A BMS testing method, device, equipment and storage medium

Abstract

The present invention relates to a BMS testing method, device, equipment and storage medium, comprising the following steps: collecting data from a BMS system of a battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data; estimating the state of charge of the battery pack based on the preprocessed data to obtain a state of charge estimation value; constructing a time series change curve based on the state of charge estimation value, and evaluating the remaining service life of the battery pack based on the time series change curve to obtain a remaining service life evaluation result; analyzing the health status of the battery pack based on the remaining service life evaluation to obtain a health status report of the battery pack; formulating a management strategy for the battery pack based on the health status report; solving the problem of complex chemical reactions and dynamic changes of batteries, and the problem of insufficient accuracy often faced by simply relying on these data to evaluate the battery status.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, and in particular to a BMS testing method, device, equipment and storage medium. Background Art

[0002] Battery Management System (BMS) is a key technology in modern electric vehicles and energy storage systems, designed to ensure the safety, efficiency and life of battery packs. However, with the popularity of electric vehicles and the increase in energy storage demand, the performance, life and safety management of battery packs have become increasingly complex. Although the current BMS system can monitor key data such as battery voltage, temperature and current, due to the complex chemical reactions and dynamic changes of the battery, relying solely on these data for battery status assessment often faces the problem of insufficient accuracy. In addition, the existing BMS methods for estimating the battery state of charge are mostly based on simple mathematical models, and fail to make full use of more dynamic information such as timing change curves for evaluation.

[0003] At present, the remaining service life assessment of battery packs still has great uncertainty, especially in the long-term operation process, the degradation speed and mode of the battery will be affected by many factors, such as working environment, charge and discharge rate, etc. These complex factors cause the existing BMS system to often deviate from the actual results in life prediction, affecting the overall management strategy of the battery. Health status analysis is still limited to basic data statistics, lacking a deep understanding of battery degradation patterns, which makes it difficult to accurately optimize subsequent management measures. The existence of these problems increases the risk of system operation, especially in large-scale energy storage applications, where the unpredictability of the battery system may lead to serious consequences.

[0004] Therefore, developing a more intelligent and accurate BMS test method has become a research hotspot. By combining data acquisition, preprocessing, timing analysis and health status assessment, a more comprehensive state monitoring and life prediction can be provided for the battery pack. This method can not only improve the accuracy of the state of charge estimation, but also make a more reliable prediction of the remaining life of the battery, providing a scientific basis for the subsequent management strategy formulation, effectively extending the battery life and improving its safety. Summary of the invention

[0005] The main purpose of the present invention is to provide a BMS testing method, device, equipment and storage medium, which solves the complex chemical reactions and dynamic changes of the battery. Simply relying on these data to evaluate the battery status often faces the problem of insufficient accuracy.

[0006] To achieve the above object, the present invention provides a BMS testing method, comprising the following steps:

[0007] Performing data collection on a BMS system of a battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data; wherein the collected data includes voltage data, current data and temperature data;

[0008] Estimating the state of charge of the battery pack based on the preprocessed data to obtain a state of charge estimation value;

[0009] Constructing a time series variation curve based on the estimated state of charge value, and evaluating the remaining service life of the battery pack based on the time series variation curve to obtain a remaining service life evaluation result;

[0010] Performing a health status analysis on the battery pack based on the remaining service life assessment to obtain a health status report of the battery pack;

[0011] A management strategy for the battery pack is formulated based on the health status report.

[0012] Further, the data collection of the BMS system of the battery pack to obtain the collected data, and the preprocessing of the collected data to obtain the preprocessed data include:

[0013] The BMS system of the battery pack is subjected to high-resolution data collection through preset sensors to obtain collected data; wherein the preset sensors include electrical sensors and temperature sensors;

[0014] The Hilbert-Huang transform technique is used to perform time-frequency analysis on the collected data to obtain multi-scale time-frequency analysis data;

[0015] Performing time series anomaly detection on the multi-scale time-frequency analysis data to obtain an anomaly detection result;

[0016] If the abnormal detection result is that the multi-scale time-frequency analysis data has abnormal data, the abnormal data is marked to form a marked data set; wherein the abnormal data includes abnormal voltage fluctuations, abnormal current spikes, and abnormal temperature changes;

[0017] Performing spatial interpolation and repair on the labeled data set by using a preset Kriging interpolation algorithm to obtain a repaired data set;

[0018] The repair data set is normalized to obtain preprocessed data.

[0019] Further, the state of charge of the battery pack is estimated based on the preprocessed data to obtain an estimated value of the state of charge, including:

[0020] Performing data fusion on the preprocessed data to obtain fused data;

[0021] Performing nonlinear observation on the state of charge of the battery pack based on the fused data to obtain a preliminary state of charge distribution;

[0022] Predicting the real-time state of the battery pack based on the preliminary state of charge distribution to obtain a preliminary state of charge estimation value;

[0023] Performing state calibration on the preliminary state-of-charge estimate to obtain a calibrated state-of-charge estimate;

[0024] Through a preset multi-task learning algorithm, a multi-dimensional analysis is performed on various states of the battery pack based on the calibrated state of charge estimation value to obtain a multi-dimensional analysis result; wherein the multi-dimensional analysis result includes the impact of the charging and discharging efficiency, power output capacity and single cell consistency on the performance of the entire battery pack;

[0025] Acquire internal connection relationship data and topological structure of the battery pack, and obtain energy consumption data of the battery pack based on the internal connection relationship data and topological structure and multi-dimensional analysis results;

[0026] The calibrated state of charge estimate is estimated based on the energy consumption data to obtain a state of charge estimate.

[0027] Further, a time series variation curve is constructed based on the estimated state of charge value, and a remaining service life of the battery pack is evaluated based on the time series variation curve to obtain a remaining service life evaluation result, including:

[0028] Decomposing the state of charge estimation value in the time-frequency domain by wavelet transform to obtain multi-scale time-frequency features; wherein the multi-scale time-frequency features include low-frequency trend features and high-frequency fluctuation features;

[0029] Constructing a state of charge change trend curve based on the multi-scale time-frequency characteristics, and using the state of charge change trend curve as a time series change curve;

[0030] Performing piecewise linear regression on the time series change curve to obtain piecewise linear fitting parameters; wherein the piecewise linear fitting parameters include a slope and an intercept;

[0031] Using a support vector regression algorithm, the remaining service life of the battery pack is predicted based on the piecewise linear fitting parameters to obtain a predicted value of the number of remaining cycles;

[0032] Correcting the predicted value of the number of remaining cycles by using a particle swarm optimization algorithm to obtain a corrected number of remaining cycles;

[0033] Obtain historical usage data of the battery pack, and comprehensively evaluate the remaining service life of the battery pack based on the historical usage data and the corrected remaining number of cycles to obtain a remaining service life evaluation result; wherein the remaining service life evaluation result includes the remaining number of cycles, the remaining cumulative number of charge and discharge times, and the remaining usage time.

