Energy storage battery module health monitoring method and system

Abstract

The invention relates to the technical field of energy storage battery module health monitoring, in particular to an energy storage battery module health monitoring method, which comprises the following steps: S1, applying a preset micro-current pulse sequence to each single battery in an echelon utilization energy storage battery module; s2, collecting voltage response data of each single battery under the excitation of the micro-current pulse sequence, and recording current temperature data of each single battery; s3, extracting a first characteristic parameter set and a second characteristic parameter set from the voltage response data; s4, a pre-established reference model is used, the model combines the characteristic parameters, the battery aging state and the temperature corresponding relation, and the real-time influence of historical damage and the current temperature is separated and quantified; and S5, based on the historical damage contribution and the current temperature instant influence contribution, evaluating the inconsistency of each single battery in the echelon utilization battery module. The method has the advantage of improving the accuracy of health state and inconsistency evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage battery module health monitoring, and in particular to a method for monitoring the health of an energy storage battery module. Background Art

[0002] Energy storage battery modules consist of multiple single cells connected in series and parallel. Due to differences in heat generation and heat dissipation conditions caused by current flowing through internal resistance, the operating temperatures of each cell within the module vary, creating a temperature gradient. Over time, this temperature gradient can exacerbate performance inconsistencies between cells, affecting parameters like capacity and internal resistance, thereby reducing the module's overall usable capacity and cycle life, and potentially posing safety risks.

[0003] Energy storage battery modules typically undergo a transition from high-demand applications (such as electric vehicles) to less demanding ones in second-life scenarios (such as stationary energy storage and communication base station backup). During this transition, detailed historical battery data is often missing or inaccurate, especially critical information such as temperature gradients and charge / discharge strategies. This makes it difficult to accurately assess the degree of damage and inconsistencies in individual cells during the second-life phase.

[0004] Existing monitoring methods typically rely on current, real-time temperature data and battery electrical parameters (such as voltage, current, and estimated internal resistance). These methods use established thermal models or empirical formulas to infer temperature distribution and correct battery parameters. These methods present a particular challenge in end-of-life scenarios: how to distinguish between inherent differences in observed single-cell performance caused by historical temperature gradients and stress factors and immediate changes caused by current operating temperatures. Failure to effectively integrate data from the previous lifecycle can lead to significant deviations in monitoring results, impacting the battery management system's balancing control strategy and the accuracy of remaining service life predictions.

[0005] Therefore, existing technologies are in urgent need of improvement, especially in terms of how to accurately separate the impact of historical accumulated damage and current temperature effects to improve the accuracy and safety of monitoring results. Summary of the Invention

[0006] In order to address the deficiencies of the prior art, the present application provides a method and system for monitoring the health of an energy storage battery module, which has the advantage of being able to accurately separate and quantify the historical cumulative damage of single cells in a second-use battery module and the immediate impact of the current temperature, thereby improving the accuracy of health status and inconsistency assessment.

[0007] The present invention adopts the following technical solutions: The present application provides a method for monitoring the health of an energy storage battery module, the method comprising the following steps: S1: Apply a preset micro-current pulse sequence to each single cell in the cascade utilization energy storage battery module; S2: Collect the voltage response data of each single cell under the stimulation of the micro-current pulse train, and record the current temperature data of each single cell; S3: extracting a first characteristic parameter set and a second characteristic parameter set from the voltage response data, wherein the first characteristic parameter set is used to characterize the historical cumulative damage of the single cell, and the second characteristic parameter set is used to characterize the immediate effect of the current temperature on the electrochemical behavior of the single cell; S4: Use a pre-established reference model that combines characteristic parameters, battery aging state, and temperature correspondence to separate and quantify the immediate impact of historical damage and current temperature; S5: Evaluate the inconsistency of each single cell in the battery module based on the historical damage contribution and the immediate impact contribution of the current temperature.

[0008] Through the above solution, the historical cumulative damage of single cells in the cascade utilization battery module and the immediate impact of the current temperature can be accurately separated and quantified, thereby improving the accuracy of health status and inconsistency assessment.

[0009] To further solve the problem, the present application further proposes that, in step S3, the step of extracting the first characteristic parameter set and the second characteristic parameter set from the voltage response data includes: S31: Obtaining individual characteristic information of a single battery, including at least one of source information, aging path information, or current response characteristic information; S32: Selecting a feature extraction rule for each single battery from a preset feature extraction rule set based on the individual characteristic information; S33: Apply the selected feature extraction rule to extract the first feature parameter set and the second feature parameter set.

[0010] Through the above solution, it is possible to select appropriate feature extraction rules according to the individual characteristic information of the single battery, thereby improving the pertinence and accuracy of feature extraction.

[0011] To improve the solution, the present application further proposes that, in step S32, the step of selecting feature extraction rules includes: S321: Standardizing individual characteristic information of the single battery; S322: Matching the standardized feature information with the applicable scope of the rules in the preset feature extraction rule set to obtain a matching result; S323: Based on the matching result, a suitable feature extraction rule is selected from the feature extraction rule set, and the rule is applied to perform feature extraction.

[0012] Through the above scheme, the selection process of feature extraction rules is further optimized through standardization and matching screening to ensure the applicability of the rules.

[0013] To improve the solution, the present application further proposes that, in step S323, the steps of screening out suitable feature extraction rules include: S3231: Setting priority levels for multiple dimensions in the preset priority selection criteria; S3232: Based on the priority level, apply the criteria of each level in order from high to low to screen multiple candidate feature extraction rules.

[0014] Through the above scheme, the priority criterion is introduced to make the selection of feature extraction rules more refined and intelligent.

[0015] To further solve the problem, the present application further proposes that, in step S4, the reference model includes: Temperature distribution information, which characterizes the temperature difference between a single cell and its surrounding cells; Based on the reference model and temperature distribution information, the historical damage and current temperature impact contributions of the single cell are separated and quantified.

[0016] Through the above scheme, the temperature distribution information is incorporated into the reference model, and the influence of temperature on damage separation is considered more comprehensively.

[0017] To further solve the problem, the present application further proposes that, in step S4, the steps of separating and quantifying historical damage and current temperature include: S41: Determine a coupling relationship between the first characteristic parameter set and the second characteristic parameter set; S42: Construct a compensation function based on the coupling relationship. The compensation function is used to characterize the nonlinear coupling effect of historical damage and current temperature on the electrochemical behavior of the battery. S43: Correcting the first characteristic parameter set or the second characteristic parameter set using the compensation function to obtain a corrected first characteristic parameter set, and / or correcting the second characteristic parameter set using the compensation function to obtain a corrected second characteristic parameter set; S44: Based on the corrected characteristic parameter set and current temperature data, separate and quantify the historical accumulated damage of the single battery and the immediate impact of the current temperature.

