be packaged. You can choose clean, non-contaminated containers or sealed bags, and ensure that there will be no chemical reaction or contamination of the sample particles. In addition, the container should be clearly marked to distinguish the sample particles under different humidity conditions.

[0076] Next, use a precise weighing tool (such as an electronic scale) to weigh an appropriate amount of sample particles from the pretreated lithium hydroxide metal particle sample to ensure that the sample quantity in each sub-container is consistent to reduce experimental error. Pour the weighed sample particles into the prepared sub-container and immediately seal it to prevent the sample from being disturbed by the external environment.

[0077] Secondly, set up the humidity simulation environment. The specific implementation steps are as follows:

[0078] Prepare appropriate humidity simulation equipment, such as a constant temperature and humidity chamber or a humidity control room, according to experimental requirements.

[0079] According to the experimental design, different humidity conditions (from 0 to 100%) are set. However, in order to simplify the explanation, this application only sets three humidity levels: low humidity, medium humidity and high humidity, where low humidity is, for example, 20% RH (relative humidity), medium humidity is, for example, 40% RH to 60% RH, and high humidity is, for example, 80% RH, to comprehensively evaluate the effect of humidity on the stability of lithium hydroxide metal particles.

[0080] Place the packaged lithium hydroxide particle samples into a simulated environment with set humidity conditions, and ensure that the samples are evenly distributed in the simulated environment to obtain accurate experimental results.

[0081] During the humidity simulation process, it is necessary to regularly observe changes in the appearance and color of the sample particles and record the relevant data.

[0082] After the humidity simulation is completed, the samples are taken out for secondary testing to obtain the samples under different humidity conditions. The specific testing content is as described above.

[0083] Integrate the initial test data and the secondary test data of the sample to form a test data set of the sample, and divide the test data set into a training set and a validation set according to a certain ratio. For example, the division ratio can be 8:2;

[0084] S403: Setting training parameters, for example, setting the learning rate to 0.001, the batch size to 32, and the number of training times to the number of samples in the training set, and training the model based on the training set. During each training process, the model calculates the stability index of each sample based on the initial test data of each sample in the training set and the secondary test data simulated under different humidity conditions to evaluate the stability of each sample, until the stability index of each sample in the training set is obtained;

[0085] Specifically, assume that the initial test data of a sample is as shown in Table 1:

[0086] Table 1

[0087] index Initial test data Weight Specific surface area detection (SA) <![CDATA[10m 2 / g]]> 0.2 Particle Size Detection (PSD) D50=5um 0.15 Color Detection White 0.05 Thermal Stability Test (TST) No decomposition at 200℃ 0.2 Moisture content detection (WC) 0.05％ 0.15 Impurity Content Detection (IC) 0.01％ 0.1 Cyclic voltammetry (CV) Peak current 10mA 0.05 Constant current charge and discharge test (CD) Capacity 1000mAh 0.1

[0088] It should be noted that the weights of the indicators in Table 1 are based on expert experience and judgment.

[0089] Furthermore, the secondary test data of the sample (taking 80% RH humidity simulation as an example) is shown in Table 2:

[0090] Table 2

[0091] index Initial test data Weight Specific surface area detection (SA) <![CDATA[9m 2 / g]]> 0.2 Particle Size Detection (PSD) D50=6um 0.15 Color Detection light yellow 0.05 Thermal Stability Test (TST) No decomposition at 200℃ 0.2 Moisture content detection (WC) 0.1％ 0.15 Impurity Content Detection (IC) 0.015％ 0.1 Cyclic voltammetry (CV) Peak current 9mA 0.05 Constant current charge and discharge test (CD) Capacity 950mAh 0.1

[0092] After obtaining the test data shown in Table 1 and Table 2, the degree of deterioration of each indicator of the sample can be calculated. The specific calculation is as follows:

[0093] 1. Calculate the degree of deterioration of each indicator:

[0094] SA deterioration = 10-9 = 1m 2 / g;

[0095] PSD deterioration = 6-5 = 1 μm;

[0096] Color deterioration (assuming conversion to quantitative scores, white = 10, pale yellow = 8) = 10-8 = 2;

[0097] TST (no change) = 0;

[0098] WC deterioration = 0.1-0.05 = 0.05%;

[0099] IC deterioration = 0.015-0.01 = 0.005%;

[0100] CV deterioration = 10-9 = 1 mA;

[0101] CD deterioration = 1000-950 = 50 mAh.

[0102] 2. Take the weighted sum of the deterioration degree of each indicator to obtain the total deterioration:

[0103] Total deterioration = (1*0.2)+(1*0.15)+(2*0.05)+(0*0.2)+(0.05*0.15)+(0.005*0.1)+(1*0.05)+(50*0.1)=5.508

[0104] 3. Calculate the stability index based on the total deterioration:

[0105] Stability index = 100 - total deterioration * 10 (multiplying by 10 here is to amplify the degree of deterioration so that stability can be expressed in the range of 0-100)

[0106] That is, the stability index of the above sample = 100-5.508*10 = 44.92.

