AGM separator thermal stability monitoring method and system

Abstract

The present application provides a method and system for monitoring the thermal stability of AGM separators, which relates to the field of thermal insulation monitoring technology, including: receiving a set of AGM separator simulation models, performing multi-level clustering analysis, and constructing static constraints; generating temperature and failure monitoring duration record data by randomly configuring temperature elements, and configuring a failure duration prediction network; obtaining the predicted failure duration; and when the predicted duration is less than or equal to a threshold, marking thermal stability anomalies. The present application can solve the technical problem that the prior art is difficult to accurately predict and monitor the thermal stability of AGM separators under different working conditions, resulting in poor thermal stability management of AGM separators. Effective management and monitoring of the thermal stability of AGM separators can be achieved, thereby improving the reliability and safety of the battery system.

Description

Technical Field

[0001] The present application relates to the field of thermal insulation monitoring technology, and in particular to a method and system for monitoring the thermal stability of an AGM separator. Background Art

[0002] The main material of AGM separator is glass fiber, which has excellent mechanical, physical and chemical properties. It plays the role of preventing positive and negative short circuits, absorbing effective acid, maintaining a certain elasticity and providing oxygen channels in the battery. Since AGM separator has a significant impact on battery performance, its performance testing is crucial.

[0003] At present, it is difficult for existing technologies to accurately predict and monitor the thermal stability of AGM separators under different working conditions, resulting in poor thermal stability management of AGM separators, which further affects the reliability and safety of the battery system. Summary of the invention

[0004] The purpose of this application is to provide a method and system for monitoring the thermal stability of AGM separators, so as to solve the technical problem that the prior art is difficult to accurately predict and monitor the thermal stability of AGM separators under different working conditions, resulting in poor thermal stability management effects of AGM separators.

[0005] In view of the above problems, the present application provides a method and system for monitoring the thermal stability of an AGM separator.

[0006] In a first aspect, the present application provides a method for monitoring the thermal stability of an AGM separator, which is implemented by an AGM separator thermal stability monitoring system, wherein the method comprises: receiving an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, an pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set and a tensile strength label set; performing multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models; traversing the multiple groups of AGM separator simulation models to extract multiple centroid thickness labels, multiple centroid aperture labels, Multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels are used to construct multiple static constraints; the multiple static constraints are traversed to randomly configure samples of temperature elements to generate multiple groups of temperature element record data and multiple groups of failure monitoring time record data; multiple failure time prediction networks are configured according to the multiple groups of temperature element record data and the multiple groups of failure monitoring time record data; the temperature element monitoring characteristic value of the first AGM partition is collected, and the multiple failure time prediction networks are selectively activated to obtain the AGM partition failure prediction time, wherein the first AGM partition belongs to the AGM partition simulation model set; when the AGM partition failure prediction time is less than or equal to the failure time threshold, the first AGM partition is marked as thermally abnormal according to the temperature element monitoring characteristic value.

[0007] In a second aspect, the present application further provides an AGM separator thermal stability monitoring system for executing the AGM separator thermal stability monitoring method as described in the first aspect, wherein the AGM separator thermal stability monitoring system comprises: a collection receiving module for receiving an AGM separator simulation model collection, wherein the AGM separator simulation model collection has a thickness label collection, an pore size label collection, a porosity label collection, a material composition ratio label collection, a wet pressure label collection and a tensile strength label collection; a multi-level clustering module for performing multi-level clustering on the AGM separator simulation model collection according to the thickness label collection, the pore size label collection, the porosity label collection, the material composition ratio label collection, the wet pressure label collection and the tensile strength label collection to obtain multiple groups of AGM separator simulation models; a constraint condition construction module for traversing the multiple groups of AGM separator simulation models to extract multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, A plurality of groups of centroid material composition ratio labels, a plurality of centroid wet pressure label sets and a plurality of centroid tensile strength labels are used to construct a plurality of static constraints; a recording data generation module is used to traverse the plurality of static constraints, perform sample random configuration of temperature elements, generate a plurality of groups of temperature element recording data and a plurality of groups of failure monitoring time recording data; a prediction network configuration module is used to configure a plurality of failure time prediction networks according to the plurality of groups of temperature element recording data and the plurality of groups of failure monitoring time recording data; a network activation module is used to collect the temperature element monitoring characteristic value of the first AGM partition, selectively activate the plurality of failure time prediction networks, and obtain the failure prediction time of the AGM partition, wherein the first AGM partition belongs to the AGM partition simulation model set; an abnormality identification module is used to identify the thermal stability abnormality of the first AGM partition according to the temperature element monitoring characteristic value when the failure prediction time of the AGM partition is less than or equal to the failure time threshold.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By receiving an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, an pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set and a tensile strength label set; performing multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models; traversing the multiple groups of AGM separator simulation models, extracting multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and constructing multiple static constraints; traversing the multiple static constraints , randomly configure samples of temperature elements, generate multiple groups of temperature element record data and multiple groups of failure monitoring time record data; configure multiple failure time prediction networks according to the multiple groups of temperature element record data and the multiple groups of failure monitoring time record data; collect the temperature element monitoring characteristic value of the first AGM separator, selectively activate the multiple failure time prediction networks, and obtain the AGM separator failure prediction time, wherein the first AGM separator belongs to the AGM separator simulation model set; when the AGM separator failure prediction time is less than or equal to the failure time threshold, the first AGM separator is marked as abnormal in thermal stability according to the temperature element monitoring characteristic value, which effectively solves the technical problem that the prior art is difficult to accurately predict and monitor the thermal stability of the AGM separator under different working conditions, resulting in poor management effect of the thermal stability of the AGM separator. It can achieve effective management and monitoring of the thermal stability of the AGM separator, thereby improving the reliability and safety of the battery system.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application 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 exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 This is a schematic diagram of the process of the AGM separator thermal stability monitoring method of this application;