[0018] Through the above scheme, by constructing and applying the compensation function, the nonlinear coupling effect between historical damage and current temperature is effectively handled, and the accuracy of separation and quantification is improved.

[0019] To improve the solution, the present application further proposes that, in step S42, the step of constructing a compensation function includes: S421: Obtaining source information or initial manufacturing characteristic information of the single battery; S422: Classify the single battery according to source information or initial manufacturing characteristics to obtain at least one battery category; S423: For each battery category, determining a coupling relationship between a first characteristic parameter set and a second characteristic parameter set within the category; S424: Based on the coupling relationship, a category-specific compensation function is constructed for each battery category to characterize the nonlinear coupling effect between the historical damage and the current temperature of the single battery of this category.

[0020] Through the above scheme, a specific compensation function is constructed according to the battery type, thereby improving the applicability and accuracy of the compensation function.

[0021] To improve the solution, the present application further proposes that, in step S423, the step of determining the coupling relationship includes: S4231: Obtain voltage response data of at least two single battery samples under different temperature conditions; S4232: Determine, under each temperature condition, a central tendency parameter and a dispersion parameter characterizing the coupling characteristics of the battery type based on voltage response data of at least two samples; S4233: Combining the central tendency parameter and the dispersion parameter determined under different temperature conditions to form a category-specific coupling relationship for the battery category.

[0022] Through the above scheme, the coupling relationship of battery categories can be determined more accurately through data analysis under multiple samples and multiple temperature conditions.

[0023] To further solve the problem, the present application also proposes a system for monitoring the health of an energy storage battery module, which applies the above-mentioned method for monitoring the health of an energy storage battery module. The system includes: An excitation module, which is used to apply a micro-current pulse sequence to the single cells in the cascade energy storage battery module; Data acquisition module, the data acquisition module is used to collect voltage response data and temperature data of the battery; Feature extraction module, the feature extraction module extracts the characteristic parameter set from the voltage response data and transmits it to the contribution separation and quantification module for analysis; Contribution separation and quantification module: The contribution separation and quantification module analyzes and separates the battery's historical accumulated damage and the immediate impact of current temperature based on the reference model; The inconsistency assessment module evaluates the health inconsistency of the batteries within the module based on the separation results.

[0024] Through the above solution, a system is provided that can implement the above health monitoring method, which is convenient for practical application.

[0025] To improve the solution, the present application also proposes that the system also includes a compensation module, which is used to adjust the battery evaluation results based on a comprehensive analysis of real-time data and historical data.

[0026] Through the above scheme, the introduction of the compensation module can further optimize and adjust the evaluation results and improve the robustness of the system.

[0027] In summary, the present application provides a method and system for monitoring the health of energy storage battery modules. By applying a microcurrent pulse sequence to single cells and analyzing their voltage responses, characteristic parameters characterizing historical damage and current temperature effects are extracted, and reference models and compensation functions are used to separate and quantify these two contributions, thereby accurately evaluating the inconsistency of batteries within the module. This effectively solves the problem of inaccurate health assessment caused by the lack of historical data of used battery modules and the influence of temperature coupling. It has the advantage of being able to accurately separate and quantify the historical cumulative damage of single cells in used battery modules and the immediate impact of current temperature, thereby improving the accuracy of health status and inconsistency assessment.

[0028] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and illustration and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is the overall flow chart of the present invention; Figure 2 Flowchart of the steps of extracting the first characteristic parameter set and the second characteristic parameter set from the voltage response data in the present invention; Figure 3 A flowchart of the steps for selecting feature extraction rules in the present invention; Figure 4 A flowchart of the steps for screening suitable feature extraction rules in the present invention; Figure 5 A flow chart of the steps of separating and quantifying historical damage and current temperature in the present invention; Figure 6 A flowchart of the steps for constructing a compensation function in the present invention; Figure 7 A flowchart of the steps for determining the coupling relationship in the present invention; Figure 8 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION

[0030] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0031] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0032] This embodiment provides a method for monitoring the health of an energy storage battery module. Figures 1 to 8 shown.

[0033] refer to Figure 1 , a method for monitoring the health of an energy storage battery module, the method comprising the following steps: S1: Apply a preset micro-current pulse sequence to each single cell in the cascade utilization energy storage battery module; S2: Collect the voltage response data of each single cell under the stimulation of the micro-current pulse train, and record the current temperature data of each single cell; S3: extracting a first characteristic parameter set and a second characteristic parameter set from the voltage response data, wherein the first characteristic parameter set is used to characterize the historical cumulative damage of the single cell, and the second characteristic parameter set is used to characterize the immediate effect of the current temperature on the electrochemical behavior of the single cell; S4: Use a pre-established reference model that combines characteristic parameters, battery aging state, and temperature correspondence to separate and quantify the immediate impact of historical damage and current temperature; S5: Evaluate the inconsistency of each single cell in the battery module based on the historical damage contribution and the immediate impact contribution of the current temperature.

[0034] Among them, the microcurrent pulse sequence is used to stimulate the electrochemical response of the battery and extract the battery state characteristics.

[0035] A pre-established reference model is a mathematical or algorithmic model constructed through offline training or historical data analysis. It encodes the relationship between the battery's characteristic parameters, aging state, and temperature. This model can use machine learning, empirical formulas, or electrochemical models. Its purpose is to provide a benchmark for parsing the currently collected characteristic parameters and separating the contributions of historical accumulated damage and the immediate impact of current temperature.

[0036] Based on the above characteristics, the core innovation of this application lies in applying a microcurrent pulse sequence to the second-hand battery and extracting a set of characteristic parameters that distinguish historical damage and current temperature effects, and combining it with a pre-established reference model. In the absence of complete historical data, the historical cumulative damage of the single cell and the immediate impact of the current temperature are separated and quantified, thereby achieving the effect of evaluating the inconsistency of the single cells in the second-hand battery module.

