Battery pack capacity attenuation characteristic decoupling analysis method

Abstract

The invention relates to the technical field of battery pack capacity attenuation analysis, and discloses a battery pack capacity attenuation characteristic decoupling analysis method, which comprises the following steps: planning an analysis process based on battery pack design parameters and working conditions, determining electrochemical impedance spectroscopy content and cyclic aging test content, and determining different discharge rate types; generating a capacity attenuation analysis process set, determining an execution sequence according to an optimal matching principle, and generating a spectrogram acquisition setting list and a test scheme list of each magnification; and finally, associating to generate an overall analysis scheme. And the analysis content is updated through real-time operation data, the process execution sequence is adjusted, and a database is established for data management and long-term tracking analysis. According to the method, decoupling analysis of the capacity attenuation characteristics of the battery pack is realized, the analysis accuracy and reliability are improved, and technical support is provided for optimization of the battery pack.

Description

Technical Field

[0001] The present invention relates to the technical field of battery pack capacity attenuation analysis, and in particular to a decoupling analysis method for battery pack capacity attenuation characteristics. Background Art

[0002] In the energy storage sector, battery packs, as key energy storage units, face capacity degradation, a significant factor restricting their application and development. With the continued advancement of new energy technologies, battery packs are increasingly being used in a wide range of fields, including electric vehicles and energy storage power stations. However, battery capacity degradation directly impacts the device's endurance, service life, and overall performance.

[0003] Current analysis methods for battery pack capacity decay have many shortcomings. Traditional analysis methods often do not fully consider the design parameters and operating conditions of the battery pack, making it difficult to fully and accurately grasp the capacity decay characteristics of the battery pack under different conditions. For example, at different discharge rates, the capacity decay patterns of the battery pack may vary significantly, and traditional methods may not be able to effectively distinguish and analyze these differences. The existing analysis process lacks systematicity and planning. When conducting capacity decay analysis, there is no clear process and method to determine the required electrochemical impedance spectroscopy content and cycle aging test content, resulting in a chaotic analysis process and the inability to obtain reliable analysis results. Moreover, the classification of different discharge rate types is not scientific and reasonable enough, and cannot accurately reflect the actual decay state of the battery pack.

[0004] There are also flaws in testing rules and data processing. Traditional testing rules are often fixed and cannot be dynamically adjusted based on actual conditions. For example, the sample size and number of repetitions for cyclic aging tests cannot be reasonably determined based on the confidence intervals of historical test data. When the measured data deviates from the expected results, supplementary testing cannot be triggered promptly and effectively. Furthermore, insufficient attention is paid to the calibration of test equipment, which can easily lead to large errors in test results.

[0005] In terms of data analysis and management, existing methods lack a comprehensive database to store and manage the various data collected during the analysis process. This makes it difficult to conduct long-term tracking and comparative analysis of battery capacity degradation analysis results, and it is impossible to accurately assess the changing trends of battery capacity degradation over time. Moreover, the analysis of factors such as charging and discharging environmental data is not in-depth enough, and the impact of environmental factors on battery capacity degradation cannot be fully considered. Summary of the Invention

[0006] The object of the present invention is to provide a method for decoupling and analyzing the capacity attenuation characteristics of a battery pack to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for decoupling and analyzing battery pack capacity attenuation characteristics, the method comprising:

[0008] Plan the capacity decay analysis process of the battery pack based on the design parameters and operating conditions of the battery pack, and determine the electrochemical impedance spectroscopy content and cycle aging test content required for the analysis. The electrochemical impedance spectroscopy content includes the internal impedance distribution characteristics of the battery, and the cycle aging test content includes obtaining the measured capacity decay data of the battery pack at different discharge rates;

[0009] Determining different discharge rate types of the battery pack according to the design parameters and operating conditions of the battery pack;

[0010] Generate a capacity fade analysis process set according to different discharge rate types, electrochemical impedance spectroscopy contents, and cycle aging test contents of the battery pack, and determine the execution order of the capacity fade analysis process set based on the optimal matching principle;

[0011] Generate a spectrum acquisition setting list for each rate at each rate according to a preset analysis cycle, different discharge rate types and capacity retention index requirements of the battery pack, and generate a test plan list for each rate at each rate according to preset test rules, different discharge rate types and capacity retention index requirements of the battery pack;

[0012] Based on the execution order of the capacity decay analysis process set, the spectrum acquisition setting list, the test plan list and the battery pack capacity decay analysis content of each rate are associated to generate an overall solution for the battery pack capacity decay feature decoupling analysis.

[0013] Preferably, the capacity decay analysis process of the battery pack is planned based on the design parameters and operating conditions of the battery pack, and the electrochemical impedance spectroscopy content and cycle aging test content required for the analysis are determined, including:

[0014] For a current rate, determining a capacity decay analysis boundary for the current rate based on historical decay data of the current rate, historical decay data of adjacent discharge rates, and a preset rate division threshold;

[0015] Extracting the capacity fade analysis data subset of the current rate according to the capacity fade analysis boundary and a preset time window;

[0016] The content of the electrochemical impedance spectroscopy and cycle aging test related data in the capacity fade analysis data subset is determined as the battery pack capacity fade analysis content at the current rate.

[0017] Preferably, different discharge rate types of the battery pack are determined according to the design parameters and operating conditions of the battery pack, including:

[0018] For the current rate, if the discharge rate of the current rate is greater than a preset rate threshold, it is determined that the current rate is in a high rate discharge state;

[0019] When the discharge rate of the current rate is not greater than the preset rate threshold, classifying the capacity decay rate and internal resistance growth rate data of the current rate based on a first preset classification algorithm to obtain decay category data, and determining a central feature of the decay category data;

[0020] determining an attenuation potential level of the current magnification according to the working conditions of the current magnification;

[0021] When the matching degree between the central feature of each attenuation category data and the preset attenuation category is greater than a preset matching threshold, determining that the current magnification is in a specific attenuation mode state;

[0022] When the matching degree is not greater than the preset matching threshold, it is determined that the current magnification is in a normal attenuation mode state.