[0107] 4. Refer to expert opinions and industry standards to set stability level evaluation thresholds:

[0108] Excellent: Stability index ≥90

[0109] Good: 80≤Stability Index<90

[0110] Generally: 70≤Stability Index<80

[0111] Poor: 60≤stability index<70

[0112] Poor: Stability index <60

[0113] The stability index of the above sample is 44.92, which is less than 60, and thus the stability of the above sample is poor.

[0114] S403: Use the validation set to verify the trained model, and use accuracy, recall and F1 score as evaluation indicators. When the accuracy reaches 90%, the recall reaches 80%, and the F1 score reaches 85%, the model verification is passed; otherwise, adjust the training parameters (for example, adjust the learning rate to 0.05 or the batch size to 64) or expand the training set samples to retrain the model until the model verification is passed.

[0115] In another embodiment, the present application also provides a lithium hydroxide particle stability detection device, such as Figure 4 As shown, the device includes: an acquisition module 100, used to obtain a lithium hydroxide particle sample to be tested; a pretreatment module 200, used to pretreatment the lithium hydroxide particle sample to be tested; a detection module 300, used to detect the pretreated lithium hydroxide particle sample to be tested to obtain detection data of the lithium hydroxide particle sample to be tested; a model construction and training module 400, used to construct and train a lithium hydroxide metal particle stability detection model; a stability detection module 500, used to input the detection data of the lithium hydroxide particle sample to be tested into the detection model to obtain a stability index of the lithium hydroxide metal particle sample.

[0116] Optionally, the pre-processing module 200 includes: a cleaning submodule for cleaning the lithium hydroxide particle sample to be tested; and a drying submodule for drying the cleaned lithium hydroxide particle sample to be tested.

[0117] Based on the above embodiments, Figure 5 , for an explanation of the computer-readable storage medium of the exemplary embodiment of the present application, please refer to Figure 5 The computer-readable storage medium shown is a CD-ROM 40, on which a computer program (i.e., a program product) is stored. When the computer program is executed by a processor, it implements the steps described in the above method implementation, for example, obtaining a lithium hydroxide particle sample to be tested; pre-treating the lithium hydroxide particle sample to be tested; testing the pre-treated lithium hydroxide particle sample to be tested to obtain test data of the lithium hydroxide particle sample to be tested; constructing and training a lithium hydroxide metal particle stability detection model; and inputting the test data of the lithium hydroxide particle sample to be tested into the detection model to obtain a stability index of the lithium hydroxide metal particle sample. The specific implementation method of each step will not be repeated here.

[0118] It should be noted that the computer-readable storage medium includes but is not limited to phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which are not listed here one by one.

[0119] Based on the above embodiments, the present application also provides an electronic device, which is described below with reference to Figure 6 An electronic device for file downloading according to an exemplary embodiment of the present application is described.

[0120] Figure 6 A block diagram of an exemplary electronic device 50 suitable for implementing the embodiments of the present application is shown. The electronic device 50 may be a computer system or a cloud server. Figure 6 The electronic device 50 shown is only an example and should not limit the functions and scope of use of the embodiments of the present application.

[0121] like Figure 6 As shown, the electronic device 50 includes but is not limited to: one or more processors or processing units 501, a system memory 502, and a bus 503 connecting different system components (including the system memory 502 and the processing unit 501).

[0122] The electronic device 50 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 50, including volatile and non-volatile media, removable and non-removable media.

[0123] The system memory 502 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 5021 and / or cache memory 5022. The electronic device 50 may further include other removable / non-removable, volatile / non-volatile computer system storage media. For example only, the ROM 5023 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 is not shown in the , usually referred to as "hard drive"). Although not in Figure 6As shown in FIG, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) can be provided. In these cases, each drive can be connected to bus 503 via one or more data medium interfaces. System memory 502 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of various embodiments of the present application.

[0124] A program / utility 5025 having a set (at least one) of program modules 5024 may be stored, for example, in system memory 502. Such program modules 5024 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 5024 generally implement the functions and / or methods of the embodiments described herein.

[0125] The electronic device 50 can also communicate with one or more external devices 504 (such as a keyboard, a pointing device, a display, etc.). Such communication can be performed through an input / output (I / O) interface 505. In addition, the electronic device 50 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and / or a public network such as the Internet) through a network adapter 506. Figure 6 As shown, the network adapter 506 communicates with other modules (such as the processing unit 501, etc.) of the electronic device 50 via the bus 503. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 50 .