[0037] To achieve these innovations, the system operates as follows: First, a pulse current generator integrated into the battery management system applies a preset microcurrent pulse train to each cell in the second-life energy storage battery module. This non-invasive stimulation method stimulates the battery's electrochemical response, collecting voltage response data for each cell under pulse stimulation and recording the current temperature of each cell. This data reflects the battery's performance under current operating conditions. Next, two key sets of characteristic parameters are extracted from the voltage response data. The first set of characteristic parameters quantifies irreversible performance degradation caused by stress accumulation during long-term service, typically including internal resistance spectrum characteristics, capacity decay rate, and voltage plateau offset. The second set of characteristic parameters characterizes the immediate impact of current temperature on the battery's electrochemical behavior, typically including impedance parameters, voltage response rate, or the difference between steady-state and transient voltages. Based on this, a pre-established reference model is used. This model combines the battery's characteristic parameters, battery aging status, and temperature information to map the immediate impact of historical damage and current temperature on battery performance. This model, which can be developed using machine learning methods, empirical formulas, or electrochemical models, aims to effectively isolate and quantify these effects. Finally, based on the separation and quantification results, the health inconsistencies of each single cell within the used battery module are assessed. This method can identify whether the battery performance inconsistencies are primarily due to historical damage or temperature differences in the current operating environment, providing more accurate and targeted health assessment results.

[0038] Through the above technical solution, this application can solve the problem that it is difficult to accurately evaluate the health status of single cells in the cascade utilization energy storage battery module due to the lack of historical service data. By applying specific micro-current pulse excitation and extracting characteristic parameters that distinguish historical damage and current temperature effects, combined with a pre-established reference model for separation and quantification, it is possible to accurately identify and quantify the historical cumulative damage of each single cell and the immediate impact of the current temperature on its electrochemical behavior in the absence of complete historical data. This makes the assessment of the inconsistency of single cells within the module more accurate, thereby providing a strong basis for the battery management system, helping to formulate balancing strategies and health management measures, improving the performance of the battery module, extending its service life, ensuring operational safety, and giving full play to the residual value of the battery.

[0039] In some of the above-mentioned embodiments of the present application, it is proposed to extract a first characteristic parameter set and a second characteristic parameter set from the voltage response data. The extraction of the first characteristic parameter set and the second characteristic parameter set from the voltage response data can be specifically performed by performing a unified Fourier transform on the collected voltage response data, and extracting its amplitude and phase information within a specific frequency range as characteristic parameters, wherein the characteristic parameters of the low-frequency part can be used to characterize the historical cumulative damage of the single cell battery, and the characteristic parameters of the high-frequency part can be used to characterize the immediate impact of the current temperature on the electrochemical behavior of the single cell battery. In this way, information reflecting the battery status can be obtained from the voltage response data. However, in its implementation process, due to the complex sources of the single cells in the cascade utilization battery module and different aging paths, directly applying a unified feature extraction rule may not accurately reflect the true state of each single cell battery, resulting in deviations in subsequent damage separation and quantification.

[0040] refer to Figure 2 In this regard, the present application further proposes that in step S3, the step of extracting the first characteristic parameter set and the second characteristic parameter set from the voltage response data includes: S31: Obtaining individual characteristic information of a single battery, including at least one of source information, aging path information, or current response characteristic information; S32: Selecting a feature extraction rule for each single battery from a preset feature extraction rule set based on the individual characteristic information; S33: Apply the selected feature extraction rule to extract the first feature parameter set and the second feature parameter set.

[0041] Among them, the individual characteristic information of a single cell refers to data reflecting the unique properties of a single cell, which can be obtained by collecting the battery's manufacturing information, historical usage records or current electrical performance parameters; source information refers to the manufacturing background of the battery, which may include the manufacturer, production batch, model, etc.; aging path information refers to the stress conditions experienced by the battery in historical use, which may include the historical number of charge and discharge cycles, cumulative discharge capacity, temperature range experienced, high and low temperature exposure time, high current charge and discharge times, etc.; current response characteristic information refers to the electrochemical characteristics exhibited by the battery in its current state, which may include the current voltage, internal resistance, capacity decay rate, self-discharge rate, etc.; the preset feature extraction rule set refers to a variety of different feature extraction algorithms or methods prepared in advance, which may include methods based on time domain analysis, frequency domain analysis, equivalent circuit model parameter extraction, machine learning model-based feature extraction methods, etc.; feature extraction rule refers to a specific feature extraction algorithm or method in the set; the first feature parameter set refers to a set of values ​​extracted from the voltage response data that mainly reflects the long-term cumulative damage state of the battery; the second feature parameter set refers to a set of values ​​extracted from the voltage response data that mainly reflects the immediate impact of the current temperature on the battery's electrochemical behavior.

[0042] The solution of the present application obtains the individual characteristic information of the single cell battery, and selects a suitable feature extraction rule for each single cell battery from a preset feature extraction rule set based on this information, and then applies the selected rule to extract the first feature parameter set and the second feature parameter set. This method no longer uses a unified feature extraction rule for all single cells, but instead matches each battery with a feature extraction method that best reflects its true state based on its personalized information. Obtaining individual characteristic information is the basis for personalized feature extraction, selecting rules based on this information is a key step in achieving self-adaptation, and applying the selected rules to extract features is the execution process for obtaining accurate parameters. It is precisely because of this processing method of selecting rules based on individual differences that the extracted characteristic parameters can more accurately reflect the historical cumulative damage of the single cell battery and the immediate impact of the current temperature, thereby providing more reliable data input for subsequent damage separation, quantification, and inconsistency assessment, thereby improving the accuracy and reliability of the entire health monitoring method.

[0043] In some preferred embodiments, the system first obtains individual characteristic information for each battery cell, such as manufacturer and historical cycle count. Based on this information, the system selects appropriate rules from a pre-set feature extraction rule set. For example, Rule 1 applies to batteries from Manufacturer A and extracts features based on the slope of the voltage response curve; Rule 2, for batteries with a high cycle count, uses a wavelet transform to extract band energy features. Based on the battery's characteristics, the system selects an appropriate rule, such as Rule 1, Rule 2, or a combination thereof. The system then processes the battery's voltage response data to extract a first set of characteristic parameters, such as capacity decay slope and internal resistance increase, and a second set of characteristic parameters, such as energy features influenced by temperature.

[0044] Through the above technical solution, feature extraction rules can be adaptively selected according to the individual characteristic information of single cells, so as to more accurately extract characteristic parameters reflecting historical cumulative damage and the immediate impact of current temperature from voltage response data, thereby improving the accuracy of subsequent damage separation, quantification and inconsistency assessment, and enhancing the reliability of health monitoring of battery modules for cascade utilization.