[0034] Further, a health status analysis is performed on the battery pack based on the remaining service life assessment to obtain a health status report of the battery pack; wherein the health status report includes electrode aging, impedance increase, and battery capacity attenuation, including:

[0035] Performing curve fitting on the remaining service life assessment result by a preset nonlinear least square method to obtain a battery pack capacity decay curve, wherein the capacity decay curve includes an exponential decay stage and a linear decay stage;

[0036] Performing electrode aging analysis on the battery pack based on the capacity decay curve to obtain the electrode aging degree, wherein the electrode aging degree includes the degree of structural change of the positive electrode material and the degree of growth of the negative electrode SEI film;

[0037] Measuring the impedance of the battery pack by electrochemical impedance spectroscopy to obtain a Nyquist diagram of the battery pack, wherein the Nyquist diagram includes a high-frequency region, a medium-frequency region, and a low-frequency region;

[0038] Performing equivalent circuit modeling on the battery pack based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes ohmic internal resistance, charge transfer impedance and diffusion impedance;

[0039] Performing parameter identification on the equivalent circuit model to obtain impedance parameters of the battery pack, wherein the impedance parameters include ohmic internal resistance, charge transfer impedance and diffusion impedance;

[0040] The health status of the battery pack is evaluated based on the electrode aging degree, capacity decay curve and impedance parameters to obtain a health status report of the battery pack, wherein the health status report includes the electrode aging degree, impedance increase amplitude and battery capacity decay rate.

[0041] Furthermore, an equivalent circuit modeling of the battery pack is performed based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack to obtain an equivalent circuit model of the battery pack, including:

[0042] Performing frequency domain analysis on the Nyquist diagram by Fourier transform to obtain a frequency response characteristic of the battery pack, wherein the frequency response characteristic includes an amplitude-frequency characteristic and a phase-frequency characteristic;

[0043] Performing complex impedance calculation on the battery pack based on the frequency response characteristic to obtain a complex impedance value of the battery pack, wherein the complex impedance value includes a real part and an imaginary part;

[0044] Performing topological analysis on the internal connection relationship data of the battery pack by a preset graph theory algorithm to obtain a topological structure matrix of the battery pack, wherein the topological structure matrix includes node relationships and edge weights, the node relationships are the connection modes between battery cells, and the edge weights are the connection impedances between battery cells;

[0045] constructing a dynamic equation of the battery pack based on the complex impedance value and the topology matrix by a state space method;

[0046] Performing circuit model simulation based on the dynamic equation to obtain a state space model of the battery pack, wherein the state space model includes a state equation and an output equation;

[0047] Estimating circuit parameters of the state space model by using a Kalman filter algorithm to obtain equivalent circuit parameters of the battery pack, wherein the equivalent circuit parameters include resistance, capacitance and inductance values;

[0048] By using the Bond diagram method, a multi-physical field coupling model of the battery pack is constructed based on the equivalent circuit parameters to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes coupling relationships among electrical, thermal and chemical reactions.

[0049] Further, formulating a management strategy for the battery pack based on the health status report includes:

[0050] Extracting features from the health status report to obtain performance indicators of the battery pack, wherein the performance indicators include internal resistance growth rate, capacity attenuation rate, and charge and discharge efficiency;

[0051] Performing hierarchical cluster analysis on the battery pack based on the performance indicators to obtain a performance grade classification result of the battery pack, wherein the performance grade classification result includes four grades: excellent, good, fair, and to be replaced;

[0052] The performance level classification results are quantitatively scored using a fuzzy comprehensive evaluation method to obtain a comprehensive performance score of the battery pack, wherein the comprehensive performance score includes a single cell score and a whole battery pack score;

[0053] Performing correlation analysis on the comprehensive performance score by grey correlation analysis method to obtain factors affecting battery pack performance, wherein the factors affecting battery pack performance include environmental humidity, discharge depth, charge and discharge cycle and temperature fluctuation;

[0054] By using a preset Markov chain Monte Carlo method, a probability prediction of the future performance of the battery pack is made based on the factors affecting the performance of the battery pack to obtain a performance attenuation trend of the battery pack;

[0055] The battery pack is optimized and analyzed based on the performance degradation trend to specify a corresponding battery pack management strategy; wherein the battery pack management strategy includes a battery pack use strategy and a battery pack maintenance strategy.

[0056] The present invention also includes a BMS testing device, comprising:

[0057] The acquisition module is used to acquire data from the BMS system of the battery pack to obtain acquired data, and pre-process the acquired data to obtain pre-processed data; wherein the acquired data includes voltage data, current data and temperature data;

[0058] An estimation module, configured to estimate the state of charge of the battery pack based on the preprocessed data to obtain an estimated value of the state of charge;

[0059] An evaluation module, configured to construct a time series variation curve based on the charge state estimation value, and to evaluate the remaining service life of the battery pack based on the time series variation curve to obtain a remaining service life evaluation result;

[0060] An analysis module, configured to perform a health status analysis on the battery pack based on the remaining service life assessment to obtain a health status report of the battery pack;

[0061] A formulation module is used to formulate a management strategy for the battery pack based on the health status report.

[0062] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0063] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0064] A BMS testing method provided by the present invention comprises the following steps: collecting data from a BMS system of a battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data; estimating the state of charge of the battery pack based on the preprocessed data to obtain a state of charge estimation value; constructing a time series change curve based on the state of charge estimation value, and evaluating the remaining service life of the battery pack based on the time series change curve to obtain a remaining service life evaluation result; analyzing the health status of the battery pack based on the remaining service life evaluation to obtain a health status report of the battery pack; formulating a management strategy for the battery pack based on the health status report; solving the problem of insufficient accuracy often faced by the complex chemical reactions and dynamic changes of the battery and the simple reliance on these data for battery status evaluation, and realizing the formulation of a more scientific and reasonable battery management strategy based on the health status report, which is helpful to improve the overall management efficiency of the battery, extend the service life and improve the safety and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic diagram of the steps of a BMS testing method in one embodiment of the present invention;

[0066] Figure 2 is a structural block diagram of a BMS testing device in one embodiment of the present invention;

[0067] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0068] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0070] like Figure 1 As shown, Figure 1 This is a schematic diagram of the steps of a BMS testing method in one embodiment of the present invention;

[0071] An embodiment of the present invention provides a BMS testing method, comprising the following steps:

[0072] Step S1, collecting data from a BMS system of a battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data; wherein the collected data includes voltage data, current data and temperature data.

