Detection and analysis method of nickel cobalt lithium manganate ternary lithium ion battery positive plate

Abstract

The invention relates to the technical field of lithium ion battery material detection, in particular to a detection and analysis method of a nickel cobalt lithium manganate ternary lithium ion battery positive plate, which comprises the following steps: S1, carrying out standardization treatment on an ion battery positive plate sample; s2, preliminarily identifying a structural stress concentration area; s3, applying different high voltages to the sample, and observing the microstructure evolution process of the connection part of the grain boundary and the particle; s4, acquiring chemical component change data of the positive plate material under voltage application; s5, in combination with the structure and component change data obtained in the S3 and the S4, performing correlation analysis on stress response and sample microstructure and chemical component change; and S6, establishing a risk assessment model for the irreversible phase change of the nickel cobalt lithium manganate ternary material. According to the method, the change of stress, microstructure and chemical components is monitored in real time, and the risk assessment model is established, so that early prediction and stability judgment of irreversible phase change of the nickel cobalt lithium manganate positive plate are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium ion battery material detection, and in particular to a detection and analysis method for a positive electrode sheet of a nickel-cobalt-lithium-manganese oxide ternary lithium ion battery. Background Art

[0002] Lithium-ion batteries are widely used in electric vehicles, energy storage devices and other fields due to their high energy density and long life. Among them, nickel cobalt manganese oxide (NCM) ternary materials have become one of the main materials for battery positive electrodes due to their good stability and cost advantages. However, NCM materials are prone to irreversible phase changes under high-rate charge and discharge and long-term cycle conditions, resulting in capacity decay and decreased electrochemical performance. Once this phase change occurs, it is usually irreversible, which seriously affects the stability and life of the battery. Therefore, its early identification and analysis are particularly critical. At present, conventional positive electrode detection methods are mostly limited to the overall electrochemical performance evaluation, and it is difficult to achieve dynamic detection and correlation analysis of structure, stress and chemical composition at the microscale, which makes the prediction and control of phase change risks have great limitations.

[0003] The existing technology lacks a systematic multi-dimensional detection method that can comprehensively evaluate and predict the risk of irreversible phase change of NCM positive electrode sheets by combining stress, microstructure and chemical composition changes. In the actual production and application process, these factors interact with each other, resulting in complex evolution of materials. Existing detection methods are often difficult to track synchronously, and lack correlation analysis tools to quantify their impact. Therefore, a detection and analysis method for nickel-cobalt-manganese-oxide lithium-ion battery positive electrode sheets is developed to solve the above problems. Summary of the invention

[0004] Based on the above purpose, the present invention provides a method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery.

[0005] A method for detecting and analyzing a positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery comprises the following steps:

[0006] S1: Standardize the positive electrode samples of nickel-cobalt-manganese-oxide ternary lithium-ion batteries, cut them into specified sizes, and remove surface impurities and oxide layers by cleaning to ensure the consistency of the surface state and microstructure of the samples;

[0007] S2: Apply a low-intensity, tiny prestress on the sample to preliminarily identify its structural stress concentration area; at the same time, the initial stress response of the positive electrode under prestress is recorded by a stress sensor to provide basic data for subsequent microstructural change detection;

[0008] S3: Use in-situ scanning transmission electron microscopy to monitor the crystal structure of the positive electrode sample at the microscale, and apply different high voltages to the sample to observe the microstructural evolution process at the grain boundary and particle connection to record the structural changes of the sample under high voltage;

[0009] S4: At the same time as step S3, the chemical composition changes on the sample surface are detected by in-situ X-ray photoelectron spectroscopy or X-ray diffraction analysis to track the distribution of nickel, cobalt, and manganese at grain boundaries and particle connections, and obtain the chemical composition change data of the positive electrode material under voltage application;

[0010] S5: Perform dynamic charge and discharge cycles on the cathode sample using an electrochemical workstation, record the stress change curve at each charge and discharge stage, and combine the structural and composition change data obtained in S3 and S4 to perform correlation analysis between the stress response and the changes in the sample microstructure and chemical composition;

[0011] S6: Based on the data obtained in S1 to S5, a risk assessment model for the irreversible phase change of nickel-cobalt-lithium manganese oxide ternary materials is established. The stress, microstructure and chemical composition changes of the positive electrode under the charge and discharge cycle are comprehensively calculated to generate a predicted evaluation value of the irreversible phase change, and then the stability of the positive electrode material is judged.

[0012] Optionally, the S1 specifically includes:

[0013] S11: Place the nickel-cobalt-manganese-oxide lithium-ion battery positive electrode sheet material on a standardized cutting device and cut it using a cutting tool with micron-level precision to make the sample size 10 mm long, 5 mm wide, and 100 μm thick to ensure the consistency of the sample size;

[0014] S12: Drying the cut samples by heating them to 80°C for 2 hours to remove moisture and volatile substances attached to the surface and ensure the dryness of the sample surface;

[0015] S13: using ultrasonic cleaning equipment to clean the sample surface under the action of ultrasonic waves with a frequency of 40kHz for 10 minutes to remove tiny impurity particles on the surface; the ultrasonic cleaning medium is an anhydrous ethanol solution with a concentration of 99.9%;

[0016] S14: placing the cleaned sample in an argon environment for low-temperature plasma surface treatment, with a plasma power of 50 W and a treatment time of 5 minutes, to remove the surface oxide layer through the activation of plasma;

[0017] S15: The treated sample is placed in a vacuum drying oven, the temperature is set to 60°C, the drying time is 4 hours, and the residual solvent is desorbed and the surface is dried to eliminate the residual liquid trace components in the cleaning process.

[0018] Optionally, the S2 specifically includes:

[0019] S21: Fix the positive electrode sheet sample of the nickel-cobalt-manganese-oxide lithium-ion battery on a micro-stress loading device, and use a piezoelectric driven loading device to apply a low-intensity prestress, and the prestress is set to 0.5 MPa to simulate the initial stress state in the actual working environment;

[0020] S22: During the process of loading the prestress, the stress response data of the sample is monitored and recorded in real time using a micro stress sensor attached to the sample. The sensitivity of the stress sensor is set to 0.01 MPa to ensure the accuracy of the stress response data and obtain the stress distribution of the positive electrode sheet under the prestress;

[0021] S23: Import the collected stress response data into a data recording device, identify the area of ​​uneven stress distribution in the positive electrode sample through data analysis, mark the stress concentration position, generate a stress distribution map of the sample, and provide basic data for subsequent microstructure change detection.