[0023] Preferably, the battery pack capacity decay analysis content also includes battery structure data and charge and discharge environment data. The different discharge rate types of the battery pack include high decay state, low decay state and critical decay state. According to the different discharge rate types, electrochemical impedance spectroscopy content and cycle aging test content of the battery pack, a capacity decay analysis process set is generated, including:

[0024] For the current rate, classify the charge and discharge environment data of the current rate based on a second preset classification algorithm to obtain various charge and discharge environment category data; for the current charge and discharge environment category data, filter the current charge and discharge environment category data based on the influence degree and change frequency of each environmental factor in the current charge and discharge environment category data to obtain a key environment subset of the current charge and discharge environment category data, and determine an influence label for the current charge and discharge environment category data based on the influence characteristics of the key environment subset and the average value of the capacity retention rate;

[0025] According to whether the discharge state of the current rate is a high attenuation state.

[0026] Preferably, after determining whether the discharge state of the current rate is a high attenuation state, the method further includes:

[0027] If the discharge state of the current rate is a high attenuation state, matching the impact labels of the charge and discharge environment category data with the capacity retention rate index to generate a targeted suppression strategy;

[0028] If the current rate discharge state is a low decay state, generating a decay control strategy based on a preset optimization rule and a decay control target;

[0029] If the discharge state of the current rate is a critical decay state, a decay stabilization strategy is generated according to historical decay fluctuations and decay stabilization rules.

[0030] Preferably, after generating the capacity decay analysis process set, the method further includes:

[0031] Update the battery pack capacity attenuation analysis content at each rate based on the real-time operating data of the battery pack;

[0032] Recalculate the discharge rate type of each rate according to the updated battery pack capacity decay analysis content;

[0033] The execution order of the capacity fade analysis process set is dynamically adjusted according to the recalculated discharge rate type.

[0034] Preferably, the preset test rules include:

[0035] Dynamically adjust the sample size and number of repetitions of the cycle aging test based on the confidence interval of the battery pack's historical test data;

[0036] When the deviation between the measured capacity decay data and the electrochemical impedance spectroscopy results exceeds the preset error threshold, the supplementary test module is automatically triggered.

[0037] Preferably, the preset test rules also include:

[0038] Establish a test equipment calibration and compensation mechanism to calibrate the test equipment for error compensation based on the baseline value of the standard battery before each test.

[0039] Preferably, after generating the overall solution for decoupling analysis of the battery pack capacity degradation characteristics, the method further includes:

[0040] Establishing a battery pack capacity decay analysis database, storing data such as the spectrum acquisition setting list, test plan list, battery pack capacity decay analysis content, and various discharge rate types in the overall solution into the database;

[0041] Regularly clean and maintain the data in the database to remove invalid and duplicate data;

[0042] Based on this database, the battery pack capacity decay analysis results are tracked and compared over a long period of time to evaluate the changing trend of battery pack capacity decay over time.

[0043] Preferably, the second preset classification algorithm is a K-means clustering algorithm, and the specific steps of classifying the current rate charge and discharge environment data include:

[0044] Calculate the Euclidean distance between each sample in the charge and discharge environment data;

[0045] Randomly initialize the cluster center points according to the preset number of clusters;

[0046] It iteratively assigns samples to the nearest cluster center and recalculates the center point position until the cluster center point no longer changes.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] By planning the capacity decay analysis process based on the design parameters and operating conditions of the battery pack, and clearly defining the content of the electrochemical impedance spectroscopy and cycle aging test, the analysis process becomes more targeted and systematic. This planning fully considers the characteristics of the battery pack under different design parameters and operating conditions, thereby more comprehensively grasping the factors affecting capacity decay. For example, when determining the capacity decay analysis boundary of the current rate, combining historical decay data with the preset rate division threshold can more accurately define the analysis scope and extract a more valuable subset of analysis data, laying the foundation for subsequent in-depth analysis.

[0049] Scientifically and rationally defining the different discharge rate types for a battery pack can more accurately reflect its actual decay state. For the current rate, by comparing the discharge rate with the preset rate threshold and analyzing the capacity decay rate and internal resistance growth rate data, it is possible to clearly distinguish between high-rate discharge states, specific decay modes, and normal decay modes. This classification method fully considers the decay characteristics of the battery pack at different discharge rates and provides an important basis for the subsequent development of targeted analysis and control strategies.

[0050] Generate a capacity decay analysis process set and determine the execution order based on the optimal matching principle, making the analysis process more orderly and efficient. At the same time, combined with the spectrum acquisition setting list and test plan list for each magnification, the accuracy and reliability of the analysis data can be ensured. For example, the preset test rules dynamically adjust the sample size and number of repetitions of the cycle aging test based on the confidence interval of historical test data, and automatically trigger the supplementary test module when the deviation between the measured data and the electrochemical impedance spectroscopy results exceeds the threshold. These measures can effectively improve the quality of test data.

[0051] After determining different discharge states, corresponding suppression strategies, control strategies, and stabilization strategies are generated. Effective measures can be taken for different battery pack decay states, thereby better controlling the battery pack's capacity decay and improving the battery pack's service life and performance. For example, when the current rate is in a high decay state, a targeted suppression strategy is generated to effectively slow the battery pack's decay rate; when it is in a low decay state, a decay control strategy is generated to further optimize the battery pack's performance; and when it is in a critical decay state, a decay stabilization strategy is generated to keep the battery pack's decay state stable.

[0052] By updating analysis content based on real-time battery pack operating data, recalculating discharge rate types, and dynamically adjusting the execution order of process sets, dynamic management of battery pack capacity decay analysis is achieved, promptly reflecting the latest status of the battery pack and improving the real-time and accuracy of the analysis. Furthermore, the establishment of a battery pack capacity decay analysis database for data storage, cleaning, maintenance, and long-term tracking and comparative analysis facilitates a deeper understanding of the changing trends of battery pack capacity decay over time, providing stronger support for the optimized design and use of battery packs.