[0073] Specifically, firstly, the BMS system of the battery pack collects data, which means collecting real-time voltage data, current data and temperature data from various monitoring points in the battery pack. These data are a direct reflection of the operating status of the battery pack and are the basis for subsequent analysis and evaluation. For example, in electric vehicles, the BMS system will continuously monitor the voltage of the battery to ensure that it is within a safe operating range. It will also record current data to determine the charging and discharging conditions of the battery and measure the temperature of the battery to prevent overheating. After the data collection is completed, the acquired data usually contains a large amount of original information, which may have noise, errors or inconsistent formats. Therefore, these acquired data need to be preprocessed. The preprocessing process includes data cleaning, denoising, normalization and other operations to ensure the accuracy and consistency of the data. Specifically, the voltage data needs to be filtered to remove instantaneous fluctuations caused by interference; the current data may need to be smoothed to eliminate high-frequency noise in the measurement; and the temperature data needs to be calibrated to correct the system error of the sensor. For example, if the collected voltage data has frequent small fluctuations, this may be caused by electrical noise or sensor error. During preprocessing, a moving average algorithm can be applied to smooth the data so that the final data can more truly reflect the actual voltage level of the battery pack. Similarly, preprocessing of temperature data may involve uniform calibration of temperature readings of different sensors to ensure consistency of readings of all sensors. In short, the purpose of data collection and preprocessing is to ensure that the results of subsequent battery state estimation, life assessment, and health status analysis based on these data are more accurate and reliable. Through accurate preprocessing, the data quality can be effectively improved, thereby improving the overall BMS system's monitoring capabilities and analysis accuracy of the battery pack status.

[0074] Step S2, estimating the state of charge of the battery pack based on the preprocessed data to obtain a state of charge estimation value.

[0075] Specifically, in this step, the state of charge of the battery pack is estimated based on the previously preprocessed voltage data, current data and temperature data. The so-called state of charge (SOC) is an important indicator for measuring the remaining power of the battery, which reflects the ratio of the actual power of the battery to its full power. Usually, the estimation of SOC is achieved through model calculation. In the above scheme, the preprocessed data is used, and these data are more accurate and consistent after noise removal and format unification, laying the foundation for the accurate estimation of SOC. Specifically, SOC estimation needs to combine voltage and current data and use a certain algorithm model to calculate the state of charge of the battery. For example, a common algorithm is a model based on coulomb counting. Coulomb counting calculates the power consumed by the battery from the full state to the current moment by integrating the discharge current of the battery, thereby obtaining the current SOC value. However, the actual performance of the battery will change due to temperature and aging, so on the basis of coulomb counting, it is also necessary to refer to temperature data for correction. Temperature data can help adjust the charging and discharging efficiency of the battery to avoid errors caused by excessively high or low temperatures. For example, suppose that in the BMS system of an electric vehicle, the current pre-processed voltage data is 3.8V, the current data is 1A, and the temperature data is 25 degrees Celsius. Through model calculation, the current SOC value of the battery is estimated to be 75%, that is, the battery pack still has about three-quarters of the power left for use. In this process, current data is used to calculate the actual power change, voltage data is used to determine the current potential of the battery, and temperature data is used to correct the impact of ambient temperature on battery performance, so as to make the SOC estimation value more accurate. Therefore, SOC estimation based on pre-processed data is not only an important part of battery pack status monitoring, but also the core input for subsequent battery life assessment and health status analysis, ensuring the scientificity and rationality of the battery management strategy.

[0076] Step S3, constructing a time series variation curve based on the state of charge estimation value, and evaluating the remaining service life of the battery pack based on the time series variation curve to obtain a remaining service life evaluation result.

[0077] Specifically, in this step, a time series change curve is first constructed based on the state of charge estimate. The state of charge (SOC) estimate represents the current power level of the battery pack. By recording and analyzing the changes of these estimates at different time points, a time series change curve can be obtained. The time series change curve is a graphical representation of the SOC value changing over time, which can show the power change trend and pattern of the battery pack during operation. The process of constructing this curve usually involves arranging the SOC estimates in consecutive time periods and connecting these data points into a curve using interpolation or smoothing techniques to better reflect the power change trend of the battery. Then, based on the constructed time series change curve, the remaining service life of the battery pack is evaluated. The core of this step is to infer the remaining service life of the battery pack by analyzing the change trend of the SOC value over time. The remaining service life evaluation of the battery pack usually relies on battery attenuation models, which can predict the future performance of the battery under current working conditions based on the characteristics of the time series change curve. For example, if the time series change curve shows that the SOC value gradually decreases over time and the rate of decrease accelerates, this may indicate that the battery is decaying at an increasing rate, thereby affecting the prediction of the remaining service life. For example, in a battery management system of an electric vehicle, it is assumed that the SOC value data of the battery pack in the past month is obtained through the state of charge estimation. These data are used to draw a curve of the SOC value changing over time. If the curve shows that the SOC value is continuously decreasing, and the magnitude of the decrease increases over time, then based on this time series change curve, the system can predict the remaining service life of the battery pack. If the curve shows that the SOC value of the battery has gradually decreased in the past two months, and the rate of decrease is accelerating, then the system will estimate that the remaining service life of the battery pack may be short, and generate the corresponding remaining service life assessment results. This method provides dynamic monitoring of the battery health status through the time series change curve, and can provide a more accurate remaining service life prediction based on the actual use and degradation trend of the battery, thereby helping users take necessary maintenance measures or replace the battery before the battery performance deteriorates.

[0078] Step S4, performing a health status analysis on the battery pack based on the remaining service life assessment to obtain a health status report of the battery pack;

[0079] Specifically, "In this step, a health status analysis of the battery pack is performed based on the remaining service life assessment. First, by evaluating the remaining service life of the battery pack, important data about the health status of the battery can be obtained. The remaining service life assessment results show the power decay and expected remaining working time of the battery pack under the current working conditions, and this information can be used to further analyze the health status of the battery. Specifically, the health status analysis of the battery pack is achieved by comparing the remaining service life assessment results with the expected performance standards of the battery. This analysis process involves evaluating the overall degradation level of the battery, performance stability, and deviations from manufacturer or design specifications. For example, if the remaining service life assessment results show that the battery pack's power is rapidly declining and the predicted remaining usage time is significantly lower than expected, it can be determined that the health status of the battery pack has significantly deteriorated. In this case, the battery pack may have a large performance degradation and may need to be repaired or replaced. Health status analysis usually takes into account multiple factors, including the number of charge and discharge cycles of the battery, temperature The impact on battery performance, as well as the potential impact of other environmental factors on battery life. By combining this information, a health status report for the battery pack can be generated, which usually includes an overall health assessment of the battery, an analysis of the main degradation factors, and recommended maintenance or replacement measures. For example, in the battery management system of an electric vehicle, suppose that through the remaining service life assessment, the system detects that the battery pack's power is decreasing at an accelerated rate, and the remaining use time is much lower than the design expectation. Based on these evaluation results, the health status analysis will compare the actual performance of the battery with the design specifications to analyze whether the battery pack has abnormal degradation or potential failures. Ultimately, the generated health status report may indicate that the health status of the battery pack is "needing attention" or "needs replacement", and provide specific maintenance recommendations or preventive measures to ensure the normal use of the battery pack and the safety of the system. Therefore, this process ensures a comprehensive assessment of the health status of the battery pack and provides a clear report to help users understand the actual operating status of the battery pack and take corresponding management measures to optimize the use of the battery.