[0022] Optionally, the S23 specifically includes:

[0023] S231: Import the stress response data collected from the micro stress sensor into the data recording device in the form of a two-dimensional matrix, and set each element of the matrix as the stress value of a microscopic area in the sample, defined as σ i,j , where σ i,j represents the stress value located in the i-th row and j-th column of the matrix, where i and j are the number of rows and columns of the matrix respectively;

[0024] S232: Calculate the two-dimensional stress matrix σ i,j The standard deviation σ std , evaluate the stress uniformity of the sample, the formula is: Among them, σ avg Represents the average stress value of the stress matrix, expressed as: Where N represents the total number of elements in the matrix; if the standard deviation σ std If the preset uniformity threshold is exceeded, it means that there are areas of uneven stress in the sample;

[0025] S233: Calculation of stress concentration factor κ i,j To identify high stress areas, the formula is: Among them, κ i,j represents the stress concentration factor located in the i-th row and j-th column of the matrix; if κ at a certain position i,jIf it is higher than a predetermined threshold, the location is marked as a stress concentration area;

[0026] S234: According to the stress concentration factor matrix κ i,j Generate a stress distribution map for the high-value areas in the image, use color coding to display the stress concentration locations, and generate a visual stress distribution map.

[0027] Optionally, the S3 specifically includes:

[0028] S31: Fixing the positive electrode sample of the nickel-cobalt-lithium-manganese oxide ternary lithium-ion battery on the sample stage of the in-situ scanning transmission electron microscope;

[0029] S32: applying a first set of voltages to the sample through a voltage control module of the electron microscope, setting an initial voltage of 3 volts to simulate the crystal structure performance of the sample under normal working conditions; performing high-resolution imaging of the grain boundaries and particle connections of the sample using the electron microscope, and recording an initial structural image of the sample under the voltage;

[0030] S33: Under the premise of keeping other conditions unchanged, gradually increasing high voltage is applied to the sample, and the voltage is adjusted to 4 volts and 5 volts in sequence. Microscopic imaging is performed at each voltage level to capture the structural evolution of grain boundaries and particle connections, and generate a sequence of structural images at different voltages;

[0031] S34: Continue to observe the sample structure at the maximum set voltage of 5 volts, especially the deformation at the grain boundary and the relaxation of the particle connection, obtain the microstructure details of the sample under high voltage, and record the crystal structure change data of the sample, and archive the crystal evolution under different voltages;

[0032] S35: Processing the collected crystal structure image sequences at each voltage level, digitally marking the changing positions of grain boundaries and particle connections, and forming a high-voltage structural change map of the sample.

[0033] Optionally, the S4 specifically includes:

[0034] S41: placing the sample in the detection area of ​​the in-situ X-ray photoelectron spectroscopy device, and determining the elemental composition distribution of the sample surface area through initial scanning, recording the initial composition data of nickel, cobalt, and manganese, marking the distribution position of each element at the grain boundary and the particle connection, and forming an initial chemical composition map of the sample;

[0035] S42: During the process of gradually increasing the voltage, in-situ XPS spectrum acquisition is performed simultaneously to record the composition changes of nickel, cobalt and manganese elements at each voltage level.

[0036] S43: Quantitatively analyze the energy spectrum data of each element and calculate the element content change rate at each voltage. The formula is: Among them, C elem (V n+1 ) represents the voltage V n+1 The content of the following elements, C elem (V n ) represents the voltage V n Element content under ΔC elem Indicates the percentage change of each element's content under different voltages;

[0037] S44: Use XPS data analysis software to visualize the element distribution of nickel, cobalt, and manganese, generate element distribution maps at various voltage levels, and use color coding to mark the chemical composition changes at different locations, highlighting the differences in the distribution states of nickel, cobalt, and manganese at grain boundaries and particle connections;

[0038] S45: Compare and analyze the chemical composition distribution maps under different voltages, mark the areas where the composition changes significantly, and archive them to generate a chemical composition change map.

[0039] Optionally, the S5 specifically includes:

[0040] S51: Connect the positive electrode sample of the nickel-cobalt-manganese-oxide lithium-ion battery to the electrode fixture of the electrochemical workstation, configure it as a three-electrode system, set the cycle voltage range to 2.5 volts to 4.5 volts to simulate the actual working conditions of the sample, and record the voltage and current data at each charge and discharge stage;

[0041] S52: At each stage of the charge and discharge cycle, the real-time stress change of the sample is recorded by the stress sensor to generate a stress change curve σ(t), where σ(t) represents the change value of stress over time; and the maximum stress value and the minimum stress value after each charge or discharge are recorded as σ max and σ min ;

[0042] S53: Synchronizing the obtained stress change curve σ(t) with the structural change map and chemical composition change map during the charge and discharge phases, marking the structural change and chemical composition change values ​​at the corresponding time points as ΔS(t) and ΔC(t), respectively, where ΔS(t) represents the structural change amount, and ΔC(t) represents the chemical composition change amount;

[0043] S54: The correlation analysis method is used to correlate the stress change curve σ(t), the structural change ΔS(t) and the chemical composition change ΔC(t). The calculation formula is:

[0044]

[0045] Among them, ρ σ,S Represents the correlation coefficient between stress and structural change, ρσ,C Represents the correlation coefficient between stress and chemical composition change, σ avg , S avg ,C avg represent the average values ​​of stress, structural change and composition change respectively;

[0046] S55: According to the correlation coefficient ρ σ,S and ρ σ,C The magnitude of stress is determined, the strength of correlation between stress changes and structural and composition changes is determined, and the results are archived to generate a correlation analysis report.