[0053] When analyzing charging and discharging environmental data, the use of methods such as the K-means clustering algorithm can deeply explore the impact of environmental factors on battery pack capacity degradation, determine key environmental subsets and impact labels, thereby more comprehensively considering the role of environmental factors and further improving the accuracy and reliability of the analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a working principle diagram of the battery pack capacity attenuation characteristic decoupling analysis method according to the present invention;

[0055] Figure 2 Flowchart for capacity fade analysis boundary determination based on historical data;

[0056] Figure 3 Flowchart generated for the decay control strategy based on discharge state;

[0057] Figure 4 Flowchart of the preset rule execution for cyclic aging test. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0059] See also Figures 1-4 The present invention relates to a method for decoupling and analyzing the capacity attenuation characteristics of a battery pack, and the specific implementation steps are as follows:

[0060] The capacity decay analysis process of the battery pack is planned based on the design parameters and operating conditions of the battery pack, and the electrochemical impedance spectroscopy content and cycle aging test content required for the analysis are determined. The electrochemical impedance spectroscopy content includes the internal impedance distribution characteristics of the battery, and the cycle aging test content includes obtaining the measured capacity decay data of the battery pack at different discharge rates.

[0061] Determine the different discharge rate types for the battery pack based on its design parameters and operating conditions. Generate a capacity fade analysis flow set based on the different discharge rate types, electrochemical impedance spectroscopy content, and cycle aging test content of the battery pack, and determine the execution order of the capacity fade analysis flow set based on the optimal matching principle.

[0062] According to the preset analysis cycle, different discharge rate types and capacity retention index requirements of the battery pack, a spectrum acquisition setting list for each rate is generated at each rate, and according to the preset test rules, different discharge rate types and capacity retention index requirements of the battery pack, a test plan list for each rate is generated at each rate.

[0063] Based on the execution order of the capacity decay analysis process set, the spectrum acquisition setting list, test plan list and battery pack capacity decay analysis content of each rate are associated to generate an overall solution for decoupling analysis of battery pack capacity decay characteristics.

[0064] Example 1:

[0065] When planning the battery pack capacity decay analysis process, the historical decay data for the current rate, the historical decay data for adjacent discharge rates, and the preset rate thresholds must be processed. The historical decay data includes records of the battery pack's capacity decay under the same current rate. These records detail the capacity trend over time or number of cycles at that rate. The historical decay data for adjacent discharge rates shows the decay at rates adjacent to the current rate. For example, if the current rate is 2C, the adjacent rates might be 1.5C and 2.5C. The decay data at these adjacent rates provides a reference for analyzing the current rate, helping to determine whether the decay at the current rate is within a reasonable range or whether there are any anomalies. The preset rate threshold is a standard value pre-set based on the battery pack's characteristics and actual application requirements. This threshold is used to define the boundary conditions for different decay analysis scenarios. Its setting should comprehensively consider factors such as battery type, design capacity, and expected usage scenario. For example, for certain power batteries, 1C might be set as the preset rate threshold to distinguish between high-rate and low-rate decay analysis boundaries.

[0066] By comprehensively analyzing the historical attenuation data of the current rate, the historical attenuation data of adjacent discharge rates, and the preset rate division threshold, the capacity attenuation analysis boundary of the current rate can be determined. Specifically, analyzing the historical attenuation data can understand the general rules and range of battery pack capacity attenuation at that rate. The historical attenuation data of adjacent discharge rates can help determine the correlation and difference between the current rate attenuation and the adjacent rates, while the preset rate division threshold is used as a benchmark to determine under what circumstances a more in-depth or more targeted analysis of the attenuation of the current rate is required. For example, when the historical attenuation data of the current rate shows that its attenuation rate is significantly higher than that of the adjacent rates and exceeds the preset rate division threshold, it may be necessary to define the analysis boundary more strictly to focus on the attenuation at that rate.

[0067] Based on the determined capacity decay analysis boundaries and preset time window, extract the capacity decay analysis data subset of the current rate. The preset time window is used to select data within an appropriate time period to ensure the validity and representativeness of the data. The setting of this time window needs to consider factors such as the battery pack's usage cycle, data update frequency, and analysis purpose. For example, if the analysis purpose is to evaluate the recent battery pack decay, the preset time window may be set to the last 3 months; if it is a long-term trend analysis, it may be set to 1 year or longer.

[0068] When extracting a data subset, first determine the data range based on the capacity decay analysis boundary, and then filter out qualified data within the range according to the preset time window. For example, assuming that the capacity decay analysis boundary determines that data with a decay rate within a certain interval needs to be analyzed in depth, and the preset time window is the last 6 months, then the data with a decay rate within the last 6 months within this interval will be filtered out from the historical data to form a targeted data subset. This data subset contains key information related to the current rate capacity decay within this time period, such as the capacity value under different cycle numbers, the corresponding operating temperature, the charge and discharge current, etc.

[0069] The content of the electrochemical impedance spectroscopy and cycle aging test related data within the capacity decay analysis data subset is determined as the battery pack capacity decay analysis content at the current rate. The electrochemical impedance spectroscopy data can reflect the impedance distribution characteristics within the battery, including ohmic impedance, polarization impedance, etc. The changes in these impedances are closely related to the battery's capacity decay. The cycle aging test related data includes the measured capacity decay data of the battery pack at different discharge rates. By analyzing these data, we can understand the capacity decay law of the battery pack during the cycle process.

[0070] Within the capacity fade analysis data subset, data related to electrochemical impedance spectroscopy and cycle aging testing must be accurately identified and extracted. For example, within the data subset, impedance data corresponding to all time points where electrochemical impedance spectroscopy testing was performed, as well as capacity fade data from cycle aging tests performed around these time points, will be screened. This data will serve as an important basis for subsequent analysis, enabling in-depth study of the characteristics and mechanisms of battery pack capacity fade at the current rate.