[0126] The processing unit 501 executes various functional applications and data processing by running programs stored in the system memory 502, such as obtaining a sample of lithium hydroxide particles to be tested; pre-processing the sample of lithium hydroxide particles to be tested; testing the pre-treated sample of lithium hydroxide particles to be tested to obtain test data of the sample of lithium hydroxide particles to be tested; constructing and training a lithium hydroxide metal particle stability detection model; and inputting the test data of the sample of lithium hydroxide particles to be tested into the detection model to obtain a stability index of the sample of lithium hydroxide metal particles. The specific implementation of each step will not be repeated here. It should be noted that although the detailed description above mentions several units / modules or sub-units / sub-modules of the concurrent file download device, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present application, the features and functions of two or more units / modules described above can be embodied in a single unit / module. Conversely, the features and functions of a single unit / module described above can be further divided to be embodied by multiple units / modules.

[0127] In the description of this application, it should be noted that the terms "first", "second" and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance.

[0128] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0129] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0130] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0131] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0132] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a cloud server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0133] The above embodiments are intended only to illustrate the technical concepts and features of this application. Their purpose is to enable those familiar with the art to understand the content of this application and implement it accordingly. They are not intended to limit the scope of protection of this application. Any equivalent changes or modifications made in accordance with the spirit of this application shall be included in the scope of protection of this application.

[0134] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for detecting the stability of lithium hydroxide particles, characterized in that: The method comprises: Obtaining a lithium hydroxide particle sample to be tested; Pre-treating the lithium hydroxide particle sample to be tested; Testing the pretreated lithium hydroxide particle sample to be tested to obtain test data of the lithium hydroxide particle sample to be tested; Construct and train a lithium hydroxide metal particle stability detection model; The detection data of the lithium hydroxide particle sample to be tested is input into the lithium hydroxide metal particle stability detection model to obtain the stability index of the lithium hydroxide metal particle sample.

2. The detection method according to claim 1, characterized in that: The lithium hydroxide particle sample to be tested is pretreated, comprising: Cleaning the lithium hydroxide particle sample to be tested; The cleaned lithium hydroxide particle sample to be tested is dried.

3. The method for detecting stability of lithium hydroxide particles according to claim 1, characterized in that: The testing of the pretreated lithium hydroxide particle sample to be tested comprises: The pretreated lithium hydroxide particle samples to be tested are subjected to physical property testing, chemical property testing and electrochemical performance testing, wherein the physical property testing includes specific surface area testing, particle size testing and color testing; the chemical property testing includes thermal stability testing, moisture content testing and impurity content testing; and the electrochemical performance testing includes cyclic voltammetry testing and constant current charge and discharge testing.

4. The method for detecting stability of lithium hydroxide particles according to claim 1, characterized in that: The lithium hydroxide metal particle stability detection model includes: A feature extraction network, used for extracting features from the detection data of the lithium hydroxide particle sample to be tested; A feature fusion network, used for fusing the features extracted by the feature extraction network; The output network is used to output the stability index of the lithium hydroxide particles to be tested.

5. The method for detecting stability of lithium hydroxide particles according to claim 1, characterized in that: The feature extraction network introduces superimposed convolutional layers to perform layer-by-layer feature extraction on the detection data of the lithium hydroxide particle sample to be tested.

6. The method for detecting stability of lithium hydroxide particles according to claim 1, characterized in that: The lithium hydroxide metal particle stability detection model is trained by the following steps: Collecting a large amount of lithium hydroxide particles and pre-treating them; Testing the pretreated lithium hydroxide particles, including performing a primary test to obtain primary test data, and performing a secondary test after simulation under different humidity conditions to obtain secondary test data; Integrate the initial detection data and the secondary detection data to form a detection data set, and divide the detection data set into a training set and a validation set in proportion; Set training parameters and train the model based on the training set until the maximum number of training times is reached; The model is verified based on the validation set, and the accuracy, recall and F1 score are used as evaluation indicators. When the accuracy, recall and F1 score reach the preset values, the model verification is passed; otherwise, the training parameters are adjusted or the training set samples are expanded to retrain the model until the model verification is passed.

7. A lithium hydroxide particle stability detection device, characterized in that: The device comprises: An acquisition module, used for acquiring a lithium hydroxide particle sample to be tested; A pretreatment module, used for pretreatment of the lithium hydroxide particle sample to be tested; A detection module, used to detect the pretreated lithium hydroxide particle sample to be tested to obtain detection data of the lithium hydroxide particle sample to be tested; Model building and training module, used to build and train the lithium hydroxide metal particle stability detection model; The stability detection module is used to input the detection data of the lithium hydroxide particle sample to be tested into the detection model to obtain the stability index of the lithium hydroxide metal particle sample.

8. The lithium hydroxide particle stability detection device according to claim 7, characterized in that: The preprocessing module comprises: A cleaning submodule, used for cleaning the lithium hydroxide particle sample to be tested; The drying submodule is used to dry the cleaned lithium hydroxide particle sample to be tested.

9. A storage medium, characterized in that: The method comprises instructions, which, when executed on a computer, enable the computer to execute the lithium hydroxide particle stability detection method according to any one of claims 1 to 6.

10. An electronic device, characterized in that: The electronic device comprises: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the lithium hydroxide particle stability detection method as described in any one of claims 1 to 6 is implemented.

Images