[0047] Optionally, the S6 specifically includes:

[0048] S61: The stress change data, microstructure change data, chemical composition change data and correlation analysis results obtained in S1 to S5 are organized into a standardized data set, wherein the stress data includes the stress change curve during the charge and discharge stage, the structure data includes the crystal structure evolution under different voltage conditions, the composition data includes the chemical composition distribution under different voltages, and the correlation analysis data includes the correlation coefficients between stress and structure change, and between stress and composition change;

[0049] S62: Extract features from the standardized data set, taking the stress extreme value and change rate in the stress data as stress features, taking the change position of grain boundaries and particle connections in the microstructure data as structural features, taking the distribution deviation of nickel, cobalt, and manganese under high voltage conditions in the chemical composition change data as composition features, and using the correlation coefficient in S5 to adjust the weight of the features to ensure that the highly correlated features have a more important weight in the model;

[0050] S63: Construct a risk assessment model based on multidimensional characteristics, take the stress characteristics, structural characteristics and composition characteristics after adjusting the weights as the input variables of the model, use multidimensional regression analysis to establish the model, and form a multidimensional irreversible phase change risk assessment model;

[0051] S64: Input the charge and discharge cycle data of the positive electrode sample into the evaluation model, and calculate it in combination with the weight coefficient of each feature to generate a phase change risk score of the sample under different voltage and stress conditions, which is used to characterize the irreversible phase change trend under different working conditions;

[0052] S65: Generate a predicted evaluation value of irreversible phase change based on the risk score, compare the score value with a preset stability threshold, and determine whether the sample has relatively stable structural characteristics; if the score value exceeds the threshold, it is determined that the sample has a risk of irreversible phase change.

[0053] Optionally, the S62 specifically includes:

[0054] S621: Extract the correlation coefficient ρ between stress and microstructure change from the correlation analysis of S5 σ,S and the correlation coefficient ρ between stress and chemical composition change σ,C , where ρ σ,S Indicates the degree of correlation between stress and structural change, ρ σ,C Indicates the degree of correlation between stress and composition change, with a value range of [-1, 1]. The larger the absolute value, the stronger the correlation.

[0055] S622: Define initial weight value w σ ,w S and w C The initial weights corresponding to stress characteristics, structural characteristics and composition characteristics are set equal to meet the initial condition w σ +w S +w C =1;

[0056] S623: Use the correlation coefficient to adjust the weight of each feature and calculate the weight adjustment coefficient α σ,S and α σ,C , respectively represent the adjustment effects of stress-structure correlation and stress-component correlation on feature weights;

[0057] S624: Calculate the feature weight value w after weight adjustment σ ′,w S ′,w C ′, the formulas are:

[0058] w S ′=w S +α σ,S × S and w C ′=w C +α σ,C

[0059] × C ; Among them, w σ ′,w S ′,w C ′ is the weight value of stress characteristics, structural characteristics and composition characteristics after weight adjustment, It represents the average influence of two correlation coefficients on stress characteristics;

[0060] S625: The weight value w after weight adjustment σ ′、w S ′、w C 'Normalization is performed to ensure that the sum of the feature weights is still equal to 1. The normalization formula is:

[0061]

[0062] Among them, w σ ″、w S ″、w C ″ is the final adjusted feature weight.

[0063] Optionally, the S63 specifically includes:

[0064] S631: The stress characteristics, structural characteristics and composition characteristics after weight adjustment are expressed as feature vectors X σ ,X S and X C ;

[0065] S632: The feature vector X σ ,X S and X C Combine to form the multi-dimensional input vector X of the model;

[0066] S633: Set the irreversible phase change risk score Y as the output variable of the model to indicate the risk of irreversible phase change of the positive electrode under different conditions; establish a multidimensional regression model to establish a linear relationship between the input vector X and the output variable Y. The model is expressed as: Among them, β0 is the constant term of the model, β i is the regression coefficient of each input feature, X i represents the i-th eigenvalue in the input vector X;

[0067] S634: Solving the regression coefficient β by the least squares method i , so that the error between the predicted value and the true value is minimized;

[0068] S635: Use the trained regression coefficients to construct a final irreversible phase change risk assessment model to predict the phase change risk score of the sample under different charging and discharging conditions.

[0069] Beneficial effects of the present invention:

[0070] The present invention, by introducing a multi-dimensional detection method, can monitor the stress, microstructure and chemical composition changes in real time, and can accurately track the dynamic evolution of the nickel cobalt manganese oxide lithium positive electrode sheet material during the charge and discharge cycle. By using high-precision detection means such as in-situ scanning transmission electron microscopy and X-ray photoelectron spectroscopy, it is possible to accurately capture the microstructure and chemical composition changes at the grain boundaries and particle connections of the material under high voltage conditions, so that the early characteristics of the irreversible phase change can be identified, providing rich basic data for subsequent stability evaluation.

[0071] The present invention constructs a risk assessment model for irreversible phase change, comprehensively quantifies stress, structure and composition change data and generates a phase change risk score, which can intuitively judge the stability of the material. This method effectively improves the prediction accuracy of irreversible phase change, makes the material life and safety management more refined, and provides important support for improving the performance and reliability of lithium-ion batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0073] Figure 1 Schematic diagram of a method for detecting and analyzing a positive electrode sheet of a battery according to an embodiment of the present invention;

[0074] Figure 2 The figure is a schematic diagram of the process of establishing a risk assessment model for irreversible phase change according to an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0076] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0077] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0078] like Figure 1-Figure 2 As shown, a method for detecting and analyzing a positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery comprises the following steps:

[0079] S1: Standardize the positive electrode samples of nickel-cobalt-manganese-oxide ternary lithium-ion batteries, cut them into specified sizes, and remove surface impurities and oxide layers by cleaning to ensure the consistency of the surface state and microstructure of the samples;

[0080] S2: Apply a low-intensity, tiny prestress on the sample to preliminarily identify its structural stress concentration area; at the same time, the initial stress response of the positive electrode under prestress is recorded by a stress sensor to provide basic data for subsequent microstructural change detection;

[0081] S3: Use in-situ scanning transmission electron microscopy to monitor the crystal structure of the positive electrode sample at the microscale, and apply different high voltages to the sample to observe the microstructural evolution process at the grain boundary and particle connection to record the structural changes of the sample under high voltage;

[0082] S4: At the same time as step S3, the chemical composition changes on the sample surface are detected by in-situ X-ray photoelectron spectroscopy or X-ray diffraction analysis to track the distribution of nickel, cobalt, and manganese at grain boundaries and particle connections, and obtain the chemical composition change data of the positive electrode material under voltage application;

[0083] S5: Perform dynamic charge and discharge cycles on the cathode sample using an electrochemical workstation, record the stress change curve at each charge and discharge stage, and combine the structural and composition change data obtained in S3 and S4 to perform correlation analysis between the stress response and the changes in the sample microstructure and chemical composition;

[0084] S6: Based on the data obtained in S1 to S5, a risk assessment model for the irreversible phase change of nickel-cobalt-lithium manganese oxide ternary materials is established. The stress, microstructure and chemical composition changes of the positive electrode under the charge and discharge cycle are comprehensively calculated to generate a predicted evaluation value of the irreversible phase change, and then the stability of the positive electrode material is judged.