[0071] Example 2:

[0072] When determining the different discharge rate types of a battery pack, for the current rate, it is necessary to first clarify the relationship between its discharge rate and the preset rate threshold. The setting of the preset rate threshold needs to be combined with the design parameters of the battery pack, such as the battery type (lithium-ion battery, nickel-metal hydride battery, etc.), the characteristics of the positive and negative electrode materials, the electrolyte formula, etc., while considering the working conditions, including the normal operating temperature range, the charge and discharge cut-off voltage, and the actual application scenarios (such as electric vehicle drive, energy storage system, etc.). For example, for a certain ternary lithium-ion battery pack, its nominal capacity is 100Ah when it is designed. In the fast charging scenario of electric vehicles, 2C (i.e., 200A discharge current) may be set as the preset rate threshold to distinguish between high-rate and non-high-rate discharge states.

[0073] When the current discharge rate is greater than the preset rate threshold, it can be directly determined to be a high-rate discharge state. This is because during high-rate discharge, the chemical reaction rate inside the battery accelerates, the ion diffusion resistance increases, and the polarization phenomenon becomes more significant, which often leads to a significant difference in the capacity decay rate compared to low-rate discharge. For example, if the current rate is 3C, which is greater than the preset 2C threshold, the internal resistance heat generation of the battery increases during the discharge process, and the structural stability of the active material may decrease faster. The capacity decay characteristics in this state are unique and need to be classified and analyzed separately.

[0074] When the current discharge rate is no greater than the preset rate threshold, the capacity decay rate and internal resistance growth rate data for the current rate are classified based on a first preset classification algorithm. The capacity decay rate refers to the decay of battery capacity per unit cycle or unit time, calculated as (initial capacity - current capacity) / initial capacity × 100%. The internal resistance growth rate is (current internal resistance - initial internal resistance) / initial internal resistance × 100%. These two parameters are key indicators of battery aging.

[0075] The first preset classification algorithm can employ a clustering or classification method suitable for processing continuous data, such as a decision tree algorithm or a support vector machine algorithm. Taking the decision tree algorithm as an example, the capacity decay rate and internal resistance growth rate data must first be preprocessed, including filling missing values, removing outliers, and normalizing. Missing values ​​can be filled by interpolating adjacent data points, outliers can be identified and removed according to the 3σ principle, and normalization maps the data to the [0, 1] interval to eliminate dimensionality effects.

[0076] The processed data is input into the decision tree model, which recursively divides the feature space, selects the optimal splitting features and splitting points, and divides the data into different attenuation categories. For example, the decision tree may first divide the data into two categories based on whether the capacity attenuation rate is greater than 5%, and then further subdivide each category based on whether the internal resistance growth rate is greater than 10%, ultimately obtaining multiple attenuation category data. Each attenuation category data corresponds to a group of samples with similar attenuation characteristics, and then determines the central feature of each attenuation category data. For example, the average capacity attenuation rate of a certain category is 8%, and the average internal resistance growth rate is 12%. This average is the central feature of the category, representing the typical attenuation feature of the samples in this category.

[0077] Next, the degradation potential level is determined based on the current rate operating conditions. These operating conditions encompass several dimensions: temperature. For example, when a battery pack operates in a temperature range of -20°C to 60°C, low temperatures can reduce the diffusion rate of lithium ions, while high temperatures can accelerate electrolyte decomposition, significantly impacting degradation potential. Charge and discharge current distribution, such as whether the pack frequently operates under high-current pulses. Humidity. High humidity can cause corrosion of the battery casing, affecting internal sealing. Vibration. In scenarios such as electric vehicles, continuous vibration can lead to poor electrode contact and increase internal resistance.

[0078] The assessment of attenuation potential requires the establishment of a multi-factor comprehensive evaluation model. For example, factors such as temperature, current, humidity, and vibration can be assigned different weights: temperature might be weighted 30%, current 40%, and humidity and vibration 15% each. Each factor is assigned a different score based on actual operating parameters. For example, an operating temperature between 45°C and 55°C is scored 80 points, while temperatures above 55°C are scored 90 points. The weights are then combined to calculate a comprehensive score, and the attenuation potential level is determined based on the score range. For example, a score of 80-100 indicates high potential, 60-80 indicates medium potential, and scores below 60 indicate low potential.

[0079] Next, the degree of match between the central features of each degradation category data and the preset degradation categories must be determined. Preset degradation categories are typical degradation patterns derived from statistical analysis of historical degradation data of battery packs under different operating conditions. For example, they can be divided into categories such as "capacity-dominated degradation," "internal resistance-dominated degradation," and "mixed degradation." Each category has a predefined range for the central features. For example, the central features of "capacity-dominated degradation" are a capacity decay rate greater than 10% and an internal resistance growth rate less than 8%.

[0080] The degree of matching can be calculated using methods such as Euclidean distance or cosine similarity. Taking Euclidean distance as an example, the Euclidean distance between the central feature of the current attenuation category and the central feature of the preset attenuation category is calculated. The smaller the distance, the higher the degree of matching. The preset matching threshold is set according to actual application requirements. For example, when the Euclidean distance is less than 0.5, it is considered that the degree of matching is greater than the preset threshold. When the degree of matching is greater than the preset matching threshold, the current magnification is determined to be a specific attenuation mode state. For example, the Euclidean distance between the central feature of a certain attenuation category and the central feature of the preset "capacity-dominated attenuation" category is 0.3, which is less than the threshold of 0.5. It can be determined as a capacity-dominated state in a specific attenuation mode; when the degree of matching is not greater than the preset matching threshold, it means that the current attenuation feature does not fit any preset typical category and is determined to be a normal attenuation mode state, that is, its attenuation feature belongs to a common attenuation mode without significant dominant factors.