[0045] During implementation, individual battery characteristic information may have different data formats and dimensions. Directly using this information in selecting feature extraction rules can lead to matching errors or inefficiencies. Furthermore, the scopes of application of rules in the pre-set feature extraction rule set may overlap or be missing. Effectively matching individual battery characteristic information with the scopes of application and selecting the most appropriate feature extraction rules is also a challenge that needs to be addressed.

[0046] refer to Figure 3 In this regard, the present application further proposes the steps of selecting feature extraction rules including: S321: Standardizing individual characteristic information of the single battery; S322: Matching the standardized feature information with the applicable scope of the rules in the preset feature extraction rule set to obtain a matching result; S323: Based on the matching result, a suitable feature extraction rule is selected from the feature extraction rule set, and the rule is applied to perform feature extraction.

[0047] Standardizing the individual characteristic information of a single battery cell refers to converting individual characteristic information in different formats and dimensions onto a unified scale, eliminating dimensional differences and making the subsequent matching process more accurate and reliable. Specifically, Z-score standardization, Min-Max scaling, or other suitable standardization methods can be used. This provides standardized input data for subsequent rule matching. Matching the standardized characteristic information with the rule applicability scopes in a preset feature extraction rule set refers to determining whether the standardized individual characteristic information meets the applicability conditions or scopes defined by the preset feature extraction rules. The rule applicability scope can be defined as a specific battery type, a specific aging stage, a specific temperature range, etc. The matching process can be performed by comparing the standardized values ​​with thresholds or intervals defined in the rule applicability scopes. This process aims to determine which rules are likely applicable to the current single battery cell. Based on the matching results, selecting suitable feature extraction rules from the feature extraction rule set refers to selecting one or more rules that best meet current requirements from the list of applicable rules obtained during the matching process. This selection can be based on preset optimization criteria, such as rule accuracy, computational efficiency, or robustness. Its purpose is to determine the final rule used for feature extraction from multiple possible rules.

[0048] This logical process of standardization, matching, and screening makes the selection of feature extraction rules more systematic and intelligent. It can dynamically adjust the feature extraction method based on the specific conditions of individual cells, thereby improving the relevance and accuracy of feature extraction. Compared to simply applying fixed rules or manual selection, this solution can better adapt to the diversity and complexity of individual cells in cascade battery modules, providing more reliable input for subsequent health status assessments.

[0049] In some preferred embodiments, it is assumed that the individual characteristic information of a single battery includes its manufacturer information and historical cycle count. A preset feature extraction rule set includes multiple rules, each with a specific scope of application. Based on the individual characteristic information of the battery, after normalization, the most appropriate feature extraction rule is selected and applied to extract the first and second feature parameter sets.

[0050] During its implementation, it is not enough to simply screen out suitable feature extraction rules, because there may be situations where multiple rules are suitable, which will lead to arbitrary selection and cannot guarantee that the selected rules are optimal, thus affecting the accuracy and reliability of health monitoring.

[0051] refer to Figure 4 In this regard, the present application further proposes the steps of screening out suitable feature extraction rules, including: S3231: Setting priority levels for multiple dimensions in the preset priority selection criteria; S3232: Based on the priority level, apply the criteria of each level in order from high to low to screen multiple candidate feature extraction rules.

[0052] This application scheme sets a priority level for multiple dimensions of the selection criteria, and applies each level of criteria in descending order to screen multiple candidate feature extraction rules. This hierarchical and orderly screening mechanism ensures that when multiple suitable rules exist, the rule with the best performance in key performance dimensions can be objectively selected, maximally adapting to the individual characteristics and current state of a specific single cell, thereby improving the quality and accuracy of feature extraction.

[0053] The multiple dimensions of the optimization criteria can include feature extraction accuracy, computational efficiency, and data integrity requirements, with priority levels set according to importance. For example, accuracy is set as priority 1, computational efficiency is set as priority 2, and data integrity requirements are set as priority 3. When screening candidate rules, the accuracy criterion is first applied to eliminate rules that fall below the accuracy threshold; then, the computational efficiency criterion is applied to eliminate rules with excessively long computation times; finally, the data integrity requirement is applied to select rules that meet the minimum data point requirement. Ultimately, the optimal rule or set of rules is selected from among the rules that pass all levels of screening.

[0054] This improved feature extraction rule selection mechanism, combined with individual feature selection rules and subsequent separation and quantification steps, forms a more refined and robust health monitoring process, effectively addressing the challenges brought about by the lack of historical information on cascade utilization batteries and the complexity of real-time operating conditions, and significantly improving the accuracy and reliability of energy storage battery module health monitoring.

[0055] In some of the above-mentioned embodiments of the present application, it is proposed to use a pre-established reference model, which combines characteristic parameters, battery aging status and temperature correspondence to separate and quantify the immediate impact of historical damage and current temperature. The use of the reference model to separate and quantify the immediate impact of historical damage and current temperature can be specifically by analyzing the first characteristic parameter set and the second characteristic parameter set extracted from the voltage response data, and combining these characteristic parameters with the preset reference model to map these characteristic parameters to the historical cumulative damage of the battery and the immediate impact contribution of the current temperature. For example, the reference model can be a lookup table or mathematical function trained based on a large amount of experimental data, which inputs characteristic parameters and current temperature and outputs historical damage contribution and current temperature impact contribution. In this way, different sources of battery performance degradation can be preliminarily distinguished. However, in its implementation process, relying solely on the reference model, the impact of the temperature difference between the single cell and the surrounding single cells on battery aging and performance is not fully considered, which may cause the separation and quantification results to be inaccurate and unable to fully reflect the true state of the battery in a complex thermal environment.

[0056] In this regard, the present application further proposes that in step S4, the reference model includes: Temperature distribution information, which characterizes the temperature difference between a single cell and its surrounding cells; Based on the reference model and temperature distribution information, the historical damage and current temperature impact contributions of the single cell are separated and quantified.

[0057] The solution of this application more accurately separates and quantifies the historical damage and current temperature impact of single cells by introducing temperature distribution information. Temperature distribution information characterizes the temperature difference between a single cell and the surrounding cells. It is obtained by arranging multiple temperature sensors inside the module, collecting temperature data of single cells at different positions, and calculating parameters such as the temperature difference or temperature gradient between the target cell and the surrounding cells. The reference model combines the characteristic parameters of the battery, the battery aging status, the temperature and the temperature distribution information, and is constructed based on machine learning algorithms or electrochemical mechanisms to analyze and separate the immediate impact of historical damage and current temperature.