[0085] S1 specifically includes:

[0086] S11: Place the nickel-cobalt-manganese-oxide lithium-ion battery positive electrode sheet material on a standardized cutting device and cut it using a cutting tool with micron-level precision to make the sample size 10 mm long, 5 mm wide, and 100 μm thick, ensuring the consistency of the sample size so as to maintain consistent structural characteristics in subsequent testing;

[0087] S12: Dry the cut samples by heating them to 80°C for 2 hours to remove moisture and volatile substances attached to the surface, ensure the dryness of the sample surface, and lay the foundation for further cleaning and removal of impurities and oxide layers;

[0088] S13: Using ultrasonic cleaning equipment, the sample surface is cleaned under the action of ultrasonic waves with a frequency of 40kHz for 10 minutes to remove tiny impurity particles on the surface; the ultrasonic cleaning medium is an anhydrous ethanol solution with a concentration of 99.9%, thereby ensuring the thoroughness of cleaning and the uniformity of the surface state;

[0089] S14: placing the cleaned sample in an argon environment for low-temperature plasma surface treatment, with a plasma power of 50 W and a treatment time of 5 minutes, to remove the surface oxide layer through the activation of plasma, ensuring that the oxide layer and impurities on the surface of the sample are completely removed to obtain a uniform surface state;

[0090] S15: Place the treated samples in a vacuum drying oven, set the temperature to 60°C, and dry for 4 hours to desorb the residual solvent and dry the surface to eliminate the residual liquid trace components in the cleaning process and ensure the dryness and stability of the surface state of the samples. Through the above standardized processing steps, the positive electrode samples have highly consistent size, surface state and microstructure, which lays a stable foundation for subsequent detection and analysis, ensuring the accuracy of the test results and the repeatability of the data.

[0091] S2 specifically includes:

[0092] S21: Fix the positive electrode sheet sample of the nickel-cobalt-manganese-oxide lithium-ion battery on a micro-stress loading device, and use a piezoelectric driven loading device to apply a low-intensity prestress, and the prestress is set to 0.5 MPa to simulate the initial stress state in the actual working environment;

[0093] S22: During the process of loading the prestress, the stress response data of the sample is monitored and recorded in real time using a micro stress sensor attached to the sample. The sensitivity of the stress sensor is set to 0.01 MPa to ensure the accuracy of the stress response data and obtain the stress distribution of the positive electrode sheet under the prestress;

[0094] S23: Import the collected stress response data into a data recording device, identify the area of ​​uneven stress distribution in the positive electrode sample through data analysis, mark the stress concentration position, generate a stress distribution map of the sample, and provide basic data for subsequent microstructure change detection.

[0095] S23 specifically includes:

[0096] S231: Import the stress response data collected from the micro stress sensor into the data recording device in the form of a two-dimensional matrix, and set each element of the matrix as the stress value of a microscopic area in the sample, defined as σ i,j , where σ i,jrepresents the stress value located in the i-th row and j-th column of the matrix, where i and j are the number of rows and columns of the matrix respectively;

[0097] S232: Calculate the two-dimensional stress matrix σ i,j The standard deviation σ std , evaluate the stress uniformity of the sample, the formula is: Among them, σ avg Represents the average stress value of the stress matrix, expressed as: Where N represents the total number of elements in the matrix; if the standard deviation σ std If the preset uniformity threshold is exceeded, it means that there are areas of uneven stress in the sample;

[0098] S233: Calculation of stress concentration factor κ i,j To identify high stress areas, the formula is: Among them, κ i,j represents the stress concentration factor located in the i-th row and j-th column of the matrix; if κ at a certain position i,j If it is higher than a predetermined threshold (e.g., 1.2), the location is marked as a stress concentration area;

[0099] S234: According to the stress concentration factor matrix κ i,j A stress distribution map is generated for the high-value areas in the sample, and the stress concentration locations are displayed using color coding to generate a visual stress distribution map in order to identify areas with uneven stress distribution in the positive electrode sample. By calculating the standard deviation and stress concentration coefficient, the stress concentration areas in the positive electrode sample can be accurately identified and marked, providing accurate stress distribution information for subsequent microstructural change detection.

[0100] S3 specifically includes:

[0101] S31: Fix the positive electrode sheet sample of the nickel-cobalt-manganese-oxide ternary lithium-ion battery on the sample stage of the in-situ scanning transmission electron microscope to ensure the stable positioning of the sample for accurate microstructure observation;

[0102] S32: applying a first set of voltages to the sample through a voltage control module of the electron microscope, setting an initial voltage of 3 volts to simulate the crystal structure performance of the sample under normal working conditions; performing high-resolution imaging of the grain boundaries and particle connections of the sample using the electron microscope, and recording an initial structural image of the sample under the voltage;

[0103] S33: Under the premise of keeping other conditions unchanged, gradually increasing high voltage is applied to the sample, and the voltage is adjusted to 4 volts and 5 volts in sequence. Microscopic imaging is performed at each voltage level to capture the structural evolution of grain boundaries and particle connections, and generate a sequence of structural images at different voltages to facilitate the observation of dynamic changes in the microstructure;

[0104] S34: Continue to observe the sample structure at the maximum set voltage of 5 volts, especially the deformation at the grain boundary and the relaxation of the particle connection, obtain the microstructure details of the sample under high voltage, and record the crystal structure change data of the sample, and archive the crystal evolution under different voltages for subsequent analysis;

[0105] S35: Process the collected crystal structure image sequences at various voltage levels to digitally mark the changing positions of grain boundaries and particle connections, forming a high-voltage structural change map of the sample, providing basic data for further analysis of irreversible phase changes.