[0081] Example 3:

[0082] When generating a capacity decay analysis process set, for the current rate, the charge and discharge environment data must be classified based on the second preset classification algorithm. Here, the second preset classification algorithm uses the K-means clustering algorithm. The charge and discharge environment data contains multiple dimensions, such as ambient temperature, humidity, charge and discharge current fluctuation range, voltage cutoff threshold, etc. Each data sample can be represented as a multidimensional vector, such as [x1,x2,x3,...,x n ], where x1 represents the ambient temperature, x2 represents the humidity, and x3 represents the charge and discharge current fluctuation range, etc.

[0083] First, calculate the Euclidean distance between each sample in the charge and discharge environment data. The calculation formula of Euclidean distance is:

[0084]

[0085] in, Representation sample and samples The Euclidean distance between is the data dimension (i.e. the number of environmental factors), Representation sample No. The value of the environmental factor, Representation sample No. The formula is used to measure the distance between two samples in multidimensional space. The smaller the distance, the more similar the environmental factor combinations of the two samples are.

[0086] The cluster centers are randomly initialized according to the preset number of clusters. The preset number of clusters is predetermined based on the distribution characteristics of the charge and discharge environment data and the analysis requirements. For example, , then initially a random selection will be made in the data space points as the center of each cluster, denoted as , each center point is also a dimensional vector, corresponding to the value of each environmental factor.

[0087] The samples are then iteratively assigned to the nearest cluster center and the center point position is recalculated until the cluster center point no longer changes. The specific iterative process is as follows: For each sample, its Euclidean distance to all cluster centers is calculated, and the sample is assigned to the cluster with the closest distance. After all samples are assigned, the mean of each environmental factor for all samples in each cluster is recalculated and used as the new cluster center point. If the difference between the new center point and the original center point is less than the preset convergence threshold (for example, the absolute value of the difference in each dimension is less than 0.01), the iteration ends. Otherwise, the above assignment and calculation process is repeated, and finally the data for each charging and discharging environment category is obtained.

[0088] The current charging and discharging environment category data obtained needs to be screened based on the impact and frequency of change of each environmental factor in this category data to obtain a critical environment subset. The impact can be measured by the correlation coefficient between each environmental factor and battery capacity decay in historical data. The larger the absolute value of the correlation coefficient, the higher the impact. The frequency of change refers to the number of fluctuations or the magnitude of changes in an environmental factor within a certain period of time. For example, the greater the temperature fluctuation range within a day, the higher the frequency of change. When screening, set the impact threshold and the change frequency threshold, retain the environmental factors that exceed both thresholds, and form the critical environment subset.

[0089] The impact label of the current charge and discharge environment category data is determined based on the impact characteristics of the key environment subset and the mean capacity retention rate. The impact characteristics include the direction of the key environmental factors on capacity decay (for example, temperature increases generally accelerate decay) and the intensity of the effect (for example, the increase in capacity decay rate for every 10°C temperature increase); the mean capacity retention rate is the average capacity retention rate of the samples corresponding to the category data, and the calculation formula is:

[0090]

[0091] in, is the number of samples in the category data, For the By comprehensively analyzing the influencing characteristics and mean values, we can determine the influencing labels, such as "high temperature and high humidity accelerated attenuation type" and "current fluctuation dominated attenuation type".

[0092] After determining whether the discharge state at the current rate is a high-attenuation state, if it is a high-attenuation state, the impact label of each charge and discharge environment category data is matched with the capacity retention rate index. The capacity retention rate index is a standard value set according to the design requirements and usage scenarios of the battery pack, such as requiring the capacity retention rate to be no less than 80% after 100 cycles. When matching, the specific impact of the combination of environmental factors corresponding to the impact label on the capacity retention rate is analyzed. For example, if an impact label is "high temperature and high humidity accelerated attenuation type", and its corresponding capacity retention rate average is 75%, which is lower than the index value of 80%, then a targeted suppression strategy is generated for high temperature and high humidity factors, such as optimizing the battery heat dissipation structure, adding a moisture-proof coating, etc.

[0093] If the current rate discharge state is low attenuation, a decay control strategy is generated based on preset optimization rules and decay control targets. The preset optimization rules include a series of optimization measures for low attenuation states, such as appropriately increasing the charge cutoff voltage to improve capacity retention when the ambient temperature is low; decay control targets, such as requiring the decay rate to be controlled within 5% for the next 100 cycles. Combining the rules and targets, specific strategies are formulated, such as adopting a constant current and constant voltage charging method and appropriately extending the constant voltage charging time when the ambient temperature is in the low attenuation range of 20°C-25°C.

[0094] If the current rate of discharge is in a critical decay state, a decay stabilization strategy is generated based on historical decay fluctuations and decay stabilization rules. Historical decay fluctuations are measured by calculating the standard deviation of the capacity decay rate over a period of time. The larger the standard deviation, the more severe the fluctuation. The decay stabilization rule is a guideline set to stabilize the decay rate. For example, when the decay rate fluctuates by more than 10%, the charge and discharge parameters need to be adjusted. For example, if the standard deviation of the historical decay rate is 8%, close to the fluctuation threshold of 10%, and the current state is critical decay, the generated strategy is to monitor the temperature and current in real time during the charge and discharge process. When the fluctuation exceeds the threshold, it automatically switches to constant current discharge mode to stabilize the decay rate.

[0095] Example 4:

[0096] After generating the capacity decay analysis process set, the capacity decay analysis content of the battery pack at each rate needs to be updated based on the real-time operation data of the battery pack. The real-time operation data of the battery pack comes from the battery management system (BMS) or other monitoring equipment, covering information from multiple dimensions. For example, during the operation of the battery pack, the real-time collected discharge current data can reflect the current discharge rate, the voltage data can reflect the change in the terminal voltage of the battery during the discharge process, the temperature data includes the temperature distribution of the battery cell and module, the cycle number records the number of charge and discharge cycles that the battery pack has completed, in addition to real-time capacity data and internal resistance data. These data are collected in real time by sensors and transmitted to the data processing unit, forming a dynamic data stream that can accurately reflect the latest status of the battery pack in actual operation.