[0013] Figure 2 This is a schematic diagram of the structure of the AGM separator thermal stability monitoring system of this application.

[0014] Description of reference numerals:

[0015] Collection receiving module 11, multi-level clustering module 12, constraint condition building module 13, record data generating module 14, prediction network configuration module 15, network activation module 16, anomaly identification module 17. DETAILED DESCRIPTION

[0016] This application provides a method and system for monitoring the thermal stability of AGM separators, which solves the technical problem that it is difficult to accurately predict and monitor the thermal stability of AGM separators under different working conditions in the prior art, resulting in poor management of the thermal stability of AGM separators. Effective management and monitoring of the thermal stability of AGM separators can be achieved, thereby improving the reliability and safety of the battery system.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] For example, please refer to the attached Figure 1 The present application provides a method for monitoring the thermal stability of an AGM separator, wherein the method for monitoring the thermal stability of an AGM separator is applied to an AGM separator thermal stability monitoring system, and the method for monitoring the thermal stability of an AGM separator specifically comprises the following steps:

[0019] S1: Receive an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, a pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set, and a tensile strength label set.

[0020] Specifically, the AGM separator simulation model set is a set containing multiple simulation models, each of which represents a different AGM separator design or property. The thickness label set represents the thickness of the AGM separator in each simulation model; the pore size label set represents the average diameter of the pores in the separator; the porosity label set represents the volume percentage of the pores in the separator; the material composition ratio label set represents the ratio of different materials in the separator, such as glass fiber, adhesive, etc.; the wet pressure label set represents the pressure characteristics of the separator in a hygroscopic state; and the tensile strength label set represents the strength of the separator when subjected to tensile force.

[0021] S2: Perform multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models.

[0022] Specifically, since different feature labels may have different dimensions and numerical ranges, the data is standardized or normalized to ensure that each feature has the same weight in the clustering process. Check and handle missing data by interpolating, deleting, or filling with default values; identify and handle outliers. Hierarchical clustering can be used to build a cluster tree to achieve multi-level grouping; each AGM partition simulation model is regarded as a separate cluster. Calculate the similarity between each pair of clusters, using a distance metric such as Euclidean distance; merge the most similar clusters into a larger cluster based on the similarity. Repeat the merging step until the preset number of clusters or similarity threshold is reached; record the merged clusters and the corresponding similarities. Based on the structure of the clustering tree, the final cluster is obtained. This can be achieved by setting a distance threshold or by evaluating clustering effectiveness indicators such as the silhouette coefficient. Through this process, multiple groups of AGM partition simulation models can be obtained.

[0023] S3: Traverse the multiple groups of AGM partition simulation models, extract multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and construct multiple static constraint conditions.

[0024] Specifically, each cluster contains a group of AGM separator simulation models with similar characteristics. For each cluster, calculate its centroid, which is the average of all model features in the cluster. For each cluster, calculate the average of all model thicknesses to obtain a centroid thickness label. Calculate the average of all model pore sizes in each cluster to obtain a centroid pore size label; calculate the average of all model porosity in each cluster to obtain a centroid porosity label; for material component ratios, calculate the average ratio of each material component to obtain a set of centroid material component ratio labels; calculate the average of all model wet pressures in each cluster to obtain a centroid wet pressure label. Calculate the average of all model tensile strengths in each cluster to obtain a centroid tensile strength label. The extracted centroid labels are used as static constraints, which will be used for subsequent failure prediction and model verification. For each centroid label, define a range or threshold, for example, the range of the centroid thickness label is [0.2mm, 0.3mm]. Combine the constraints of all centroid labels into a set, and each cluster corresponds to a constraint set.

[0025] S4: traversing the plurality of static constraint conditions, performing random sample configuration on the temperature elements, and generating a plurality of groups of temperature element record data and a plurality of groups of failure monitoring duration record data.