[0058] Incorporating temperature distribution information allows the reference model to more comprehensively understand the thermal environment of individual cells. For example, if a historically damaged battery is currently in a low-temperature region, its performance degradation may not be fully apparent; whereas, if a historically healthy battery is in a high-temperature region, its performance may rapidly decline. By incorporating temperature distribution information, the model can compensate for the effects of temperature differences between cells, more accurately separating the effects of historical damage and current temperature on battery performance, significantly improving the accuracy of health assessments.

[0059] In some preferred embodiments, as a specific implementation, temperature sensors, such as NTC thermistors, can be placed on the side or top of each cell within the energy storage battery module. For a cell to be evaluated, in addition to collecting its own temperature data, temperature data for its four immediately adjacent cells is also collected. Temperature distribution information can be calculated as the difference between the temperature of the cell and the average temperature of its four neighboring cells. A reference model can employ a feedforward neural network structure, where the input layer receives a first set of feature parameters, a second set of feature parameters, the current cell temperature, and calculated temperature distribution information. The hidden layer can include multiple layers of nonlinear activation functions. The output layer outputs two values, representing the historical damage contribution and the current temperature contribution. This neural network model can be obtained through offline training, with training data including performance data and known aging states of batteries operating under different temperature gradients. During actual monitoring, the real-time collected feature parameters, current temperature, and temperature distribution information are input into the trained neural network to obtain the separated and quantified historical damage contribution and current temperature contribution.

[0060] This technical solution allows for the separation and quantification of the contributions of historical damage and current temperature effects on individual cells, while fully accounting for temperature differences between individual cells and their surroundings. This enables a more refined and accurate assessment of battery health, more effectively identifying both inherent performance degradation caused by historical cumulative damage and immediate performance fluctuations caused by the current local thermal environment. This improves the accuracy of health monitoring for inconsistent individual cells in end-of-life battery modules.

[0061] During its implementation, the accuracy of separation and quantification is affected because the influence of historical damage and current temperature on the electrochemical behavior of the battery is not a simple linear superposition, but a complex nonlinear coupling effect.

[0062] refer to Figure 5 In this regard, the present application further proposes the steps of separating and quantifying historical damage and current temperature, including: S41: Determine a coupling relationship between the first characteristic parameter set and the second characteristic parameter set; S42: Construct a compensation function based on the coupling relationship. The compensation function is used to characterize the nonlinear coupling effect of historical damage and current temperature on the electrochemical behavior of the battery. S43: Correcting the first characteristic parameter set or the second characteristic parameter set using the compensation function to obtain a corrected first characteristic parameter set, and / or correcting the second characteristic parameter set using the compensation function to obtain a corrected second characteristic parameter set; S44: Based on the corrected characteristic parameter set and current temperature data, separate and quantify the historical accumulated damage of the single battery and the immediate impact of the current temperature.

[0063] The coupling relationship refers to the mutual influence and interaction between the first characteristic parameter set and the second characteristic parameter set, which can be determined by mathematical models, statistical analysis or machine learning methods, and its purpose is to provide a basis for constructing the compensation function; The compensation function refers to a mathematical model or mapping relationship used to simulate and quantify the nonlinear coupling effect of historical damage and current temperature on the electrochemical behavior of the battery. Specifically, it can be constructed using polynomial functions, nonlinear regression models, or neural networks. Its purpose is to eliminate or weaken the influence of nonlinear coupling effects on characteristic parameters. Correction refers to adjusting or modifying the original characteristic parameter set based on the compensation function, with the aim of making the characteristic parameters more accurately reflect the historical damage of the battery and the current temperature impact.

[0064] This application solution constructs a compensation function by determining the coupling relationship between the first and second characteristic parameter sets to characterize the nonlinear coupling effect of historical damage and current temperature on the battery's electrochemical behavior. Because historical damage and current temperature interact and jointly influence the battery's electrochemical behavior, the compensation function effectively simulates this nonlinear coupling effect, thereby more accurately reflecting the battery's actual state.

[0065] The compensation function corrects the characteristic parameter set to eliminate or weaken the impact of nonlinear coupling effects, allowing the characteristic parameters to more accurately reflect the battery's historical damage and the immediate impact of current temperature. Based on the corrected characteristic parameter set and current temperature data, the historical cumulative damage of individual cells and the immediate impact of current temperature can be more accurately separated and quantified, providing a more reliable basis for health assessment of battery modules undergoing cascade utilization.

[0066] In a preferred embodiment, the coupling relationship is established by collecting voltage response data from operating batteries under different historical damage and temperature conditions, extracting first and second characteristic parameter sets, and then establishing a mathematical model between these two parameter sets using nonlinear regression or machine learning algorithms (such as support vector regression or neural networks). Based on this coupling relationship, a compensation function corrects the characteristic parameters using a trained nonlinear model (such as a multilayer perceptron neural network). The corrected characteristic parameter set can be further input into the historical damage assessment model and the temperature impact model, thereby achieving an accurate assessment of the battery's health status.

[0067] Through this technical solution, the present application can effectively handle the nonlinear coupling of historical damage and current temperature on battery behavior, significantly improve the accuracy of battery health monitoring, and provide a more reliable data basis for subsequent inconsistency assessment and management.

[0068] In some of the above-mentioned embodiments of the present application, it is proposed to construct a compensation function based on a coupling relationship to characterize the nonlinear coupling effect of historical damage and current temperature on the electrochemical behavior of the battery. The construction of this compensation function can be specifically by collecting performance data of a batch of battery samples at different temperatures, establishing a unified mathematical model or lookup table, taking characteristic parameters and temperature as input, and outputting a correction factor or corrected characteristic parameters. In this way, the characteristic parameters can be corrected. However, in actual applications, the coupling relationship between historical damage and current temperature may vary for single cells with different sources or initial manufacturing characteristics. If these differences are not taken into account, directly using a unified compensation function for correction may lead to inaccurate correction results, thereby affecting the accuracy of health monitoring.

[0069] refer to Figure 6 In this regard, the present application further proposes that the steps of constructing a compensation function include: S421: Obtaining source information or initial manufacturing characteristic information of the single battery; S422: Classify the single battery according to source information or initial manufacturing characteristics to obtain at least one battery category; S423: For each battery category, determining a coupling relationship between a first characteristic parameter set and a second characteristic parameter set within the category; S424: Based on the coupling relationship, a category-specific compensation function is constructed for each battery category to characterize the nonlinear coupling effect between the historical damage and the current temperature of the single battery of this category.