[0097] During the update process, real-time operating data must first be preprocessed. This includes data filtering to remove noise that may occur during the acquisition process, such as using a Kalman filter algorithm to smooth voltage and current data. It also includes outlier detection and elimination, identifying abnormal data by setting a reasonable threshold range. For example, if the temperature of a single battery cell suddenly rises to 100°C, it is clearly outside the normal operating range and can be identified as an outlier and eliminated. Simultaneously, data normalization is performed to convert data of different dimensions to a unified scale for subsequent analysis, such as normalizing temperature data to the range [0, 1].

[0098] The pre-processed real-time operating data will be integrated with the original capacity decay analysis content. The original analysis content includes historical decay data, electrochemical impedance spectroscopy data, cycle aging test data, etc. During fusion, the real-time data is inserted into the historical data sequence in chronological order to form an updated data set. For example, the original capacity decay curve is drawn based on the data of the past 100 cycles. When there is new real-time data for the 101st cycle, it is added to the data sequence and the capacity decay curve containing the latest data is regenerated. In this way, the capacity decay analysis content can reflect the current decay characteristics of the battery pack in a timely manner. For example, if the real-time data shows that the capacity decay rate of the last few cycles has significantly accelerated, the updated analysis content will highlight this change.

[0099] The discharge rate type for each rate is recalculated based on the updated battery pack capacity decay analysis content. The recalculation process is basically the same as the logic for initially determining the discharge rate type, but is based on more comprehensive and up-to-date data. First, for each rate, the updated capacity decay rate and internal resistance growth rate data are obtained. The capacity decay rate is calculated based on the updated initial capacity and the current real-time capacity, using the formula (initial capacity - current real-time capacity) / initial capacity × 100%; the internal resistance growth rate is calculated based on the initial internal resistance and the real-time internal resistance, i.e. (real-time internal resistance - initial internal resistance) / initial internal resistance × 100%.

[0100] For each rate, the relationship between its discharge rate and the preset rate threshold is determined. If the current discharge rate is greater than the preset rate threshold, it is still determined to be a high-rate discharge state; if it is not greater than the preset rate threshold, the capacity decay rate and internal resistance growth rate data are again classified based on the first preset classification algorithm. The first preset classification algorithm here is the same as before, for example, using a decision tree algorithm to re-feature and classify the updated data, obtain new decay category data, and determine the central feature of each decay category data.

[0101] The degradation potential level at the current rate is re-evaluated based on the operating condition information in the updated real-time operating data, such as the latest temperature, charge and discharge current, and ambient humidity. Changes in operating conditions may cause changes in the degradation potential level. For example, if the battery pack originally operated at 25°C with a low degradation potential level, but the real-time data shows that the ambient temperature has risen to 40°C, the degradation potential level needs to be recalculated and may be raised to a medium potential level.

[0102] The degree of match between the central feature of each decay category data and the preset decay category is re-evaluated. Due to data updates, the central feature may change, and the degree of match will also change accordingly. For example, if the original central feature does not sufficiently match the preset decay category, indicating a normal decay mode, but the updated data causes the central feature to match a preset decay category (such as "Internal Resistance Dominant Decay") above the preset matching threshold, the discharge rate type for that rate will be re-determined as a specific decay mode.

[0103] Dynamically adjust the execution order of the capacity decay analysis process set based on the recalculated discharge rate type. The capacity decay analysis process set contains multiple analysis steps or modules, and different discharge rate types may correspond to different priorities or execution orders. For example, for the rate of high-rate discharge state, its decay analysis process may need to prioritize electrochemical impedance spectroscopy analysis to quickly locate the impact of internal impedance changes on capacity decay; and for the rate of specific decay mode state, it may be necessary to first perform a cluster analysis of the charge and discharge environment data to determine the impact of environmental factors.

[0104] When dynamically adjusting the execution order, a set of adjustment rules must be established. The rules can be based on the priority of the discharge rate type, for example, a high-rate discharge state has a higher priority than a specific decay mode state, and a specific decay mode state has a higher priority than a normal decay mode state; they can also be based on the urgency of the decay. For example, when the decay rate of a certain rate suddenly accelerates and is recalculated to be a high decay state, the corresponding analysis process is executed in advance. During the adjustment process, the logical coherence between the various analysis processes must be ensured. For example, before performing a cycle aging test, the electrochemical impedance spectroscopy analysis should be completed to obtain the initial state of the battery's internal impedance.

[0105] Through the above updating, recalculating and dynamic adjustment process based on real-time operating data, the decoupling analysis method of battery pack capacity attenuation characteristics can have real-time adaptability, respond promptly to changes in the battery pack operating status, and ensure the accuracy and effectiveness of the analysis results.

[0106] Example 5:

[0107] In the decoupling analysis of battery pack capacity attenuation characteristics, pre-set test rules are a key link in ensuring the accuracy and reliability of test data. First, the sample size and number of repetitions of the cycle aging test are dynamically adjusted based on the confidence interval of the battery pack's historical test data. The historical test data contains the cycle aging test results of the battery pack under different operating conditions and different rates in the past. For example, in the 100 cycle tests conducted at a 2C discharge rate, the capacity attenuation data and internal resistance change data of each test were obtained. The confidence interval is calculated based on statistical methods and is used to represent the reliability range of historical test data. It usually estimates the interval of the overall parameter with a certain probability (such as 95%).

[0108] Specifically, when the confidence interval of historical test data is small, it indicates that the data has a low degree of dispersion and high reliability. In this case, the sample size and number of repetitions of the cyclic aging test can be appropriately reduced to improve test efficiency. For example, if the confidence interval of a certain type of test data is [0.85, 0.90], it means that there is a 95% probability that the population parameter falls within this interval, and the interval range is small. The sample size can be reduced from 50 to 30, and the number of repetitions can be adjusted from 3 to 2. Conversely, when the confidence interval is large, it indicates that the data has a high degree of dispersion and low reliability. The sample size and number of repetitions need to be increased to reduce the impact of random errors. For example, if the confidence interval is [0.70, 0.95] and the interval range is wide, the sample size can be increased to 80 and the number of repetitions can be increased to 4. More test samples can be used to narrow the confidence interval and improve the credibility of the data.