[0026] Specifically, define the possible range of temperature elements, which can be based on the temperature changes in the actual working environment of the AGM separator. For example, the temperature range may be from -20°C to 60°C. For each static constraint, randomly generate a certain number of temperature samples. Create a list or array to store the temperature element record data and the failure monitoring time record data. Use a random number generator to generate temperature samples within the defined temperature range; a uniform distribution or a normal distribution can be used to simulate different temperature conditions. Store the generated temperature samples in the temperature element record data structure. Define the possible range of failure monitoring time based on the expected life and working conditions of the AGM separator. For each temperature sample, randomly generate a failure monitoring time. The failure monitoring time can be generated based on historical data or empirical distribution. Associate the generated failure monitoring time with the corresponding temperature sample and store it in the failure monitoring time record data structure.

[0027] S5: configuring a plurality of failure time prediction networks according to the plurality of groups of temperature element record data and the plurality of groups of failure monitoring time record data.

[0028] Specifically, select one or more machine learning models suitable for regression tasks; including linear regression, decision tree regression, random forest regression, gradient boosting machine, neural network; before training the model, preprocess the data to ensure that all features are on the same scale, and standardization or normalization can be used. Divide the data into training sets and test sets to evaluate the generalization ability of the model. Use the training set data to train the selected model, use the training data to train the model, and use techniques such as cross-validation to find the optimal model parameters. Use the test set data to verify the performance of the model, and use indicators such as mean square error, root mean square error, R² score, etc. to evaluate the performance of the model. Select the model that performs best in the verification phase and save the trained model to a file or database.

[0029] S6: collecting the temperature factor monitoring characteristic value of the first AGM separator, selectively activating the multiple failure time prediction networks, and obtaining the predicted failure time of the AGM separator, wherein the first AGM separator belongs to the AGM separator simulation model set.

[0030] Specifically, the temperature elements of the AGM separator are collected in real time according to the temperature sensor or other monitoring equipment; the temperature data provided by the sensor is recorded, and these data will be used as input features for the failure time prediction network; abnormal or invalid data points are removed; and the trained and saved failure time prediction network model is loaded. According to the characteristics of the first AGM separator, such as thickness, pore size, etc., the pre-processed temperature element monitoring characteristic values ​​are input into the prediction network. The prediction operation is performed to obtain the failure prediction time of the AGM separator.

[0031] S7: When the predicted failure duration of the AGM separator is less than or equal to a failure duration threshold, the first AGM separator is marked as having abnormal thermal stability according to the temperature element monitoring characteristic value.

[0032] Specifically, a failure time threshold is defined, which is based on the design life of the AGM separator, historical failure data or other engineering standards, and the predicted failure time is compared with the preset failure time threshold. If the predicted time is less than or equal to the threshold, a thermal stability abnormality event is recorded in the monitoring system. A thermal stability abnormality flag is generated for the first AGM separator, which can be a warning sign, an alarm code or any other form of flag.

[0033] Furthermore, step S2 of the present application also includes:

[0034] Obtain thickness label threshold, pore size label threshold, porosity label threshold, material composition ratio label deviation threshold, wet pressure label threshold and tensile strength label threshold; according to the i-th attribute label threshold, based on the i-th attribute label set, traverse the i-1 level clustering results of the AGM diaphragm simulation model to perform i-level clustering analysis to obtain the i-level clustering results of the AGM diaphragm simulation model; wherein the i-th attribute label threshold belongs to the thickness label threshold, the pore size label threshold, the porosity label threshold, the material composition ratio label deviation threshold, the wet pressure label threshold and the tensile strength label threshold, the i-th attribute label The label set belongs to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set; when i is equal to the total number of label attributes, the i-level clustering result of the AGM partition simulation model is set to the multiple groups of AGM partition simulation models; when i is less than the total number of label attributes, according to the i+1th attribute label threshold and based on the i+1th attribute label set, the i-level clustering result of the AGM partition simulation model is traversed to perform i+1-level clustering analysis to obtain the i+1-level clustering result of the AGM partition simulation model, and the initial value of i is equal to 1.

[0035] Specifically, according to the performance requirements and design standards of the AGM separator, a threshold is set for each attribute label, thickness, pore size, porosity, material composition ratio, wet pressure and tensile strength. These thresholds can be set by a technician in this field based on historical data. Select an initial clustering algorithm, such as K-means or hierarchical clustering, as the basis for the first-level clustering. Perform initial clustering on the AGM separator simulation model set to obtain the first-level clustering results. Traverse each attribute label, thickness, pore size, porosity, material composition ratio, wet pressure and tensile strength. For each attribute label, use its corresponding threshold to guide the clustering analysis. Based on the threshold of the current attribute label, perform i-level clustering analysis on the AGM separator simulation model. Merge the clustering results of the current attribute label with the previous results to form a new clustering result. When i is less than the total number of label attributes, use the threshold of the next attribute label to perform i+1-level clustering analysis on the current clustering result. Add the i+1-level clustering results to the model set. When i is equal to the total number of label attributes, use the i-level clustering results as the final multiple groups of AGM separator simulation models. At this point, the model collection includes multiple AGM separator simulation models obtained through multi-level clustering analysis.