[0080] Step S5: formulating a management strategy for the battery pack based on the health status report.

[0081] Specifically, in this step, a management strategy for the battery pack is formulated based on the health status report. The health status report provides the current health status, performance degradation degree and potential problems of the battery pack, which is an important basis for formulating the battery management strategy. The formulation process of the management strategy needs to comprehensively consider the actual health status of the battery to ensure that the battery pack can operate safely and efficiently within its remaining service life and extend the service life of the battery as much as possible. Specifically, the management strategy may include optimizing the charging and discharging process of the battery pack to avoid overcharging or over-discharging, which are all formulated based on the battery health problems reflected in the health status report. For example, if the report shows that some cells of the battery pack have overheating or voltage imbalance problems, the management strategy may adjust the charging current of these cells, or take equalization charging measures when necessary. In addition, the management strategy may also involve adjusting the battery usage environment, such as taking additional cooling measures to prevent the battery from overheating when the temperature is high, thereby protecting the health of the battery. The formulation of the management strategy should also take into account the expected working conditions and application scenarios of the battery pack. For example, in an industrial application with frequent use and harsh environment, the management strategy may be more stringent, including more frequent health status monitoring, more cautious charging and discharging control, and preventive replacement of battery cells. In contrast, in a mild environment, the strategy may focus more on extending the life of the battery rather than replacing it frequently. For example, in an electric vehicle, suppose the health status report shows that some cells of the battery pack have shown significant performance degradation due to long-term exposure to high temperature environments. In this case, the management strategy may include limiting the maximum charging voltage of these cells and performing more frequent temperature monitoring of the battery pack in future use. If the health of the battery pack deteriorates further, the strategy may recommend battery replacement or repair after a specified number of mileage to ensure safe operation of the vehicle. Through management strategies based on health status reports, the battery pack can be better protected during its life cycle, reduce the occurrence of unexpected failures, and ensure stable performance under various usage conditions. This strategy not only helps to extend the service life of the battery, but also improves the reliability and safety of the entire system.

[0082] In a specific embodiment, the data collection of the BMS system of the battery pack to obtain the collected data, and the preprocessing of the collected data to obtain the preprocessed data include:

[0083] The BMS system of the battery pack is subjected to high-resolution data collection through preset sensors to obtain collected data; wherein the preset sensors include electrical sensors and temperature sensors;

[0084] The Hilbert-Huang transform technique is used to perform time-frequency analysis on the collected data to obtain multi-scale time-frequency analysis data;

[0085] Performing time series anomaly detection on the multi-scale time-frequency analysis data to obtain an anomaly detection result;

[0086] If the abnormal detection result is that the multi-scale time-frequency analysis data has abnormal data, the abnormal data is marked to form a marked data set; wherein the abnormal data includes abnormal voltage fluctuations, abnormal current spikes, and abnormal temperature changes;

[0087] Performing spatial interpolation and repair on the labeled data set by using a preset Kriging interpolation algorithm to obtain a repaired data set;

[0088] The repair data set is normalized to obtain preprocessed data.

[0089] Specifically, in this step, data acquisition is performed on the BMS system of the battery pack to obtain the collected data, and the data are preprocessed to obtain preprocessed data. First, high-resolution data acquisition is performed on the BMS system of the battery pack through preset sensors, including electrical sensors and temperature sensors. These sensors can provide accurate voltage, current and temperature data, which provide the basis for subsequent analysis. Subsequently, the Hilbert-Huang transform technology is used to perform time-frequency analysis on the collected data. This technology can analyze the data in two dimensions of time and frequency to generate multi-scale time-frequency analysis data. The Hilbert-Huang transform decomposes the non-stationary signal and extracts the frequency information at different time scales, thereby revealing the time-varying characteristics and potential abnormal patterns in the data. Next, the multi-scale time-frequency analysis data is subjected to time series anomaly detection to obtain anomaly detection results. If abnormal data is detected in the multi-scale time-frequency analysis data, these abnormal data need to be annotated to form a labeled data set. These abnormal data include abnormal voltage fluctuations, abnormal current spikes, and abnormal temperature changes, which may indicate potential problems or failures in the battery pack. After the labeled data set is formed, the preset Kriging interpolation algorithm is used to spatially interpolate and repair these labeled data. Kriging interpolation is a statistical interpolation method that can estimate and complete missing or abnormal data in space, thereby repairing outliers in the data set and making the data more complete and accurate. After the repaired data set is spatially interpolated, the repaired data set obtained contains the processed and completed data points. Finally, the repaired data set is normalized to obtain preprocessed data. Normalization standardizes the data to a uniform range, making the data more consistent and comparable in subsequent analysis. For example, scaling the data to a range of 0 to 1 eliminates the dimensional differences and scale effects in the original data, making the data more suitable for further analysis and modeling. For example, assume that the BMS system of the battery pack collects a large amount of voltage, current and temperature data during operation, which may contain some abnormal voltage fluctuations and temperature changes. After the Hilbert-Huang transform is used to perform time-frequency analysis on the data, these abnormal data can be identified and these outliers can be repaired through Kriging interpolation. The normalized data will provide a more accurate and reliable basis for battery pack health status analysis and performance evaluation.

[0090] In a specific embodiment, estimating the state of charge of the battery pack based on the preprocessed data to obtain an estimated state of charge value includes:

[0091] Performing data fusion on the preprocessed data to obtain fused data;

[0092] Performing nonlinear observation on the state of charge of the battery pack based on the fused data to obtain a preliminary state of charge distribution;

[0093] Predicting the real-time state of the battery pack based on the preliminary state of charge distribution to obtain a preliminary state of charge estimation value;

[0094] Performing state calibration on the preliminary state-of-charge estimate to obtain a calibrated state-of-charge estimate;

[0095] Through a preset multi-task learning algorithm, a multi-dimensional analysis is performed on various states of the battery pack based on the calibrated state of charge estimation value to obtain a multi-dimensional analysis result; wherein the multi-dimensional analysis result includes the impact of the charging and discharging efficiency, power output capacity and single cell consistency on the performance of the entire battery pack;

[0096] Acquire internal connection relationship data and topological structure of the battery pack, and obtain energy consumption data of the battery pack based on the internal connection relationship data and topological structure and multi-dimensional analysis results;

[0097] The calibrated state of charge estimate is estimated based on the energy consumption data to obtain a state of charge estimate.