[0109] When the deviation between the measured capacity decay data and the electrochemical impedance spectroscopy results exceeds the preset error threshold, the supplementary test module is automatically triggered. The preset error threshold is set according to the test accuracy requirements of the battery pack and the actual application scenario, for example, it is set to 5%. The measured capacity decay data is the decay of the battery capacity with the number of cycles obtained through the cycle aging test, and the electrochemical impedance spectroscopy results reflect the distribution and changes of the internal impedance of the battery. When the deviation between the two exceeds the threshold, it may mean that there is an abnormality in the test process, such as a failure of the test equipment, fluctuations in environmental conditions, or an abnormality in the battery itself.

[0110] After the supplementary test module is triggered, the test environment and equipment will be checked first to ensure that the environmental conditions such as temperature and humidity meet the test requirements and that the accuracy and stability of the equipment meet the standards. Then the relevant cycle aging test and electrochemical impedance spectroscopy test will be re-performed to obtain new test data. For example, if the measured capacity decay rate in the first test is 12%, and the equivalent capacity decay rate obtained by electrochemical impedance spectroscopy analysis is 18%, the deviation reaches 6%, which exceeds the preset 5% threshold. The supplementary test module will be triggered and the test will be performed again under the same conditions. If the deviation of the second test is reduced to 3%, it is considered that the first deviation may be caused by accidental factors, and the second test data will be used; if the deviation is still large, it is necessary to further investigate the cause, such as checking whether the battery has internal micro-short circuits and other problems.

[0111] The pre-set test rules also include establishing a calibration and compensation mechanism for test equipment. Before each test, the test equipment is calibrated for error compensation based on the baseline values ​​of a standard battery. A standard battery is a battery that has been calibrated with high precision, with parameters such as capacity and internal resistance having known, accurate baseline values. For example, a standard battery has a rated capacity of 20Ah and an internal resistance of 10mΩ. Establishing the calibration and compensation mechanism for test equipment requires first determining the equipment's error model. By performing multiple measurements on the standard battery, recording the differences between the equipment's measured values ​​and the baseline values, and establishing a mapping relationship between the measured values ​​and the actual values, the system then establishes a mapping relationship between the measured values ​​and the actual values.

[0112] Before each test, a standard battery is connected to the test equipment for measurement. The equipment's measured values ​​for the standard battery are obtained. For example, if the measured capacity of the standard battery is 19.8Ah and the internal resistance is 10.5mΩ, the capacity deviation is -1% and the internal resistance deviation is +5% compared to the baseline values. Based on a pre-established error model, the equipment's measured data is compensated and calibrated. For example, in subsequent tests, the measured capacity value is multiplied by 1.01 (1÷(1-1%)) and the measured internal resistance value is multiplied by 0.95 (1÷(1+5%)) to eliminate the impact of equipment errors. This calibration and compensation mechanism ensures that the test equipment maintains high measurement accuracy during each test, providing reliable raw data for subsequent capacity decay analysis.

[0113] After generating an overall solution for decoupling analysis of battery pack capacity decay characteristics, it is necessary to establish a battery pack capacity decay analysis database to store data such as the spectrum acquisition setting list, test plan list, battery pack capacity decay analysis content, and various discharge rate types in the overall solution. The structural design of the database needs to consider the classified storage and rapid retrieval of data. For example, the spectrum acquisition setting list can be classified and stored according to different rates and different analysis cycles, and the test plan list can be archived by test type (cyclic aging test, electrochemical impedance spectroscopy test, etc.) and test conditions. The battery pack capacity decay analysis content includes historical decay data, real-time operation data, analysis results, etc. The data of each discharge rate type contains information such as classification results and decay patterns at different rates.

[0114] Regularly clean and maintain the data in the database to remove invalid and duplicate data. Invalid data includes abnormal data caused by equipment failure, human error, and other reasons during the test process. For example, a piece of capacity decay data shows a sudden increase in capacity, which clearly does not conform to the law of battery aging and can be determined as invalid data and deleted. Duplicate data refers to the same data collected multiple times under the same test conditions, such as capacity data recorded repeatedly at the same rate and the same number of cycles. Only one piece of duplicate data needs to be retained, and the remaining duplicate data should be deleted to save storage space and improve data query efficiency. The cleaning and maintenance cycle can be set according to the frequency of data updates, such as performing a comprehensive data cleanup once a month.

[0115] Based on this database, the battery pack capacity decay analysis results are tracked and compared over a long period of time to evaluate the changing trend of battery pack capacity decay over time. Long-term tracking and analysis requires regular extraction of the latest capacity decay data from the database and comparison with historical data, such as extracting data once a quarter, drawing a capacity decay curve, and observing the slope of the curve to understand the change in decay rate. Comparative analysis can be performed between battery packs at different rates and different working conditions, such as comparing the capacity decay curves at 2C discharge rate and 1C discharge rate to analyze the impact of high rate on decay; it can also be performed between different time periods of the same battery pack, such as comparing the decay data of the battery pack after 1 year of use and after 2 years of use to evaluate the decay trend after long-term use.

[0116] Long-term tracking and comparative analysis can uncover potential patterns and anomalies in battery pack capacity degradation. For example, if tracking and analysis reveal a sudden acceleration in the degradation rate of a battery pack under normal operating conditions, this could indicate early failure and require timely maintenance or replacement. Comparative analysis of degradation data from different batches of battery packs can reveal differences in design or manufacturing processes, providing a basis for optimizing and improving the battery pack. This database-based long-term analysis mechanism provides data support for battery pack reliability assessment, lifespan prediction, and maintenance strategy formulation, helping to improve the efficiency and cost-effectiveness of battery packs.