[0036] Furthermore, step S3 of the present application also includes:

[0037] According to the multiple groups of AGM partition simulation models, a first group of thickness label sets of the first group of AGM partition simulation models are extracted; the first group of thickness label sets are normalized and distributed on the number axis to obtain a one-dimensional distribution coordinate set; according to the one-dimensional distribution coordinate set, K-means clustering analysis is performed on the first group of thickness label sets to obtain K initial centroid coordinates; the K initial centroid coordinates are traversed, and the one-dimensional distribution coordinate set is traversed to perform distance mean analysis to obtain K outlier indices; the initial centroid coordinate with the minimum value of the K outlier indices is taken from the K initial centroid coordinates, and a first centroid thickness label is set, which is added to the multiple centroid thickness labels.

[0038] Specifically, a first group of models is selected from multiple groups of AGM diaphragm simulation models, and a set of thickness labels is extracted from the first group of models. The maximum and minimum values ​​of the data are determined to create a normalized number axis, and each thickness label is converted into a normalized value relative to the data range. According to the normalized thickness labels, a one-dimensional number axis is created, and each normalized value is mapped to a point on the number axis to form a one-dimensional distribution coordinate set. K points are randomly selected as the initial centroids, the distance between each point and all centroids is calculated, each point is assigned to the cluster where the nearest centroid is located, the centroid of each cluster is recalculated, and the above steps are repeated until the centroid position converges. A distance metric, such as Euclidean distance, is selected to calculate the distance between each point and the centroid. For each centroid, the mean distance of all its member points is calculated, and for each centroid, the ratio of the distance between each member point and the centroid to the mean distance is calculated to obtain the outlier index. For each initial centroid, its outlier index is calculated, and the centroid with the smallest outlier index is selected from all centroids. Extract the coordinates of the selected centroid and use the extracted coordinates as the first centroid thickness label. Create a list to store all centroid thickness labels and add the first centroid thickness label to the list.

[0039] Furthermore, step S4 of the present application also includes:

[0040] According to the multiple static constraints, extract the first static constraint; use the first static constraint to limit the AGM partition, retrieve the partition test temperature and the partition failure record duration, and obtain a number of partition test temperature record data and a number of groups of partition failure record durations; obtain the first partition test temperature record data of the several partition test temperature record data; based on the first partition test temperature record data, extract the first group of partition failure record durations from the several groups of partition failure record durations; count the mode value of the first group of partition failure record durations and set it as the first partition failure record duration; add the first partition test temperature record data to the first group of temperature element record data, and add the first partition failure record duration to the first group of temperature element record data; add the first group of temperature element record data to the multiple groups of temperature element record data, and add the first group of temperature element record data to the multiple groups of failure monitoring duration record data.

[0041] Specifically, the first static constraint condition is selected from multiple static constraints as a limiting condition for subsequent operations. Using the first static constraint condition as a limiting condition, the partition test temperature and partition failure record duration that meet the conditions are retrieved in the database or test record. The retrieval includes a query on the partition performance test history to find the test data that meets the first static constraint condition. After the retrieval is completed, one or more groups of partition test temperature record data and one or more groups of partition failure record durations are obtained, and these data should meet the first static constraint condition. The first partition test temperature record data is selected from the retrieved partition test temperature record data as the starting point for subsequent analysis. Using the first partition test temperature record data, the first group of partition failure record durations are extracted from the retrieved partition failure record durations, and these records should correspond to the first partition test temperature record data. Statistical analysis is performed on the first group of partition failure record durations to find the mode value, that is, the failure record duration with the most occurrences. The mode value of the first group of partition failure record durations is set as the first partition failure record duration as the basis for subsequent analysis and prediction. Add the first partition test temperature record data and the first partition failure record duration to the corresponding temperature element record data to ensure that each set of record data is associated with a specific temperature condition. Add the temperature element record data containing the first partition test temperature record data and the first partition failure record duration to multiple sets of temperature element record data for subsequent analysis and use. Add the first set of temperature element record data containing the first partition test temperature record data and the first partition failure record duration to multiple sets of failure monitoring duration record data for subsequent analysis and use.

[0042] Furthermore, the present application also includes:

[0043] Add the first partition test temperature record data into the partition test temperature record data set; when the data volume of the partition test temperature record data set meets the preset data volume, analyze the discrete series of the partition test temperature record data set, wherein the discrete series is the mean of the centroid distance between clusters after cluster analysis of the partition test temperature record data set; when the discrete series is less than or equal to the discrete series threshold, update the partition test temperature record data set; when the discrete series is greater than the discrete series threshold, set the partition test temperature record data set to the first group of temperature element record data.