[0106] The specific steps include:

[0107] S351: Perform image difference processing on the image sequences under different voltages, and calculate the intensity change ΔI of each pixel using the following formula: i,j :ΔI i,j =I i,j (V n+1 )-I i,j (V n ), where I i,j (V n+1 ) indicates that position i, j is at voltage V n+1 The pixel intensity under i,j (V n ) means the voltage V n The pixel intensity under ΔI i,j Indicates the intensity difference at different voltages;

[0108] S352: Setting threshold ΔI th Binarize the difference image. If ΔI i,j >ΔI th Then the marked position i, j is the position where the grain boundary or particle connection changes, and the area where the difference value exceeds the threshold value indicates that the microstructure of this position has changed significantly under the voltage change;

[0109] S353: All marked change positions are stored in digital coding form and their coordinates are located to form a high-voltage structural change map including the grain boundary and particle connection change areas, which is used to visualize the structural changes of the sample under different high voltages; through the above steps, the structural changes of the positive electrode sample under high voltage can be clearly marked through digital image difference and binarization processing technology to form an accurate high-voltage structural change map, providing detailed and intuitive microstructure data for irreversible phase change analysis.

[0110] S4 specifically includes:

[0111] S41: placing the sample in the detection area of ​​the in-situ X-ray photoelectron spectroscopy device, and determining the elemental composition distribution of the sample surface area through initial scanning, recording the initial composition data of nickel, cobalt, and manganese, marking the distribution position of each element at the grain boundary and the particle connection, and forming an initial chemical composition map of the sample;

[0112] S42: During the process of gradually increasing the voltage (in the voltage range from 3V to 5V), in-situ XPS spectrum acquisition is performed synchronously to record the composition changes of nickel (Ni), cobalt (Co), and manganese (Mn) elements at each voltage level, so as to capture the effect of voltage application on chemical composition distribution in real time;

[0113] S43: Quantitatively analyze the energy spectrum data of each element and calculate the element content change rate at each voltage. The formula is: Among them, C elem (V n+1 ) represents the voltage V n+1 The content of the following elements, C elem (V n ) represents the voltage V n Element content under ΔC elem Indicates the percentage change of each element's content under different voltages;

[0114] S44: Use XPS data analysis software to visualize the element distribution of nickel, cobalt, and manganese, generate element distribution maps at various voltage levels, and use color coding to mark the chemical composition changes at different locations, highlighting the differences in the distribution states of nickel, cobalt, and manganese at grain boundaries and particle connections;

[0115] S45: Compare and analyze the chemical composition distribution maps under different voltages, mark the areas with significant composition changes, and archive to generate chemical composition change maps to characterize the composition evolution of the positive electrode material under high voltage conditions, and provide accurate chemical composition data for subsequent irreversible phase change analysis; through the above steps, the composition changes of nickel, cobalt and manganese can be accurately tracked through in-situ X-ray photoelectron spectroscopy technology, and the composition change data under different voltage conditions can be combined to generate maps, which clearly show the chemical composition evolution of grain boundaries and particle connections, and provide detailed chemical information for subsequent phase change analysis.

[0116] S5 specifically includes:

[0117] S51: Connect the positive electrode sample of the nickel-cobalt-manganese-oxide lithium-ion battery to the electrode fixture of the electrochemical workstation, configure it as a three-electrode system, set the cycle voltage range to 2.5 volts to 4.5 volts to simulate the actual working conditions of the sample, and record the voltage and current data at each charge and discharge stage;

[0118] S52: At each stage of the charge and discharge cycle, the real-time stress change of the sample is recorded by the stress sensor to generate a stress change curve σ(t), where σ(t) represents the change value of stress over time; and the maximum stress value and the minimum stress value after each charge or discharge are recorded as σ max and σ min ;

[0119] S53: Synchronize the obtained stress change curve σ(t) with the structural change map and chemical composition change map during the charge and discharge phase (from steps S3 and S4, respectively), and mark the structural change and chemical composition change values ​​at the corresponding time points as ΔS(t) and ΔC(t), respectively, where ΔS(t) represents the structural change amount, and ΔC(t) represents the chemical composition change amount;

[0120] S54: The correlation analysis method is used to correlate the stress change curve σ(t), the structural change ΔS(t) and the chemical composition change ΔC(t). The calculation formula is:

[0121]

[0122] Among them, ρ σ,S Represents the correlation coefficient between stress and structural change, ρ σ,C Represents the correlation coefficient between stress and chemical composition change, σ avg , S avg ,C avg represent the average values ​​of stress, structural change and composition change respectively;

[0123] S55: According to the correlation coefficient ρ σ,S and ρ σ,C The size of the stress change can be determined to determine the strength of the correlation between the stress change and the structural and composition changes, and the results can be archived to generate a correlation analysis report to provide a basis for the subsequent irreversible phase change risk assessment. Through the above steps, the stress change curve in the dynamic charge and discharge cycle can be combined with the structural and composition change data to accurately analyze the correlation between stress, structure and chemical composition, providing reliable correlation data support for the early warning of irreversible phase change.

[0124] S6 specifically includes:

[0125] S61: The stress change data, microstructure change data, chemical composition change data and correlation analysis results obtained in S1 to S5 are organized into a standardized data set, wherein the stress data includes the stress change curve during the charge and discharge stage, the structure data includes the crystal structure evolution under different voltage conditions, the composition data includes the chemical composition distribution under different voltages, and the correlation analysis data includes the correlation coefficients between stress and structure change, and between stress and composition change;

[0126] S62: Extract features from the standardized data set, taking the stress extreme value and change rate in the stress data as stress features, taking the change position of grain boundaries and particle connections in the microstructure data as structural features, taking the distribution deviation of nickel, cobalt, and manganese under high voltage conditions in the chemical composition change data as composition features, and using the correlation coefficient in S5 to adjust the weight of the features to ensure that the highly correlated features have a more important weight in the model;

[0127] S63: Construct a risk assessment model based on multidimensional characteristics, take the stress characteristics, structural characteristics and composition characteristics after adjusting the weights as the input variables of the model, use multidimensional regression analysis to establish the model, and form a multidimensional irreversible phase change risk assessment model;

[0128] S64: Input the charge and discharge cycle data of the positive electrode sample into the evaluation model, and calculate it in combination with the weight coefficient of each feature to generate a phase change risk score of the sample under different voltage and stress conditions, which is used to characterize the irreversible phase change trend under different working conditions;

[0129] S65: Generate a predicted evaluation value of irreversible phase change based on the risk score, compare the score with the preset stability threshold, and determine whether the sample has relatively stable structural characteristics; if the score exceeds the threshold, it is determined that the sample has an irreversible phase change risk, and it is recommended to further evaluate its stability; through the above steps, a risk assessment model can be constructed using standardized data sets and multi-dimensional feature weight adjustment, and the feature weights can be optimized in combination with the correlation coefficient to achieve phase change risk prediction and stability judgment of the sample, providing a reliable quantitative basis for early warning of irreversible phase change.