[0117] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0118] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for decoupling analysis of battery pack capacity attenuation characteristics, characterized in that: The method comprises: Plan the capacity decay analysis process of the battery pack based on the design parameters and operating conditions of the battery pack, and determine the electrochemical impedance spectroscopy content and cycle aging test content required for the analysis. The electrochemical impedance spectroscopy content includes the internal impedance distribution characteristics of the battery, and the cycle aging test content includes obtaining the measured capacity decay data of the battery pack at different discharge rates; Determining different discharge rate types of the battery pack according to the design parameters and operating conditions of the battery pack; Generate a capacity fade analysis process set according to different discharge rate types, electrochemical impedance spectroscopy contents, and cycle aging test contents of the battery pack, and determine the execution order of the capacity fade analysis process set based on the optimal matching principle; Generate a spectrum acquisition setting list for each rate at each rate according to a preset analysis cycle, different discharge rate types and capacity retention index requirements of the battery pack, and generate a test plan list for each rate at each rate according to preset test rules, different discharge rate types and capacity retention index requirements of the battery pack; Based on the execution order of the capacity decay analysis process set, the spectrum acquisition setting list, the test plan list and the battery pack capacity decay analysis content of each rate are associated to generate an overall solution for the battery pack capacity decay feature decoupling analysis.

2. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 1, characterized in that: Plan the capacity decay analysis process of the battery pack based on the design parameters and operating conditions of the battery pack, and determine the electrochemical impedance spectroscopy content and cycle aging test content required for the analysis, including: For a current rate, determining a capacity decay analysis boundary for the current rate based on historical decay data of the current rate, historical decay data of adjacent discharge rates, and a preset rate division threshold; Extracting the capacity fade analysis data subset of the current rate according to the capacity fade analysis boundary and a preset time window; The content of the electrochemical impedance spectroscopy and cycle aging test related data in the capacity fade analysis data subset is determined as the battery pack capacity fade analysis content at the current rate.

3. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 1, characterized in that: Different discharge rate types of the battery pack are determined according to the design parameters and operating conditions of the battery pack, including: For the current rate, if the discharge rate of the current rate is greater than a preset rate threshold, it is determined that the current rate is in a high rate discharge state; When the discharge rate of the current rate is not greater than the preset rate threshold, classifying the capacity decay rate and internal resistance growth rate data of the current rate based on a first preset classification algorithm to obtain decay category data, and determining a central feature of the decay category data; determining an attenuation potential level of the current magnification according to the working conditions of the current magnification; When the matching degree between the central feature of each attenuation category data and the preset attenuation category is greater than a preset matching threshold, determining that the current magnification is in a specific attenuation mode state; When the matching degree is not greater than the preset matching threshold, it is determined that the current magnification is in a normal attenuation mode state.

4. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 3, characterized in that: The battery pack capacity decay analysis content also includes battery structure data and charge and discharge environment data. The different discharge rate types of the battery pack include high decay state, low decay state and critical decay state. According to the different discharge rate types, electrochemical impedance spectroscopy content and cycle aging test content of the battery pack, a capacity decay analysis process set is generated, including: For the current rate, classify the charge and discharge environment data of the current rate based on a second preset classification algorithm to obtain various charge and discharge environment category data; for the current charge and discharge environment category data, filter the current charge and discharge environment category data based on the influence degree and change frequency of each environmental factor in the current charge and discharge environment category data to obtain a key environment subset of the current charge and discharge environment category data, and determine an influence label for the current charge and discharge environment category data based on the influence characteristics of the key environment subset and the average value of the capacity retention rate; According to whether the discharge state of the current rate is a high attenuation state.

5. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 4, characterized in that: After determining whether the discharge state of the current rate is a high attenuation state, the method further includes: If the discharge state of the current rate is a high attenuation state, matching the impact labels of the charge and discharge environment category data with the capacity retention rate index to generate a targeted suppression strategy; If the current rate discharge state is a low decay state, generating a decay control strategy based on a preset optimization rule and a decay control target; If the discharge state of the current rate is a critical decay state, a decay stabilization strategy is generated according to historical decay fluctuations and decay stabilization rules.

6. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 5, characterized in that: After generating the capacity fade analysis process set, the following steps are also included: Update the battery pack capacity attenuation analysis content at each rate based on the real-time operating data of the battery pack; Recalculate the discharge rate type of each rate according to the updated battery pack capacity decay analysis content; The execution order of the capacity fade analysis process set is dynamically adjusted according to the recalculated discharge rate type.

7. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 1, characterized in that: The preset test rules include: Dynamically adjust the sample size and number of repetitions of the cycle aging test based on the confidence interval of the battery pack's historical test data; When the deviation between the measured capacity decay data and the electrochemical impedance spectroscopy results exceeds the preset error threshold, the supplementary test module is automatically triggered.

8. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 7, characterized in that: The preset test rules also include: Establish a test equipment calibration and compensation mechanism to calibrate the test equipment for error compensation based on the baseline value of the standard battery before each test.

9. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 1, characterized in that: After generating the overall solution for decoupling analysis of the battery pack capacity degradation characteristics, the following steps are also included: Establishing a battery pack capacity decay analysis database, storing data such as the spectrum acquisition setting list, test plan list, battery pack capacity decay analysis content, and various discharge rate types in the overall solution into the database; Regularly clean and maintain the data in the database to remove invalid and duplicate data; Based on this database, the battery pack capacity decay analysis results are tracked and compared over a long period of time to evaluate the changing trend of battery pack capacity decay over time.

10. The battery pack capacity attenuation characteristic decoupling analysis method according to claim 4, characterized in that: The second preset classification algorithm is a K-means clustering algorithm, and the specific steps of classifying the current rate charge and discharge environment data include: Calculate the Euclidean distance between each sample in the charge and discharge environment data; Randomly initialize the cluster center points according to the preset number of clusters; It iteratively assigns samples to the nearest cluster center and recalculates the center point position until the cluster center point no longer changes.

Images