[0044] Specifically, the extracted first partition test temperature record data is added to the existing partition test temperature record data set. When the data volume of the partition test temperature record data set reaches the preset data volume, its discrete series is analyzed. The discrete series is measured by the mean of the centroid distance between clusters after cluster analysis, which reflects the degree of difference in the performance of the partition under different temperature conditions in the data set. If the discrete series is less than or equal to the preset discrete series threshold, it means that the temperature conditions in the data set are relatively similar and no further update is required. The partition test temperature record data set is kept unchanged and the partition performance under other temperature conditions is continued to be monitored. If the discrete series is greater than the preset threshold, it means that there are significant differences in the temperature conditions in the data set, and the partition test temperature record data set needs to be updated. The partition test temperature record data set is updated to a set containing the latest data so as to more accurately reflect the performance of the partition under different temperature conditions. If the discrete series is greater than the discrete series threshold, the updated partition test temperature record data set is set as the first set of temperature element record data. This set of data will be used for subsequent analysis and prediction, providing a basis for the performance evaluation of AGM partitions under different temperature conditions.

[0045] Furthermore, step S5 of the present application also includes:

[0046] The real-time temperature information of the electrolyte after heat exchange with the first AGM separator is collected through the internal temperature sensor; the charge and discharge status information of the battery deployed with the AGM separator is obtained; the real-time temperature information of the environment is collected through the external temperature sensor; the AGM separator temperature is predicted based on the real-time temperature information of the environment, the charge and discharge status information and the real-time temperature information of the electrolyte, and the temperature element monitoring characteristic value is generated.

[0047] Specifically, a temperature sensor is installed inside the AGM partition to monitor the electrolyte temperature after heat exchange with the partition in real time. The charge and discharge status information of the battery deployed with the AGM partition is obtained through the battery management system or other monitoring equipment. This information includes the current voltage, current, charge and discharge status of the battery, etc., which is crucial for predicting the temperature change of the AGM partition. A temperature sensor is installed outside the battery to monitor the ambient temperature in real time. The ambient temperature has an important impact on the performance and temperature change of the AGM partition, so it needs to be accurately monitored. Combined with the real-time temperature information of the environment, the charge and discharge status information and the real-time temperature information of the electrolyte, the AGM partition temperature is predicted; machine learning, deep learning or physical models can be used to accurately predict the temperature change trend of the partition. After the prediction is completed, a temperature element monitoring characteristic value containing the ambient temperature, charge and discharge status and electrolyte temperature information is generated. These characteristic values ​​can serve as the basis for subsequent analysis and decision-making, helping users better understand the temperature behavior of the AGM partition.

[0048] Furthermore, the present application also includes:

[0049] The electrolyte real-time temperature information includes electrolyte real-time temperature information at multiple locations; the mean temperature of the electrolyte real-time temperature information at the multiple locations is calculated to generate electrolyte characteristic temperature information; based on the battery model information, the historical AGM partition temperature is backtracked for the environmental real-time temperature information, the charge and discharge status information and the electrolyte characteristic temperature information to obtain an AGM partition temperature backtracking temperature set; the mode value of the AGM partition temperature backtracking temperature set is counted and set as the temperature element monitoring characteristic value.

[0050] Specifically, multiple temperature sensors are installed in the AGM partition to monitor the temperature of the electrolyte at different locations in real time. These sensors can be distributed in different areas of the partition to ensure the comprehensiveness and accuracy of the temperature data. The real-time temperature data of the electrolyte collected from multiple locations are processed; the average value of these temperatures is calculated to obtain the characteristic temperature information of the electrolyte, which is a comprehensive indicator of the electrolyte temperature. Combined with the battery model information, the real-time temperature information of the environment, the charge and discharge status information, and the characteristic temperature information of the electrolyte are retrospectively analyzed for historical data. The purpose of the retrospective analysis is to understand the temperature performance of the AGM partition under different historical conditions and provide a reference for the current temperature prediction. Through the analysis of historical data, a set containing the temperature of the AGM partition under different historical conditions is obtained; it reflects the temperature change trend of the AGM partition under different environmental conditions and battery conditions. The temperature values ​​in the retrospective temperature set of the AGM partition temperature are statistically analyzed; the temperature value with the most occurrences, that is, the mode value, is found. This value represents the most common temperature of the AGM partition under similar conditions. The mode value of the AGM partition temperature backtracking temperature set is set as the temperature element monitoring characteristic value; this characteristic value can be used as a reference point for current temperature prediction and monitoring to help users better understand the temperature behavior of the AGM partition.

[0051] In summary, the AGM separator thermal stability monitoring method provided in this application has the following technical effects:

[0052] By receiving an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, an pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set and a tensile strength label set; performing multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models; traversing the multiple groups of AGM separator simulation models, extracting multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and constructing multiple static constraints; traversing the multiple static constraints , randomly configure samples of temperature elements, generate multiple groups of temperature element record data and multiple groups of failure monitoring time record data; configure multiple failure time prediction networks according to the multiple groups of temperature element record data and the multiple groups of failure monitoring time record data; collect the temperature element monitoring characteristic value of the first AGM separator, selectively activate the multiple failure time prediction networks, and obtain the AGM separator failure prediction time, wherein the first AGM separator belongs to the AGM separator simulation model set; when the AGM separator failure prediction time is less than or equal to the failure time threshold, the first AGM separator is marked as abnormal in thermal stability according to the temperature element monitoring characteristic value, which effectively solves the technical problem that the prior art is difficult to accurately predict and monitor the thermal stability of the AGM separator under different working conditions, resulting in poor management effect of the thermal stability of the AGM separator. It can achieve effective management and monitoring of the thermal stability of the AGM separator, thereby improving the reliability and safety of the battery system.