[0070] Among them, source information or initial manufacturing characteristic information refers to the data of single cells at the initial stage of production or use that can reflect their intrinsic properties or potential behavioral differences. Specifically, it may include the brand, model, production batch, production date, initial capacity, initial internal resistance, electrode material type, electrolyte composition, etc. Its purpose is to provide a basis for the subsequent refined classification of single cells; Battery categories refer to groups of batteries with similar properties or behavior patterns, divided based on the source information or initial manufacturing characteristics of individual batteries. Specifically, individual batteries can be classified into different groups through clustering algorithms, rule matching, or manual designation. The purpose is to group batteries with similar coupling characteristics together for targeted analysis and modeling. The coupling relationship refers to the correlation and interaction between the historical cumulative damage level of a single battery and the impact of the current operating temperature on its electrochemical behavior parameters, such as voltage response and internal resistance. Specifically, this correlation can be quantified through methods such as experimental data analysis, physical model establishment, or data-driven model training. Its purpose is to accurately capture the behavioral patterns of different battery types under different temperatures and aging conditions. A category-specific compensation function refers to a mathematical model or mapping relationship constructed for a specific battery category and used to correct the characteristic parameters of single cells in that category. Specifically, it can be a mathematical formula, a lookup table or a machine learning model. Its input may include a first characteristic parameter set, a second characteristic parameter set and / or the current temperature, and the output is the corrected characteristic parameters. Its purpose is to provide more accurate characteristic parameter correction based on the inherent characteristics of different battery categories, thereby improving the accuracy of health monitoring.

[0071] This application solution captures the inherent differences in batteries by obtaining information about the source or initial manufacturing characteristics of individual cells. Based on this information, individual cells are first classified, with cells with similar characteristics grouped together. This step is crucial because different battery categories may have significantly different coupling relationships between historical damage and current temperature due to differences in materials, processes, or usage history.

[0072] Next, for each battery class, the coupling relationship between the first and second characteristic parameter sets within that class is determined. This eliminates the assumption that all batteries follow the same coupling rules and instead quantifies the unique coupling characteristics of each battery class. Finally, based on the coupling relationship determined for each class, a class-specific compensation function is constructed that accurately characterizes the nonlinear coupling effect between the historical damage and current temperature of the individual cells in that class.

[0073] In practical applications, when the characteristic parameters of a single battery cell need to be corrected, the battery category is first determined, and then a compensation function specific to that category is used for correction. This refined compensation function construction process avoids the errors that might be introduced by a unified compensation function, thereby improving the accuracy of characteristic parameter correction and enhancing the precision of separating historical damage from the immediate effects of current temperature, ultimately enhancing the accuracy of the entire energy storage battery module health monitoring method.

[0074] For example, when constructing a compensation function, batteries can be categorized by comparing their source information and initial manufacturing characteristics. For each battery category, complete historical or experimental data is then used to analyze how the first and second characteristic parameter sets change under different temperature and aging conditions. By building a model, such as a polynomial or neural network model, correction factors are output, ultimately constructing a category-specific compensation function.

[0075] This solution addresses the differences in historical damage and temperature impacts among different batteries through classification and targeted modeling, ensuring high accuracy and reliability of health monitoring.

[0076] During its implementation, this solution aims to accurately determine the coupling relationship between the first and second characteristic parameter sets within each battery type in order to construct a more precise compensation function. If the coupling relationship is inaccurate, the constructed compensation function will not accurately reflect the nonlinear coupling effect of historical damage and current temperature, thus affecting the accuracy of health monitoring.

[0077] refer to Figure 7 In this regard, the present application further proposes that the steps for determining the coupling relationship include: S4231: Obtain voltage response data of at least two single battery samples under different temperature conditions; S4232: Determine, under each temperature condition, a central tendency parameter and a dispersion parameter characterizing the coupling characteristics of the battery type based on voltage response data of at least two samples; S4233: Combining the central tendency parameter and the dispersion parameter determined under different temperature conditions to form a category-specific coupling relationship for the battery category.

[0078] The voltage response data refers to the record of the terminal voltage change over time after the single cell is stimulated by a micro-current pulse sequence, which can be represented by a discrete voltage sampling point sequence or a continuous voltage waveform curve.

[0079] Different temperature conditions refer to multiple preset temperature points or temperature ranges with differences, which can be achieved by controlling the ambient temperature using a constant temperature box or by controlling the temperature of the battery through a heating / cooling device.

[0080] The central tendency parameter refers to a statistic used to describe the trend in a set of data. Specifically, it is the typical value of the voltage response data of the battery category at a specific temperature. It can be determined by the mean, median or mode. Its purpose is to reflect the general electrochemical behavior characteristics of the battery category at the temperature.

[0081] The dispersion parameter refers to a statistic used to describe the degree of dispersion of a set of data. Specifically, it is the degree of variation in the voltage response data of the battery category at a specific temperature. It can be determined by standard deviation, variance, range or interquartile range. Its purpose is to reflect the consistency or difference of the electrochemical behavior of the battery category at this temperature.

[0082] The category-specific coupling relationship refers to the regular characterization of the mutual correlation and influence between the first characteristic parameter set and the second characteristic parameter set within the battery category. Specifically, it is a combination of central trend parameters and dispersion parameters under different temperature conditions. It can be represented by a multidimensional parameter set, a mathematical model or a lookup table. Its purpose is to provide a basis for constructing an accurate compensation function.

[0083] This application solution provides a data basis for analyzing the effect of temperature on the electrochemical behavior of batteries by obtaining voltage response data of at least two single battery samples under different temperature conditions. Under each temperature condition, based on the voltage response data of these samples, the central tendency parameter and dispersion parameter characterizing the coupling characteristics of the battery category are calculated and determined. The central tendency parameter reflects the typical behavior of the battery of this category at this temperature, while the dispersion parameter captures individual differences or variability. By combining the two, the electrochemical response of the battery category at a specific temperature can be more comprehensively and accurately characterized.

[0084] Next, the central tendency and dispersion parameters under different temperature conditions are combined into a class-specific coupling relationship, forming a complete model that can reflect the impact of temperature changes on the battery's electrochemical behavior. This coupling relationship provides the basis for constructing a class-specific compensation function, which is used to accurately characterize the nonlinear coupling effect between the battery's historical damage and the current temperature. In this way, the complex behavior of battery classes under different temperatures and aging conditions can be more accurately reflected, making subsequent correction of characteristic parameters based on the compensation function more precise.