[0098] Specifically, in this step, the state of charge of the battery pack is estimated based on the preprocessed data to obtain the estimated state of charge value. First, the fused data is obtained by performing data fusion on the preprocessed data. The data fusion process involves integrating data from different sensors (such as data from electrical sensors and temperature sensors) to form a comprehensive data set, which contains complete information about the battery pack under different conditions and can provide a more comprehensive view of the battery state. Next, the state of charge of the battery pack is nonlinearly observed based on the fused data to obtain a preliminary state of charge distribution. Nonlinear observation refers to the use of complex mathematical models or algorithms to process the fused data and extract a preliminary estimated distribution of the state of charge of the battery pack. This distribution represents the state of charge distribution of the battery pack under current conditions, but may require further adjustment and optimization. After obtaining the preliminary state of charge distribution, the real-time state of the battery pack is predicted based on this distribution to obtain a preliminary state of charge estimate. This process involves applying the preliminary state of charge distribution to the real-time data to predict the current state of charge of the battery pack. The prediction result can provide an estimate of the current power level of the battery pack, but in order to improve accuracy, further state calibration is required. The preliminary state of charge estimate is calibrated to obtain a calibrated state of charge estimate. State calibration is to adjust and optimize the preliminary state of charge estimate by comparing it with the actual measurement value or other reference value to reduce the prediction error and obtain a more accurate state of charge estimate. The calibration process may include adjusting model parameters, applying correction algorithms, etc. to ensure the accuracy of the estimate. Subsequently, a multi-dimensional analysis is performed on the various states of the battery pack based on the calibrated state of charge estimate through a preset multi-task learning algorithm to obtain a multi-dimensional analysis result. The multi-task learning algorithm can handle multiple tasks at the same time, such as analyzing the charge and discharge efficiency, power output capacity and single cell consistency of the battery pack. Through these analyses, the impact of these factors on the performance of the entire battery pack can be evaluated, thereby providing an in-depth understanding of the overall performance of the battery pack. For example, the analysis may reveal the change in the charge and discharge efficiency of the battery pack under different loads, and the impact of the consistency of different single cells on the overall performance of the battery pack. After obtaining the internal connection relationship data and topological structure of the battery pack, the energy consumption data of the battery pack is obtained based on these data and the multi-dimensional analysis results. The internal connection relationship data describes how the individual battery cells in the battery pack are connected, while the topology describes the overall layout and structure of the battery pack. By combining this information with the results of multi-dimensional analysis, the energy consumption of each part of the battery pack in actual use can be calculated to obtain energy consumption data. Finally, the calibrated state of charge estimate is further estimated based on the energy consumption data to obtain the final state of charge estimate. This step involves applying the energy consumption data to the calculation model of the calibrated state of charge estimate to comprehensively consider the impact of energy consumption on the battery pack state of charge, thereby obtaining a more accurate state of charge estimate.This estimate can reflect the power level of the battery pack in actual operation, provide accurate battery status information for the battery management system, and support the optimization of the battery pack operation and management strategy. For example, in the battery management system of an electric vehicle, the voltage, current and temperature data of the battery are integrated through data fusion, and the preliminary state of charge distribution is obtained using a nonlinear observation model. Then, the preliminary state of charge estimate is obtained by predicting the real-time data, and the state is calibrated to correct the estimation error. The charge and discharge efficiency and power output capacity of the battery pack are analyzed by a multi-task learning algorithm, and the energy consumption data is calculated by combining the internal connection relationship and topological structure of the battery pack. Finally, the calibrated state of charge estimate is obtained by analyzing the energy consumption data, thereby providing accurate battery pack power level information. These steps ensure that the battery management system can accurately monitor and predict the state of charge of the battery pack and optimize the use and maintenance of the battery.

[0099] In a specific embodiment, a time series variation curve is constructed based on the state of charge estimation value, and a remaining service life of the battery pack is evaluated based on the time series variation curve to obtain a remaining service life evaluation result, including:

[0100] Decomposing the state of charge estimation value in the time-frequency domain by wavelet transform to obtain multi-scale time-frequency features; wherein the multi-scale time-frequency features include low-frequency trend features and high-frequency fluctuation features;

[0101] Constructing a state of charge change trend curve based on the multi-scale time-frequency characteristics, and using the state of charge change trend curve as a time series change curve;

[0102] Performing piecewise linear regression on the time series change curve to obtain piecewise linear fitting parameters; wherein the piecewise linear fitting parameters include a slope and an intercept;

[0103] Using a support vector regression algorithm, the remaining service life of the battery pack is predicted based on the piecewise linear fitting parameters to obtain a predicted value of the number of remaining cycles;

[0104] Correcting the predicted value of the number of remaining cycles by using a particle swarm optimization algorithm to obtain a corrected number of remaining cycles;

[0105] Obtain historical usage data of the battery pack, and comprehensively evaluate the remaining service life of the battery pack based on the historical usage data and the corrected remaining number of cycles to obtain a remaining service life evaluation result; wherein the remaining service life evaluation result includes the remaining number of cycles, the remaining cumulative number of charge and discharge times, and the remaining usage time.

[0106] Specifically, in this step, a time series change curve is first constructed based on the estimated state of charge value, and the remaining service life of the battery pack is evaluated using the curve to obtain the remaining service life evaluation result. The specific operation includes the following steps. First, the estimated state of charge value is decomposed in the time-frequency domain by wavelet transform to obtain multi-scale time-frequency features. Wavelet transform is a technology that decomposes a signal into low-frequency trend features and high-frequency fluctuation features at different scales. The low-frequency trend feature reflects the long-term change trend of the state of charge, while the high-frequency fluctuation feature reveals the short-term fluctuation and noise of the state of charge. These multi-scale time-frequency features help to fully understand the change of the battery state. After obtaining the multi-scale time-frequency features, a state of charge change trend curve is constructed based on these features and used as a time series change curve. The state of charge change trend curve shows the change trend of the state of charge of the battery pack over time, which can reflect the performance degradation of the battery during use. The time series change curve provides basic data for the subsequent remaining service life evaluation. Then, a piecewise linear regression is performed on the time series change curve to obtain piecewise linear fitting parameters. Segmented linear regression is a technique that divides the time series change curve into multiple segments and fits a linear model to each segment, where the segmented linear fitting parameters include slope and intercept. The slope describes the speed of the state of charge change, while the intercept represents the initial value of the state of charge. These fitting parameters help quantify the change law of the state of charge, thereby providing a basis for the prediction of the remaining service life. Based on the segmented linear fitting parameters, the support vector regression algorithm is used to predict the remaining service life of the battery pack and obtain the remaining cycle number prediction value. Support vector regression is a technique that uses support vector machines for regression analysis. It predicts the change trend of the state of charge of the battery pack through a training model to obtain the remaining cycle number prediction value of the battery pack, that is, the number of times the battery pack can continue to charge and discharge. Subsequently, the remaining cycle number prediction value is corrected by the particle swarm optimization algorithm to obtain the corrected remaining cycle number. The particle swarm optimization algorithm is an optimization algorithm based on swarm intelligence. It optimizes the prediction results by simulating the movement of particles in the search space to reduce the prediction error and obtain a more accurate remaining cycle number. Finally, the historical usage data of the battery pack is obtained, and based on this data and the corrected remaining number of cycles, the remaining service life of the battery pack is comprehensively evaluated to obtain the remaining service life evaluation result. The historical usage data includes the battery pack's past charge and discharge records and usage environment data. These data can be combined with the corrected remaining number of cycles to comprehensively evaluate the remaining service life of the battery pack, and obtain detailed evaluation results including the remaining number of cycles, the remaining cumulative number of charge and discharge times, and the remaining usage time. For example, assuming that after prediction and correction, the remaining number of cycles of the battery pack is 500 times, the remaining cumulative number of charge and discharge times is 2000 times, and the remaining usage time is 6 months, these data can help evaluate the future performance and life of the battery pack and guide users' maintenance and replacement decisions.