[0053] Embodiment 2: Based on the AGM separator thermal stability monitoring method in the above embodiment, the present application also provides an AGM separator thermal stability monitoring system, please refer to the attached Figure 2 , the AGM separator thermal stability monitoring system comprises:

[0054] The set receiving module 11 is used to receive an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, a pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set and a tensile strength label set.

[0055] The multi-level clustering module 12 is used to perform multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models.

[0056] The constraint condition construction module 13 is used to traverse the multiple groups of AGM partition simulation models, extract multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and construct multiple static constraint conditions.

[0057] The record data generation module 14 is used to traverse the multiple static constraint conditions, perform random sample configuration on the temperature elements, and generate multiple groups of temperature element record data and multiple groups of failure monitoring duration record data.

[0058] The prediction network configuration module 15 is used to configure a plurality of failure time prediction networks according to the plurality of groups of temperature element record data and the plurality of groups of failure monitoring time record data.

[0059] The network activation module 16 is used to collect the temperature factor monitoring characteristic value of the first AGM partition, selectively activate the multiple failure time prediction networks, and obtain the AGM partition failure prediction time, wherein the first AGM partition belongs to the AGM partition simulation model set.

[0060] The abnormality identification module 17 is used to identify the thermal stability abnormality of the first AGM separator according to the temperature element monitoring characteristic value when the predicted failure time of the AGM separator is less than or equal to the failure time threshold.

[0061] Furthermore, the multi-level clustering module 12 in the AGM separator thermal stability monitoring system is also used for:

[0062] Obtain thickness label threshold, pore size label threshold, porosity label threshold, material composition ratio label deviation threshold, wet pressure label threshold and tensile strength label threshold; according to the i-th attribute label threshold, based on the i-th attribute label set, traverse the i-1 level clustering results of the AGM diaphragm simulation model to perform i-level clustering analysis to obtain the i-level clustering results of the AGM diaphragm simulation model; wherein the i-th attribute label threshold belongs to the thickness label threshold, the pore size label threshold, the porosity label threshold, the material composition ratio label deviation threshold, the wet pressure label threshold and the tensile strength label threshold, the i-th attribute label The label set belongs to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set; when i is equal to the total number of label attributes, the i-level clustering result of the AGM partition simulation model is set to the multiple groups of AGM partition simulation models; when i is less than the total number of label attributes, according to the i+1th attribute label threshold and based on the i+1th attribute label set, the i-level clustering result of the AGM partition simulation model is traversed to perform i+1-level clustering analysis to obtain the i+1-level clustering result of the AGM partition simulation model, and the initial value of i is equal to 1.

[0063] Furthermore, the constraint condition building module 13 in the AGM separator thermal stability monitoring system is also used for:

[0064] According to the multiple groups of AGM partition simulation models, a first group of thickness label sets of the first group of AGM partition simulation models are extracted; the first group of thickness label sets are normalized and distributed on the number axis to obtain a one-dimensional distribution coordinate set; according to the one-dimensional distribution coordinate set, K-means clustering analysis is performed on the first group of thickness label sets to obtain K initial centroid coordinates; the K initial centroid coordinates are traversed, and the one-dimensional distribution coordinate set is traversed to perform distance mean analysis to obtain K outlier indices; the initial centroid coordinate with the minimum value of the K outlier indices is taken from the K initial centroid coordinates, and a first centroid thickness label is set, which is added to the multiple centroid thickness labels.

[0065] Furthermore, the recording data generating module 14 in the AGM separator thermal stability monitoring system is also used for:

[0066] According to the multiple static constraints, a first static constraint is extracted; with the first static constraint as the AGM partition limit, the partition test temperature and the partition failure record duration are retrieved to obtain a plurality of partition test temperature record data and a plurality of groups of partition failure record durations; the first partition test temperature record data of the plurality of partition test temperature record data is obtained; based on the first partition test temperature record data, a first group of partition failure record durations is extracted from the plurality of groups of partition failure record durations; the mode value of the first group of partition failure record durations is counted and set as the first partition failure record duration; the first partition test temperature record data is added to the first group of temperature element record data, and the first partition failure record duration is added to the first group of temperature element record data; the first group of temperature element record data is added to the plurality of groups of temperature element record data, and the first group of temperature element record data is added to the plurality of groups of failure monitoring duration record data.

[0067] Furthermore, the AGM separator thermal stability monitoring system further includes a collection update module, which is used to:

[0068] Add the first partition test temperature record data into the partition test temperature record data set; when the data volume of the partition test temperature record data set meets the preset data volume, analyze the discrete series of the partition test temperature record data set, wherein the discrete series is the mean of the centroid distance between clusters after cluster analysis of the partition test temperature record data set; when the discrete series is less than or equal to the discrete series threshold, update the partition test temperature record data set; when the discrete series is greater than the discrete series threshold, set the partition test temperature record data set to the first group of temperature element record data.