[0130] The weight adjustment of features in S62 specifically includes:

[0131] S621: Extract the correlation coefficient ρ between stress and microstructure change from the correlation analysis of S5 σ,S and the correlation coefficient ρ between stress and chemical composition change σ,C , where ρ σ,S Indicates the degree of correlation between stress and structural change, ρ σ,C Indicates the degree of correlation between stress and composition change, with a value range of [-1, 1]. The larger the absolute value, the stronger the correlation.

[0132] S622: Define initial weight value w σ ,w S and w C The initial weights corresponding to stress characteristics, structural characteristics and composition characteristics are set equal to meet the initial condition w σ +w S +w C =1;

[0133] S623: Use the correlation coefficient to adjust the weight of each feature and calculate the weight adjustment coefficient α σ,S and α σ,C , respectively represent the adjustment effect of stress-structure correlation and stress-component correlation on feature weights, and the calculation formula is as follows: α σ,S =|ρ σ,S |;α σ,C =|ρ σ,C |, where |ρ σ,S | and |ρ σ,C | are the absolute values ​​of the correlation coefficients, which are used to avoid negative correlation affecting weight adjustment. The larger the absolute value, the higher the weight adjustment of the feature;

[0134] S624: Calculate the feature weight value w after weight adjustment σ ′,w S ′,w C ′, the formulas are:

[0135] w S ′=w S +α σ,S × S and w C ′=w C +α σ,C

[0136] × C ; Among them, w σ ′,w S ′,w C ′ is the weight value of stress characteristics, structural characteristics and composition characteristics after weight adjustment, It represents the average influence of two correlation coefficients on stress characteristics;

[0137] S625: The weight value w after weight adjustment σ ′、w S ′、w C 'Normalization is performed to ensure that the sum of the feature weights is still equal to 1. The normalization formula is:

[0138]

[0139] Among them, w σ ″、w S ″、w C ″ is the final adjusted feature weight, which is used for the feature input of the model; through the above steps, S62 can accurately adjust the weight of each feature according to the correlation coefficient, ensuring that the highly correlated features occupy a more important weight in the model, thereby improving the prediction accuracy and stability of the model for irreversible phase change.

[0140] The risk assessment model constructed in S63 specifically includes:

[0141] S631: The stress characteristics, structural characteristics and composition characteristics after weight adjustment are expressed as feature vectors X σ ,X S and X C , where X σ =[σ max , σ min , σ rate ] represents the stress characteristic vector, including the maximum stress value σ max , minimum stress value σ min , stress change rate σ rate ;X S =σ max , σ min , σ rate ], represents the structural feature vector, including the structural change position S loc , change distribution S dist and the rate of change S rate ;X C =[C Ni , C Co , C Mn ] represents the composition characteristic vector, including the content changes of nickel, cobalt and manganese C Ni ,C Co , C Mn ;

[0142] S632: The feature vector X σ ,X S and X C The multidimensional input vector X of the model is combined and expressed as: X = [X σ ,X S ,X C ]=[σ max , σ min , σ rate , σ max , σ min , σ rate , C Ni , C Co , C Mn ], the position of each feature in the input vector is arranged according to the weight adjustment result, so that high-weight features are preferentially involved in model training;

[0143] S633: Set the irreversible phase change risk score Y as the output variable of the model to indicate the risk of irreversible phase change of the positive electrode under different conditions; establish a multidimensional regression model to establish a linear relationship between the input vector X and the output variable Y. The model is expressed as: Among them, β0 is the constant term of the model, β iis the regression coefficient of each input feature, X i represents the i-th eigenvalue in the input vector X;

[0144] S634: Solving the regression coefficient β by the least squares method i , so as to minimize the error between the predicted value and the true value; the least squares method objective function is expressed as: Among them, Y k is the actual risk score of the kth sample, X i,k is the ith eigenvalue of the kth sample, N1 is the total number of samples, and the optimal regression coefficient β is obtained by solving the minimization problem i ;

[0145] S635: Using the trained regression coefficients, a final irreversible phase change risk assessment model is constructed to predict the phase change risk score of the sample under different charging and discharging conditions; the expression of the risk assessment model is: Y 预测 =β0+β1σ max +β2σ min +β3σ rate +β4S loc +β5S dist +β6S rate +β7C Ni +β8C Co +β9C Mn ; Through the above steps, the multidimensional regression analysis in S63 can associate stress, structure and composition characteristic data with phase change risk scores, build a complete irreversible phase change risk assessment model, and provide a reliable quantitative tool for stability prediction and phase change risk analysis of positive electrode materials.