[0107] In a specific embodiment, a health status analysis of the battery pack is performed based on the remaining service life assessment to obtain a health status report of the battery pack; wherein the health status report includes electrode aging, impedance increase, and battery capacity attenuation, including:

[0108] Performing curve fitting on the remaining service life assessment result by a preset nonlinear least square method to obtain a battery pack capacity decay curve, wherein the capacity decay curve includes an exponential decay stage and a linear decay stage;

[0109] Performing electrode aging analysis on the battery pack based on the capacity decay curve to obtain the electrode aging degree, wherein the electrode aging degree includes the degree of structural change of the positive electrode material and the degree of growth of the negative electrode SEI film;

[0110] Measuring the impedance of the battery pack by electrochemical impedance spectroscopy to obtain a Nyquist diagram of the battery pack, wherein the Nyquist diagram includes a high-frequency region, a medium-frequency region, and a low-frequency region;

[0111] Performing equivalent circuit modeling on the battery pack based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes ohmic internal resistance, charge transfer impedance and diffusion impedance;

[0112] Performing parameter identification on the equivalent circuit model to obtain impedance parameters of the battery pack, wherein the impedance parameters include ohmic internal resistance, charge transfer impedance and diffusion impedance;

[0113] The health status of the battery pack is evaluated based on the electrode aging degree, capacity decay curve and impedance parameters to obtain a health status report of the battery pack, wherein the health status report includes the electrode aging degree, impedance increase amplitude and battery capacity decay rate.

[0114] Specifically, in this solution, the health status of the battery pack is analyzed based on the remaining service life assessment. First, the remaining service life assessment results of the battery pack are processed. In this process, the assessment results are curve-fitted by the preset nonlinear least squares method to obtain the capacity decay curve. This capacity decay curve is an important manifestation of the health status of the battery pack, reflecting the changing trend of the battery capacity over time. The capacity decay curve can usually be divided into two stages, namely the exponential decay stage and the linear decay stage. In the exponential decay stage, the battery capacity decays rapidly, while in the linear decay stage, the decay rate is relatively slow. Therefore, the capacity decay curve can help analyze the future use of the battery pack and provide a basis for further health status assessment. Based on this capacity decay curve, the electrode aging analysis will be carried out next. The degree of electrode aging is one of the important reasons for the change in battery performance. In the electrode aging analysis, the degree of electrode aging mainly includes the degree of change in the structure of the positive electrode material and the degree of growth of the negative electrode SEI film (solid electrolyte interface film). Changes in the structure of the positive electrode material will cause the capacity of the battery to decay. In particular, during the battery charging and discharging process, the irreversible loss of the material will significantly affect the life of the battery; while the growth of the SEI film will increase the internal resistance of the battery and affect the charging and discharging efficiency. These aging phenomena gradually intensify with the long-term use of the battery. By analyzing the capacity decay curve, the degree of electrode aging can be determined and a basis for the health status report can be provided. In order to further understand the performance of the battery pack, it is also necessary to measure the impedance of the battery pack through electrochemical impedance spectroscopy technology to obtain the Nyquist plot. The Nyquist plot is a frequency response diagram that can reflect the impedance changes of the battery pack at different frequencies. The Nyquist plot usually contains high-frequency, medium-frequency and low-frequency regions, each of which corresponds to a different electrochemical process. The high-frequency region mainly reflects the ohmic internal resistance, which is the contact resistance between the electrode and the electrolyte inside the battery; the medium-frequency region is usually related to the charge transfer impedance, indicating the electrochemical reaction rate inside the battery; and the low-frequency region reflects the diffusion impedance, representing the diffusion process of charge in the electrode. Therefore, the Nyquist plot can provide important frequency domain information for the equivalent circuit modeling of the battery pack. Based on the Nyquist diagram, the internal connection relationship data of the battery pack, and the topological structure, the battery pack is modeled by equivalent circuit, and the equivalent circuit model of the battery pack can be obtained. This model contains multiple resistance components, such as ohmic internal resistance, charge transfer impedance, and diffusion impedance. The ohmic internal resistance represents the basic resistance of the battery pack, while the charge transfer impedance reflects the difficulty of the electrochemical reaction, and the diffusion impedance describes the efficiency of charge movement within the battery pack. The equivalent circuit model can accurately simulate the electrical characteristics of the battery pack and help analyze the operating status and health level of the battery pack. Parameter identification of the equivalent circuit model is a crucial step. The goal of parameter identification is to obtain the impedance parameters of the battery pack, including the ohmic internal resistance value, charge transfer impedance value, and diffusion impedance value.Through these impedance parameters, the change of the internal resistance of the battery pack can be quantitatively evaluated, and the performance degradation of the battery pack can be further revealed. These parameters play an important role in the evaluation of the health status of the battery pack, because the increase of impedance is usually an important sign of battery aging and performance degradation. As the battery pack is used for a longer time, the change of impedance parameters can help predict the remaining life of the battery. Finally, based on the electrode aging degree, capacity decay curve and impedance parameters, the health status of the battery pack is comprehensively evaluated to obtain the health status report of the battery pack. This report will include key information such as the electrode aging degree, impedance increase amplitude and battery capacity decay rate. Through this report, you can fully understand the current health status of the battery pack and provide data support for future use and maintenance decisions. For example, if the report shows that the battery capacity decay rate is high and the impedance increase amplitude is significant, it may indicate that the battery pack is close to the limit of its service life and needs to be maintained or replaced as soon as possible. Through the analysis of the health status report, an optimized use strategy can be formulated to extend the service life of the battery pack and improve its overall operating efficiency. For example, a battery pack's health status report shows that its capacity decay curve is in the linear decay stage, and the electrode aging analysis shows that the positive electrode material structure has undergone significant irreversible changes. At the same time, the electrochemical impedance measurement results show that both the ohmic internal resistance and the charge transfer impedance have increased significantly. This information indicates that the performance of the battery pack has significantly declined. Through further Markov chain Monte Carlo prediction methods, it can be predicted that the performance of the battery pack will continue to deteriorate in the future. Therefore, based on this health status report, managers can decide to replace the battery pack in advance or strengthen its maintenance.