[0069] Further, the network activation module 16 in the AGM separator thermal stability monitoring system is also used for:

[0070] The real-time temperature information of the electrolyte after heat exchange with the first AGM separator is collected through the internal temperature sensor; the charge and discharge status information of the battery deployed with the AGM separator is obtained; the real-time temperature information of the environment is collected through the external temperature sensor; the AGM separator temperature is predicted based on the real-time temperature information of the environment, the charge and discharge status information and the real-time temperature information of the electrolyte, and the temperature element monitoring characteristic value is generated.

[0071] Furthermore, the AGM separator thermal stability monitoring system further includes a monitoring characteristic value acquisition module, which is used to:

[0072] The electrolyte real-time temperature information includes electrolyte real-time temperature information at multiple locations; the mean temperature of the electrolyte real-time temperature information at the multiple locations is calculated to generate electrolyte characteristic temperature information; based on the battery model information, the historical AGM partition temperature is backtracked for the environmental real-time temperature information, the charge and discharge status information and the electrolyte characteristic temperature information to obtain an AGM partition temperature backtracking temperature set; the mode value of the AGM partition temperature backtracking temperature set is counted and set as the temperature element monitoring characteristic value.

[0073] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The AGM separator thermal stability monitoring method and specific examples in the first embodiment are also applicable to the AGM separator thermal stability monitoring system of the present embodiment. Through the above detailed description of the AGM separator thermal stability monitoring method, those skilled in the art can clearly understand the AGM separator thermal stability monitoring system of the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0074] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0075] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. AGM separator thermal stability monitoring method, characterized in that: include: Receive an AGM separator simulation model set, wherein the AGM separator simulation model set has a thickness label set, a pore size label set, a porosity label set, a material composition ratio label set, a wet pressure label set, and a tensile strength label set; Perform multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set, and the tensile strength label set to obtain multiple groups of AGM separator simulation models; Traversing the multiple groups of AGM separator simulation models, extracting multiple centroid thickness labels, multiple centroid pore diameter labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and constructing multiple static constraint conditions; Traversing the multiple static constraint conditions, performing random sample configuration on the temperature elements, and generating multiple groups of temperature element record data and multiple groups of failure monitoring duration record data; According to the multiple groups of temperature element record data and the multiple groups of failure monitoring duration record data, multiple failure duration prediction networks are configured; Collecting a temperature element monitoring characteristic value of a first AGM separator, selectively activating the plurality of failure time prediction networks, and obtaining an AGM separator failure prediction time, wherein the first AGM separator belongs to the AGM separator simulation model set; When the predicted failure duration of the AGM separator is less than or equal to a failure duration threshold, the first AGM separator is marked as having abnormal thermal stability according to the temperature element monitoring characteristic value.

2. The method for monitoring thermal stability of AGM separator according to claim 1, characterized in that: The AGM separator simulation model set is subjected to multi-level clustering according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set, and the tensile strength label set to obtain multiple groups of AGM separator simulation models, including: Obtain thickness label threshold, pore size label threshold, porosity label threshold, material composition ratio label deviation threshold, wet pressure label threshold and tensile strength label threshold; According to the i-th attribute label threshold, based on the i-th attribute label set, traverse the i-1 level clustering results of the AGM partition simulation model to perform i-level clustering analysis to obtain the i-level clustering results of the AGM partition simulation model; Wherein, the i-th attribute label threshold value belongs to the thickness label threshold value, the pore size label threshold value, the porosity label threshold value, the material composition ratio label deviation threshold value, the wet pressure label threshold value and the tensile strength label threshold value, and the i-th attribute label set belongs to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set; When i is equal to the total number of label attributes, the i-level clustering result of the AGM separator simulation model is set as the multiple groups of AGM separator simulation models; When i is less than the total number of label attributes, according to the i+1th attribute label threshold and based on the i+1th attribute label set, the i+1th level clustering results of the AGM partition simulation model are traversed to perform i+1th level clustering analysis to obtain the i+1th level clustering results of the AGM partition simulation model, and the initial value of i is equal to 1.

3. The method for monitoring thermal stability of AGM separator according to claim 1, characterized in that: Traversing the multiple groups of AGM separator simulation models, extracting multiple centroid thickness labels, multiple centroid pore diameter labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and constructing multiple static constraint conditions, including: Extracting a first set of thickness label sets of a first set of AGM separator simulation models according to the plurality of sets of AGM separator simulation models; Normalizing the first set of thickness label sets and distributing them on a numerical axis to obtain a one-dimensional distribution coordinate set; According to the one-dimensional distribution coordinate set, performing K-means cluster analysis on the first set of thickness label sets to obtain K initial centroid coordinates; Traversing the K initial centroid coordinates, traversing the one-dimensional distribution coordinate set to perform distance mean analysis, and obtaining K outlier indices; The initial centroid coordinate of the minimum value of the K outlier indices is taken from the K initial centroid coordinates, a first centroid thickness label is set, and the label is added to the plurality of centroid thickness labels.