[0146] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0147] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery, characterized in that: The following steps are involved: S1: Standardize the positive electrode samples of nickel-cobalt-manganese-oxide ternary lithium-ion batteries, cut them into specified sizes, and remove surface impurities and oxide layers by cleaning to ensure the consistency of the surface state and microstructure of the samples; S2: Apply a low-intensity, tiny prestress on the sample to preliminarily identify its structural stress concentration area; at the same time, the initial stress response of the positive electrode under prestress is recorded by a stress sensor to provide basic data for subsequent microstructural change detection; S3: Use in-situ scanning transmission electron microscopy to monitor the crystal structure of the positive electrode sample at the microscale, and apply different high voltages to the sample to observe the microstructural evolution process at the grain boundary and particle connection to record the structural changes of the sample under high voltage; S4: At the same time as step S3, the chemical composition changes on the sample surface are detected by in-situ X-ray photoelectron spectroscopy or X-ray diffraction analysis to track the distribution of nickel, cobalt, and manganese at grain boundaries and particle connections, and obtain the chemical composition change data of the positive electrode material under voltage application; S5: Perform dynamic charge and discharge cycles on the cathode sample using an electrochemical workstation, record the stress change curve at each charge and discharge stage, and combine the structural and composition change data obtained in S3 and S4 to perform correlation analysis between the stress response and the changes in the sample microstructure and chemical composition; S6: Based on the data obtained in S1 to S5, a risk assessment model for the irreversible phase change of nickel-cobalt-lithium manganese oxide ternary materials is established. The stress, microstructure and chemical composition changes of the positive electrode under the charge and discharge cycle are comprehensively calculated to generate a predicted evaluation value of the irreversible phase change, and then the stability of the positive electrode material is judged.

2. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 1, characterized in that: The S1 specifically includes: S11: Place the nickel-cobalt-manganese-oxide lithium-ion battery positive electrode sheet material on a standardized cutting device and cut it using a cutting tool with micron-level precision to make the sample size 10 mm long, 5 mm wide, and 100 μm thick to ensure the consistency of the sample size; S12: Drying the cut samples by heating them to 80°C for 2 hours to remove moisture and volatile substances attached to the surface and ensure the dryness of the sample surface; S13: using ultrasonic cleaning equipment to clean the sample surface under the action of ultrasonic waves with a frequency of 40kHz for 10 minutes to remove tiny impurity particles on the surface; the ultrasonic cleaning medium is an anhydrous ethanol solution with a concentration of 99.9%; S14: placing the cleaned sample in an argon environment for low-temperature plasma surface treatment, with a plasma power of 50 W and a treatment time of 5 minutes, to remove the surface oxide layer through the activation of plasma; S15: The treated sample is placed in a vacuum drying oven, the temperature is set to 60°C, the drying time is 4 hours, and the residual solvent is desorbed and the surface is dried to eliminate the residual liquid trace components in the cleaning process.

3. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 1, characterized in that: The S2 specifically includes: S21: Fix the positive electrode sheet sample of the nickel-cobalt-manganese-oxide lithium-ion battery on a micro-stress loading device, and use a piezoelectric driven loading device to apply a low-intensity prestress, and the prestress is set to 0.5 MPa to simulate the initial stress state in the actual working environment; S22: During the process of loading the prestress, the stress response data of the sample is monitored and recorded in real time using a micro stress sensor attached to the sample. The sensitivity of the stress sensor is set to 0.01 MPa to ensure the accuracy of the stress response data and obtain the stress distribution of the positive electrode sheet under the prestress; S23: Import the collected stress response data into a data recording device, identify the area of ​​uneven stress distribution in the positive electrode sample through data analysis, mark the stress concentration position, generate a stress distribution map of the sample, and provide basic data for subsequent microstructure change detection.

4. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium ternary lithium-ion battery according to claim 3, characterized in that: The S23 specifically includes: S231: Import the stress response data collected from the micro stress sensor into the data recording device in the form of a two-dimensional matrix, and set each element of the matrix as the stress value of a microscopic area in the sample, defined as σ i,j , where σ i,j represents the stress value located in the i-th row and j-th column of the matrix, where i and j are the number of rows and columns of the matrix respectively; S232: Calculate the two-dimensional stress matrix σ i,j The standard deviation σ std , evaluate the stress uniformity of the sample, the formula is: Among them, σ avg Represents the average stress value of the stress matrix, expressed as: Where N represents the total number of elements in the matrix; if the standard deviation σ std If the preset uniformity threshold is exceeded, it means that there are areas of uneven stress in the sample; S233: Calculation of stress concentration factor κ i,j To identify high stress areas, the formula is: Among them, κ i,j represents the stress concentration factor located in the i-th row and j-th column of the matrix; if κ at a certain position i,j If it is higher than a predetermined threshold, the location is marked as a stress concentration area; S234: According to the stress concentration factor matrix κ i,j Generate a stress distribution map for the high-value areas in the image, use color coding to display the stress concentration locations, and generate a visual stress distribution map.

5. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 1, characterized in that: The S3 specifically includes: S31: Fixing the positive electrode sample of the nickel-cobalt-lithium-manganese oxide ternary lithium-ion battery on the sample stage of the in-situ scanning transmission electron microscope; S32: applying a first set of voltages to the sample through a voltage control module of the electron microscope, setting an initial voltage of 3 volts to simulate the crystal structure performance of the sample under normal working conditions; performing high-resolution imaging of the grain boundaries and particle connections of the sample using the electron microscope, and recording an initial structural image of the sample under the voltage; S33: Under the premise of keeping other conditions unchanged, gradually increasing high voltage is applied to the sample, and the voltage is adjusted to 4 volts and 5 volts in sequence. Microscopic imaging is performed at each voltage level to capture the structural evolution of grain boundaries and particle connections, and generate a sequence of structural images at different voltages; S34: Continue to observe the sample structure at the maximum set voltage of 5 volts, especially the deformation at the grain boundary and the relaxation of the particle connection, obtain the microstructure details of the sample under high voltage, and record the crystal structure change data of the sample, and archive the crystal evolution under different voltages; S35: Processing the collected crystal structure image sequences at each voltage level, digitally marking the changing positions of grain boundaries and particle connections, and forming a high-voltage structural change map of the sample.

6. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 1, characterized in that: The S4 specifically includes: S41: placing the sample in the detection area of ​​the in-situ X-ray photoelectron spectroscopy device, and determining the elemental composition distribution of the sample surface area through initial scanning, recording the initial composition data of nickel, cobalt, and manganese, marking the distribution position of each element at the grain boundary and the particle connection, and forming an initial chemical composition map of the sample; S42: During the process of gradually increasing the voltage, in-situ XPS spectrum acquisition is performed simultaneously to record the composition changes of nickel, cobalt and manganese elements at each voltage level; S43: Quantitatively analyze the energy spectrum data of each element and calculate the element content change rate at each voltage. The formula is: Among them, C elem (V n+1 ) represents the voltage V n+1 The content of the following elements, C elem (V n ) represents the voltage V n Element content under ΔC elem Indicates the percentage change of each element's content under different voltages; S44: Use XPS data analysis software to visualize the element distribution of nickel, cobalt, and manganese, generate element distribution maps at various voltage levels, and use color coding to mark the chemical composition changes at different locations, highlighting the differences in the distribution states of nickel, cobalt, and manganese at grain boundaries and particle connections; S45: Compare and analyze the chemical composition distribution maps under different voltages, mark the areas where the composition changes significantly, and archive them to generate a chemical composition change map.

7. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 1, characterized in that: The S5 specifically includes: S51: Connect the positive electrode sample of the nickel-cobalt-manganese-oxide lithium-ion battery to the electrode fixture of the electrochemical workstation, configure it as a three-electrode system, set the cycle voltage range to 2.5 volts to 4.5 volts to simulate the actual working conditions of the sample, and record the voltage and current data at each charge and discharge stage; S52: At each stage of the charge and discharge cycle, the real-time stress change of the sample is recorded by the stress sensor to generate a stress change curve σ(t), where σ(t) represents the change value of stress over time; and the maximum stress value and the minimum stress value after each charge or discharge are recorded as σ max and σ min ; S53: Synchronizing the obtained stress change curve σ(t) with the structural change map and chemical composition change map during the charge and discharge phases, marking the structural change and chemical composition change values ​​at the corresponding time points as ΔS(t) and ΔC(t), respectively, where ΔS(t) represents the structural change amount, and ΔC(t) represents the chemical composition change amount; S54: The correlation analysis method is used to correlate the stress change curve σ(t), the structural change ΔS(t) and the chemical composition change ΔC(t). The calculation formula is: Among them, ρ σ,S Represents the correlation coefficient between stress and structural change, ρ σ,C Represents the correlation coefficient between stress and chemical composition change, σ avg , S avg ,C avg represent the average values ​​of stress, structural change and composition change respectively; S55: According to the correlation coefficient ρ σ,S and ρ σ,C The magnitude of stress is determined, the strength of correlation between stress changes and structural and composition changes is determined, and the results are archived to generate a correlation analysis report.

8. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 7, characterized in that: The S6 specifically includes: S61: The stress change data, microstructure change data, chemical composition change data and correlation analysis results obtained in S1 to S5 are organized into a standardized data set, wherein the stress data includes the stress change curve during the charge and discharge stage, the structure data includes the crystal structure evolution under different voltage conditions, the composition data includes the chemical composition distribution under different voltages, and the correlation analysis data includes the correlation coefficients between stress and structure change, and between stress and composition change; S62: Extract features from the standardized data set, taking the stress extreme value and change rate in the stress data as stress features, taking the change position of grain boundaries and particle connections in the microstructure data as structural features, taking the distribution deviation of nickel, cobalt, and manganese under high voltage conditions in the chemical composition change data as composition features, and using the correlation coefficient in S5 to adjust the weight of the features to ensure that the highly correlated features have a more important weight in the model; S63: Construct a risk assessment model based on multidimensional characteristics, take the stress characteristics, structural characteristics and composition characteristics after adjusting the weights as the input variables of the model, use multidimensional regression analysis to establish the model, and form a multidimensional irreversible phase change risk assessment model; S64: Input the charge and discharge cycle data of the positive electrode sample into the evaluation model, and calculate it in combination with the weight coefficient of each feature to generate a phase change risk score of the sample under different voltage and stress conditions, which is used to characterize the irreversible phase change trend under different working conditions; S65: Generate a predicted evaluation value of irreversible phase change based on the risk score, compare the score value with a preset stability threshold, and determine whether the sample has relatively stable structural characteristics; if the score value exceeds the threshold, it is determined that the sample has a risk of irreversible phase change.

9. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 8, characterized in that: The S62 specifically includes: S621: Extract the correlation coefficient ρ between stress and microstructure change from the correlation analysis of S5 σ,S and the correlation coefficient ρ between stress and chemical composition change σ,C , where ρ σ,S Indicates the correlation between stress and structural change, ρ σ,C Indicates the degree of correlation between stress and composition change, with a value range of [-1, 1]. The larger the absolute value, the stronger the correlation. S622: Define initial weight value w σ ,w S and w C The initial weights corresponding to stress characteristics, structural characteristics and composition characteristics are set equal to meet the initial condition w σ +w S +w C =1; S623: Use the correlation coefficient to adjust the weight of each feature and calculate the weight adjustment coefficient α σ,S and α σ,C , respectively represent the adjustment effects of stress-structure correlation and stress-component correlation on feature weights; S624: Calculate the feature weight value w after weight adjustment σ ′,w S ′,w C ′, the formulas are: w S ′ = w S + α σ,S × w S and w C ′ = w C + α σ,C × C ; Among them, w σ ′,w S ′,w C ′ is the weight value of stress characteristics, structural characteristics and composition characteristics after weight adjustment, It represents the average influence of two correlation coefficients on stress characteristics; S625: The weight value w after weight adjustment σ ′、w S ′、w C 'Normalization is performed to ensure that the sum of the feature weights is still equal to 1. The normalization formula is: Among them, w σ ″、w S ″、w C ″ is the final adjusted feature weight.

10. The method for detecting and analyzing the positive electrode sheet of a nickel-cobalt-manganese-oxide lithium-ion battery according to claim 9, characterized in that: The S63 specifically includes: S631: The stress characteristics, structural characteristics and composition characteristics after weight adjustment are expressed as feature vectors X σ ,X S and X C ; S632: The feature vector X σ ,X S and X C Combine to form the multi-dimensional input vector X of the model; S633: Set the irreversible phase change risk score Y as the output variable of the model to indicate the risk of irreversible phase change of the positive electrode under different conditions; establish a multidimensional regression model to establish a linear relationship between the input vector X and the output variable Y. The model is expressed as: Among them, β0 is the constant term of the model, β i is the regression coefficient of each input feature, X i represents the i-th eigenvalue in the input vector X; S634: Solving the regression coefficient β by the least squares method i , so that the error between the predicted value and the true value is minimized; S635: Use the trained regression coefficients to construct a final irreversible phase change risk assessment model to predict the phase change risk score of the sample under different charging and discharging conditions.

Images