[0115] In a specific embodiment, an equivalent circuit modeling of the battery pack is performed based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack to obtain an equivalent circuit model of the battery pack, including:

[0116] Performing frequency domain analysis on the Nyquist diagram by Fourier transform to obtain a frequency response characteristic of the battery pack, wherein the frequency response characteristic includes an amplitude-frequency characteristic and a phase-frequency characteristic;

[0117] Performing complex impedance calculation on the battery pack based on the frequency response characteristic to obtain a complex impedance value of the battery pack, wherein the complex impedance value includes a real part and an imaginary part;

[0118] Performing topological analysis on the internal connection relationship data of the battery pack by a preset graph theory algorithm to obtain a topological structure matrix of the battery pack, wherein the topological structure matrix includes node relationships and edge weights, the node relationships are the connection modes between battery cells, and the edge weights are the connection impedances between battery cells;

[0119] constructing a dynamic equation of the battery pack based on the complex impedance value and the topology matrix by a state space method;

[0120] Performing circuit model simulation based on the dynamic equation to obtain a state space model of the battery pack, wherein the state space model includes a state equation and an output equation;

[0121] Estimating circuit parameters of the state space model by using a Kalman filter algorithm to obtain equivalent circuit parameters of the battery pack, wherein the equivalent circuit parameters include resistance, capacitance and inductance values;

[0122] By using the Bond diagram method, a multi-physical field coupling model of the battery pack is constructed based on the equivalent circuit parameters to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes coupling relationships among electrical, thermal and chemical reactions.

[0123] Specifically, in this step, based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack, the battery pack is modeled as an equivalent circuit to obtain an equivalent circuit model of the battery pack. First, the Nyquist diagram is analyzed in the frequency domain by Fourier transform to obtain the frequency response characteristics of the battery pack. These frequency response characteristics include amplitude-frequency characteristics and phase-frequency characteristics, where the amplitude-frequency characteristics represent the amplitude response of the battery pack to signals of different frequencies, and the phase-frequency characteristics describe the change in phase. These characteristics help to understand the electrical behavior of the battery pack at different frequencies. After obtaining the frequency response characteristics, the complex impedance of the battery pack is calculated based on these characteristics to obtain the complex impedance value of the battery pack. The complex impedance value includes a real part and an imaginary part, where the real part represents the resistance of the battery pack and the imaginary part represents the reactance of the battery pack. The complex impedance value provides the comprehensive impedance characteristics of the battery pack under an AC signal, reflecting its electrical performance. Next, the internal connection relationship data of the battery pack is topologically analyzed by a preset graph theory algorithm to obtain a topological structure matrix of the battery pack. The topological structure matrix includes node relationships and edge weights, where the node relationships describe the connection mode between battery cells and the edge weights represent the connection impedance between battery cells. This information helps to build a structural model of the battery pack, taking into account the electrical connection between battery cells and their impedance. Subsequently, the dynamic equations of the battery pack are constructed based on the complex impedance values ​​and the topological structure matrix through the state space method. The state space method describes the dynamic behavior of the system by establishing state equations and output equations. The state equation is used to describe the state changes inside the battery pack, while the output equation describes the output response of the system. The constructed dynamic equation can fully reflect the behavior of the battery pack under dynamic conditions. The circuit model is simulated based on the dynamic equations to obtain the state space model of the battery pack. The state space model includes state equations and output equations, which are used to describe the response characteristics of the battery pack under different working conditions. Through simulation, the performance of the battery pack in actual use can be predicted and its performance can be optimized. The circuit parameters of the state space model are estimated through the Kalman filter algorithm to obtain the equivalent circuit parameters of the battery pack. The equivalent circuit parameters include resistance, capacitance and inductance values, which are used to characterize the electrical characteristics of the battery pack and help optimize its performance. Finally, the multi-physics coupling model of the battery pack is constructed based on the equivalent circuit parameters through the Bond diagram method, thereby obtaining the equivalent circuit model of the battery pack. The equivalent circuit model includes the coupling relationship of electrical, thermal and chemical reactions, and fully considers the coupling effects of multiple physical fields involved in the actual work of the battery pack. This model can more accurately describe the comprehensive performance of the battery pack, including electrical behavior, thermal management and chemical reactions, and provide a scientific basis for the design and optimization of the battery pack. For example, in the development process of the battery pack, the frequency response characteristics of the battery pack are obtained through Nyquist diagram analysis, and then the complex impedance value is calculated, and the connection relationship between the battery cells is analyzed through graph theory algorithms.Based on this information, dynamic equations were constructed and circuit model simulation was performed to obtain the state space model of the battery pack. The parameters of the model were estimated using the Kalman filter algorithm to obtain the resistance, capacitance and inductance values. Finally, these parameters were used to construct a multi-physics coupling model of the battery pack using the Bond diagram method, taking into account the coupling relationship between electrical, thermal and chemical reactions, thereby obtaining an accurate equivalent circuit model.

[0124] In a specific embodiment, formulating a management strategy for the battery pack based on the health status report includes:

[0125] Extracting features from the health status report to obtain performance indicators of the battery pack, wherein the performance indicators include internal resistance growth rate, capacity attenuation rate, and charge and discharge efficiency;

[0126] Performing hierarchical cluster analysis on the battery pack based on the performance indicators to obtain a performance grade classification result of the battery pack, wherein the performance grade classification result includes four grades: excellent, good, fair, and to be replaced;

[0127] The performance level classification results are quantitatively scored using a fuzzy comprehensive evaluation method to obtain a comprehensive performance score of the battery pack, wherein the comprehensive performance score includes a single cell score and a whole battery pack score;

[0128] Performing correlation analysis on the comprehensive performance score by grey correlation analysis method to obtain factors affecting battery pack performance, wherein the factors affecting battery pack performance include environmental humidity, discharge depth, charge and discharge cycle and temperature fluctuation;

[0129] By using a preset Markov chain Monte Carlo method, a probability prediction of the future performance of the battery pack is made based on the factors affecting the performance of the battery pack to obtain a performance attenuation trend of the battery pack;

[0130] The battery pack is optimized and analyzed based on the performance degradation trend to specify a corresponding battery pack management strategy; wherein the battery pack management strategy includes a battery pack use strategy and a battery pack maintenance strategy.