4. The method for monitoring thermal stability of AGM separator according to claim 1, characterized in that: Traversing the multiple static constraints, performing random sample configuration on the temperature elements, and generating multiple groups of temperature element record data and multiple groups of failure monitoring duration record data, including: Extracting a first static constraint condition according to the plurality of static constraint conditions; The first static constraint condition is limited to the AGM separator, the separator test temperature and the separator failure record duration are retrieved, and several separator test temperature record data and several groups of separator failure record durations are obtained; Obtaining first baffle test temperature record data of the plurality of baffle test temperature record data; Extracting the first group of partition failure record durations from the plurality of groups of partition failure record durations based on the first partition test temperature record data; Counting the mode value of the first group of partition failure record durations, setting it as the first partition failure record duration; Add the first partition test temperature record data to the first group of temperature element record data, and add the first partition failure record duration to the first group of temperature element record data; The first set of temperature element record data is added into the multiple sets of temperature element record data, and the first set of temperature element record data is added into the multiple sets of failure monitoring duration record data.

5. The method for monitoring thermal stability of AGM separator according to claim 4, characterized in that: Adding the first partition test temperature record data to the first group of temperature element record data, and adding the first partition failure record duration to the first group of temperature element record data, includes: Adding the first partition test temperature record data into the partition test temperature record data set; When the data volume of the partition test temperature record data set meets the preset data volume, analyzing the discrete series of the partition test temperature record data set, wherein the discrete series is the mean value of the centroid distance between clusters after cluster analysis of the partition test temperature record data set; When the discrete level is less than or equal to the discrete level threshold, updating the partition test temperature record data set; When the discrete level is greater than the discrete level threshold, the partition test temperature record data set is set as the first set of temperature element record data.

6. The method for monitoring thermal stability of AGM separator according to claim 1, characterized in that: Collect the temperature element monitoring characteristic values ​​of the first AGM separator, including: The real-time temperature information of the electrolyte after heat exchange with the first AGM separator is collected through the internal temperature sensor; Obtain the charge and discharge status information of the AGM separator deployed battery; Collect real-time temperature information of the environment through external temperature sensors; The AGM separator temperature is predicted based on the real-time environmental temperature information, the charge and discharge state information and the real-time electrolyte temperature information to generate the temperature element monitoring characteristic value.

7. The method for monitoring thermal stability of an AGM separator according to claim 6, characterized in that: The AGM separator temperature is predicted according to the real-time environmental temperature information, the charge-discharge state information and the real-time electrolyte temperature information to generate the temperature element monitoring characteristic value, including: The electrolyte real-time temperature information includes electrolyte real-time temperature information at multiple locations; Calculating the average temperature of the real-time temperature information of the electrolyte at the multiple locations to generate characteristic temperature information of the electrolyte; Based on the battery model information, the historical AGM separator temperature is traced back for the real-time temperature information of the environment, the charge and discharge state information, and the characteristic temperature information of the electrolyte to obtain an AGM separator temperature traceback temperature set; The mode value of the AGM partition temperature tracing temperature set is counted and set as the temperature factor monitoring characteristic value.

8. AGM separator thermal stability monitoring system, characterized in that: The steps for implementing the AGM separator thermal stability monitoring method according to any one of claims 1 to 7, the AGM separator thermal stability monitoring system comprising: A collection receiving module, used for receiving an AGM separator simulation model collection, wherein the AGM separator simulation model collection comprises a thickness label collection, an aperture label collection, a porosity label collection, a material composition ratio label collection, a wet pressure label collection, and a tensile strength label collection; A multi-level clustering module, used for performing multi-level clustering on the AGM separator simulation model set according to the thickness label set, the pore size label set, the porosity label set, the material composition ratio label set, the wet pressure label set and the tensile strength label set to obtain multiple groups of AGM separator simulation models; A constraint condition construction module is used to traverse the multiple groups of AGM separator simulation models, extract multiple centroid thickness labels, multiple centroid pore size labels, multiple centroid porosity labels, multiple groups of centroid material composition ratio labels, multiple centroid wet pressure label sets and multiple centroid tensile strength labels, and construct multiple static constraint conditions; A record data generation module, used to traverse the multiple static constraint conditions, perform random sample configuration on the temperature elements, and generate multiple groups of temperature element record data and multiple groups of failure monitoring duration record data; A prediction network configuration module, configured to configure a plurality of failure time prediction networks according to the plurality of groups of temperature element record data and the plurality of groups of failure monitoring time record data; a network activation module, for collecting temperature element monitoring characteristic values ​​of a first AGM separator, selectively activating the plurality of failure time prediction networks, and obtaining a predicted failure time of the AGM separator, wherein the first AGM separator belongs to the AGM separator simulation model set; The abnormality identification module is used to identify the thermal stability abnormality of the first AGM separator according to the temperature factor monitoring characteristic value when the predicted failure time of the AGM separator is less than or equal to the failure time threshold.

Images