[0085] In a preferred embodiment, at least five single-cell battery samples are first obtained and subjected to the same microcurrent pulse sequence at three temperature points, such as 0°C, 25°C, and 50°C. The voltage response data for each sample at each temperature point are then collected. Next, the central tendency parameter (e.g., mean) and dispersion parameter (e.g., standard deviation) of the voltage response are calculated at each temperature point. Finally, the mean and standard deviation at different temperatures are combined into a six-element vector representing the class-specific coupling relationship for that battery class. This vector can be stored in a lookup table or used to fit a polynomial function.

[0086] This method comprehensively and accurately captures the electrochemical behavior and variability of battery types at different temperatures by combining data under multiple temperature conditions, thereby providing a precise coupling relationship, laying a solid foundation for constructing category-specific compensation functions, improving the accuracy of separating historical damage from current temperature effects, and ultimately enhancing the reliability of health monitoring of battery modules undergoing cascade utilization.

[0087] refer to Figure 8 , this application further proposes a storage battery module health monitoring system, the system comprising: An excitation module, which is used to apply a micro-current pulse sequence to the single cells in the cascade energy storage battery module; Data acquisition module, the data acquisition module is used to collect voltage response data and temperature data of the battery; Feature extraction module, the feature extraction module extracts the characteristic parameter set from the voltage response data and transmits it to the contribution separation and quantification module for analysis; Contribution separation and quantification module: The contribution separation and quantification module analyzes and separates the battery's historical accumulated damage and the immediate impact of current temperature based on the reference model; The inconsistency assessment module evaluates the health inconsistency of the batteries within the module based on the separation results.

[0088] The solution provided in this application effectively monitors and assesses the health status of second-life energy storage battery modules through the collaborative operation of hardware modules. The excitation module actively stimulates the battery's electrochemical response by applying a sequence of microcurrent pulses and acquires dynamic response data. The data acquisition module measures and records the battery's voltage and temperature in real time, providing a foundation for subsequent analysis. The feature extraction module converts raw data into a set of characteristic parameters that characterize the battery's status.

[0089] The core component is the Contribution Separation and Quantification module, which analyzes characteristic parameters based on a preset reference model to separate and quantify the contributions of historical cumulative damage and current temperature effects. This module addresses the lack of historical data for end-of-life batteries, accurately assessing the battery's health and the impact of the current environment under unknown historical operating conditions. Finally, the Inconsistency Assessment module uses this separated data to accurately assess the health differences between individual cells within a module, identifying potential weak cells or risk points.

[0090] Through the combination of hardware and software, the system provides precise control and efficient analysis, overcoming the problems of low monitoring efficiency and poor real-time performance in existing technologies, making it possible to accurately assess the health of batteries with complex histories, thereby providing a reliable basis for battery management and cascade utilization strategies.

[0091] In some preferred embodiments, the energy storage battery module health monitoring system can be implemented as follows: The excitation module utilizes a high-precision, programmable current source circuit to output a microcurrent pulse train with a preset waveform (e.g., square wave, sine wave, or pulse train) and amplitude, which is connected to each cell in the module via a switch matrix. The data acquisition module includes multiple high-speed analog-to-digital converters (ADCs) and temperature sensors (e.g., thermistors) positioned near each cell to collect the battery's voltage response and temperature data, respectively. The feature extraction module, contribution separation and quantification module, and inconsistency assessment module are integrated into an embedded processing unit (e.g., a high-performance microcontroller or digital signal processor). This processing unit receives voltage and temperature data and runs a feature extraction algorithm (e.g., time-domain feature extraction or frequency-domain impedance spectroscopy) to obtain a set of feature parameters. The processing unit then uses a stored reference model and separation and quantification algorithm to analyze the feature parameters and separate the contribution of historical cumulative damage from the contribution of current temperature influence. Finally, the processing unit calculates the health inconsistency index for each cell in the module based on the separation results and outputs or stores the results.

[0092] Through this technical solution, the system provides precise hardware-level stimulation and data collection capabilities, overcoming the limitations of software-only monitoring. By applying specific analytical methods, the system effectively distinguishes and quantifies the historical cumulative damage of end-of-life batteries and the immediate impact of current temperature, enabling accurate assessment of health inconsistencies within individual cells within a module and reliable monitoring even in the absence of complete historical data.

[0093] During its implementation, battery evaluation results may be affected by various factors, such as ambient temperature and battery aging. Relying solely on real-time data for evaluation may lead to deviations in the evaluation results, thus affecting the decision-making of the battery management system.

[0094] In this regard, the present application further proposes an energy storage battery module health monitoring system, which also includes: Compensation module,The compensation module is used to adjust the battery evaluation results based on the comprehensive analysis of real-time data and historical data.

[0095] Among them, the compensation module refers to a functional unit that receives real-time monitoring data and historical operation data of the battery, and based on the analysis of these data, corrects the preliminary evaluation results output by the contribution separation quantification module or the inconsistency evaluation module. It can be implemented by a software algorithm module, a hardware processing unit or a combination of software and hardware. Its purpose is to improve the accuracy of the battery health assessment results.

[0096] This solution introduces a compensation module that adjusts the battery health assessment results by comprehensively analyzing real-time battery data and historical operating data. The compensation module receives preliminary results on historical damage and current temperature impacts from the contribution separation and quantification module, as well as the preliminary assessment from the inconsistency assessment module. It also accesses historical battery data, such as temperature exposure, number of charge and discharge cycles, and health assessment records. Based on this historical data, the compensation module identifies factors that may affect the accuracy of the current assessment. For example, if the battery has been exposed to high temperatures for an extended period, the compensation module will adjust the assessment results appropriately based on known relationships between high temperature and accelerated aging, potentially lowering the health score or increasing the risk of inconsistency. Conversely, if the battery has been used in a mild environment, the compensation module will appropriately increase the health score. This historical data-based adjustment overcomes the limitations of relying solely on real-time data, resulting in more comprehensive and accurate assessment results. The compensation module works together with the contribution separation and quantification module and the inconsistency assessment module to form a correction feedback loop, improving the overall system assessment accuracy.

[0097] In some preferred embodiments, the compensation module can be a software program running on the system's main control unit. This program can access a battery history database that stores key operating parameters and environmental data for each battery cell in the module since its commissioning. After the contribution separation and quantification module and the inconsistency assessment module complete their preliminary assessment, they send the results to the compensation module. The compensation module queries the battery's historical data based on its unique identifier. For example, it can query the battery's maximum temperature, cumulative high-temperature exposure duration, and total charge and discharge capacity during its previous or current life cycle. The compensation module can include a built-in prediction model based on the relationship between historical data and aging, such as a regression model or a machine learning model. This model receives the preliminary assessment results and historical data as input and outputs an adjusted assessment result. For example, if the preliminary assessment results indicate a low level of inconsistency for a battery cell, but historical data indicates that the battery cell has been exposed to temperatures significantly above normal operating temperatures for an extended period, the compensation module's model may determine that the battery has a potential risk of accelerated aging and adjust the inconsistency assessment result upward by a preset amount or based on a correction value calculated by the model. The adjusted evaluation results can then be used for decision-making in the battery management system, such as the development of balancing strategies or the prediction of remaining life.

[0098] Through the above technical solution, the battery assessment results can be adjusted based on the comprehensive analysis of real-time data and historical data, thereby improving the accuracy of the energy storage battery module health monitoring system in assessing the battery health status, providing a more reliable decision-making basis for the battery management system, helping to optimize battery operation management, extend battery life, and improve the overall performance and safety of the energy storage system.

[0099] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for monitoring the health of an energy storage battery module, characterized in that: The method comprises the following steps: S1: Apply a preset micro-current pulse sequence to each single cell in the cascade utilization energy storage battery module; S2: Collect the voltage response data of each single cell under the stimulation of the micro-current pulse train, and record the current temperature data of each single cell; S3: extracting a first characteristic parameter set and a second characteristic parameter set from the voltage response data, wherein the first characteristic parameter set is used to characterize the historical cumulative damage of the single cell, and the second characteristic parameter set is used to characterize the immediate effect of the current temperature on the electrochemical behavior of the single cell; S4: Use a pre-established reference model that combines characteristic parameters, battery aging state, and temperature correspondence to separate and quantify the immediate impact of historical damage and current temperature; S5: Evaluate the inconsistency of each single cell in the battery module based on the historical damage contribution and the immediate impact contribution of the current temperature.

2. The method for monitoring the health of an energy storage battery module according to claim 1, wherein: In step S3, the step of extracting the first characteristic parameter set and the second characteristic parameter set from the voltage response data includes: S31: Obtaining individual characteristic information of a single battery, including at least one of source information, aging path information, or current response characteristic information; S32: Selecting a feature extraction rule for each single battery from a preset feature extraction rule set based on the individual characteristic information; S33: Apply the selected feature extraction rule to extract the first feature parameter set and the second feature parameter set.

3. The method for monitoring the health of an energy storage battery module according to claim 2, wherein: In step S32, the step of selecting feature extraction rules includes: S321: Standardizing individual characteristic information of the single battery; S322: Matching the standardized feature information with the applicable scope of the rules in the preset feature extraction rule set to obtain a matching result; S323: Based on the matching result, a suitable feature extraction rule is selected from the feature extraction rule set, and the rule is applied to perform feature extraction.

4. The method for monitoring the health of an energy storage battery module according to claim 3, wherein: In step S323, the steps of screening out suitable feature extraction rules include: S3231: Setting priority levels for multiple dimensions in the preset priority selection criteria; S3232: Based on the priority level, apply the criteria of each level in order from high to low to screen multiple candidate feature extraction rules.

5. The method for monitoring the health of an energy storage battery module according to claim 1, wherein: In step S4, the reference model includes: Temperature distribution information, which characterizes the temperature difference between a single cell and its surrounding cells; Based on the reference model and temperature distribution information, the historical damage and current temperature impact contributions of the single cell are separated and quantified.

6. The method for monitoring the health of an energy storage battery module according to claim 1, wherein: In step S4, the steps of separating and quantifying historical damage and current temperature include: S41: Determine a coupling relationship between the first characteristic parameter set and the second characteristic parameter set; S42: Construct a compensation function based on the coupling relationship. The compensation function is used to characterize the nonlinear coupling effect of historical damage and current temperature on the electrochemical behavior of the battery. S43: Correcting the first characteristic parameter set or the second characteristic parameter set using the compensation function to obtain a corrected first characteristic parameter set, and / or correcting the second characteristic parameter set using the compensation function to obtain a corrected second characteristic parameter set; S44: Based on the corrected characteristic parameter set and current temperature data, separate and quantify the historical accumulated damage of the single battery and the immediate impact of the current temperature.

7. The method for monitoring the health of an energy storage battery module according to claim 6, wherein: In step S42, the step of constructing a compensation function includes: S421: Obtaining source information or initial manufacturing characteristic information of the single battery; S422: Classify the single battery according to source information or initial manufacturing characteristics to obtain at least one battery category; S423: For each battery category, determining a coupling relationship between a first characteristic parameter set and a second characteristic parameter set within the category; S424: Based on the coupling relationship, a category-specific compensation function is constructed for each battery category to characterize the nonlinear coupling effect between the historical damage and the current temperature of the single battery of this category.

8. The method for monitoring the health of an energy storage battery module according to claim 7, wherein: In step S423, the step of determining the coupling relationship includes: S4231: Obtain voltage response data of at least two single battery samples under different temperature conditions; S4232: Determine, under each temperature condition, a central tendency parameter and a dispersion parameter characterizing the coupling characteristics of the battery type based on voltage response data of at least two samples; S4233: Combining the central tendency parameter and the dispersion parameter determined under different temperature conditions to form a category-specific coupling relationship for the battery category.

9. An energy storage battery module health monitoring system, using the energy storage battery module health monitoring method according to claim 1, characterized in that: The system includes: An excitation module, which is used to apply a micro-current pulse sequence to the single cells in the cascade energy storage battery module; A data acquisition module, configured to acquire voltage response data and temperature data of the battery; Feature extraction module, the feature extraction module extracts the characteristic parameter set from the voltage response data and transmits it to the contribution separation and quantification module for analysis; A contribution separation and quantification module, which analyzes and separates the battery's historical accumulated damage and the immediate impact of current temperature based on a reference model; An inconsistency assessment module is used to assess the health inconsistency of batteries in the module based on the separation result.

10. The energy storage battery module health monitoring system according to claim 9, characterized in that: The system also includes a compensation module for adjusting the battery evaluation result based on a comprehensive analysis of real-time data and historical data.

Images