[0131] Specifically, in this step, the process of formulating a management strategy for the battery pack based on the health status report is as follows. First, feature extraction of the health status report is the key first step, and this process aims to extract the performance indicators of the battery pack from the health status report. The performance indicators include internal resistance growth rate, capacity decay rate, and charge and discharge efficiency. The internal resistance growth rate is used to measure the increase in the internal resistance of the battery during use, the capacity decay rate reflects the degree of reduction of the battery capacity with the use time, and the charge and discharge efficiency describes the efficiency of energy conversion of the battery during the charge and discharge process. These performance indicators can comprehensively reflect the overall health status of the battery pack and its performance changes. After obtaining these performance indicators, the next step is a hierarchical clustering analysis based on performance indicators. This analysis method can classify the battery pack according to the performance indicators and obtain the performance level classification results of the battery pack. The performance level classification results include four levels: excellent, good, general, and to be replaced. Through hierarchical clustering analysis, the battery pack can be divided into different levels according to its performance level, so as to understand the performance of each battery group and determine which battery packs can still maintain a good state and which need to be paid attention to or replaced. After obtaining the performance level classification results, the fuzzy comprehensive evaluation method is used to quantify these results and obtain the comprehensive performance score of the battery pack. The comprehensive performance score includes the score of a single cell and the score of the entire battery pack. The fuzzy comprehensive evaluation method can quantitatively evaluate the performance of the battery pack by defining fuzzy sets and weight coefficients, making the evaluation results more accurate and reasonable. These scores can comprehensively reflect the performance of the battery pack in various performance indicators and help judge its overall state. Subsequently, the comprehensive performance score is correlated by the gray correlation analysis method to obtain the factors affecting the performance of the battery pack. The factors affecting the performance of the battery pack include environmental humidity, discharge depth, charge and discharge cycle and temperature fluctuation. The gray correlation analysis method can reveal how these factors affect the performance of the battery pack by calculating the correlation between each factor and the performance of the battery pack, providing a basis for subsequent optimization analysis. For example, finding the degree of influence of environmental humidity and temperature fluctuations on battery performance can help determine the use strategy of the battery pack under different environmental conditions. After identifying the factors affecting the performance of the battery pack, the preset Markov chain Monte Carlo method is used to probabilistically predict the future performance of the battery pack to obtain the performance attenuation trend of the battery pack. The Markov Chain Monte Carlo method is a statistical method used to simulate and predict the behavior of complex systems. It can predict the future performance degradation of battery packs by simulating the performance changes of battery packs under different factors. This prediction result can provide the performance trend of the battery pack in future use and help evaluate its long-term performance. Based on the performance degradation trend, the battery pack is optimized and analyzed to formulate corresponding management strategies. The optimization analysis can evaluate the predicted performance degradation trend and formulate corresponding battery pack usage and maintenance strategies.For example, if the prediction results show that the performance of the battery pack decays rapidly in a high temperature environment, it may be recommended to reduce the frequency of use of the battery pack or strengthen cooling measures in a high temperature environment. At the same time, the maintenance strategy may include regular inspection of the performance of the battery pack and timely replacement of the battery cells with degraded performance to ensure the stability and reliability of the entire battery pack. Ultimately, these management strategies can effectively extend the service life of the battery pack and improve its operating efficiency. For example, in actual applications, if the health status report shows that the capacity decay rate of the battery pack is high and the internal resistance growth rate has also increased, then the battery pack is classified into "general" or "to be replaced" levels through hierarchical clustering analysis, combined with the comprehensive performance score obtained by the fuzzy comprehensive evaluation method, it may be found that the overall performance of the battery pack is poor. In this case, the gray correlation analysis method may reveal the significant impact of ambient humidity and charge and discharge cycles on battery performance. The prediction results of the Markov chain Monte Carlo method show that the performance of the battery pack will decay significantly in the future. Therefore, the optimization analysis may recommend minimizing high humidity and long-cycle operations in the next use, and strengthening maintenance measures to ensure the stability and reliability of the battery pack in actual applications.

[0132] A BMS testing method in an embodiment of the present invention is described above. A BMS testing device in an embodiment of the present invention is described below. Figure 2 In one embodiment of the present invention, a BMS testing device includes:

[0133] The acquisition module 21 is used to acquire data from the BMS system of the battery pack to obtain acquired data, and pre-process the acquired data to obtain pre-processed data; wherein the acquired data includes voltage data, current data and temperature data;

[0134] An estimation module 22, configured to estimate the state of charge of the battery pack based on the preprocessed data to obtain an estimated value of the state of charge;

[0135] An evaluation module 23, configured to construct a time series variation curve based on the charge state estimation value, and perform a remaining service life evaluation on the battery pack based on the time series variation curve to obtain a remaining service life evaluation result;

[0136] An analysis module 24, configured to perform a health status analysis on the battery pack based on the remaining service life assessment to obtain a health status report of the battery pack;

[0137] The formulation module 25 is used to formulate a management strategy for the battery pack based on the health status report.

[0138] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0139] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0140] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0141] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0142] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0143] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0144] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A BMS testing method, characterized in that: The following steps are involved: Performing data collection on a BMS system of a battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data; wherein the collected data includes voltage data, current data and temperature data; estimating the state of charge of the battery pack based on the preprocessed data to obtain a state of charge estimation value; Constructing a time series variation curve based on the estimated state of charge value, and evaluating the remaining service life of the battery pack based on the time series variation curve to obtain a remaining service life evaluation result; Performing a health status analysis on the battery pack based on the remaining service life assessment result to obtain a health status report of the battery pack; formulating a management strategy for the battery pack based on the health status report; A time series variation curve is constructed based on the estimated state of charge value, and a remaining service life evaluation is performed on the battery pack based on the time series variation curve to obtain a remaining service life evaluation result, including: Decomposing the state of charge estimation value in the time-frequency domain by wavelet transform to obtain multi-scale time-frequency features; wherein the multi-scale time-frequency features include low-frequency trend features and high-frequency fluctuation features; Constructing a state of charge change trend curve based on the multi-scale time-frequency characteristics, and using the state of charge change trend curve as a time series change curve; Performing piecewise linear regression on the time series change curve to obtain piecewise linear fitting parameters; wherein the piecewise linear fitting parameters include a slope and an intercept; Using a support vector regression algorithm, the remaining service life of the battery pack is predicted based on the piecewise linear fitting parameters to obtain a predicted value of the number of remaining cycles; Correcting the predicted value of the number of remaining cycles by using a particle swarm optimization algorithm to obtain a corrected number of remaining cycles; Obtain historical usage data of the battery pack, and comprehensively evaluate the remaining service life of the battery pack based on the historical usage data and the corrected remaining number of cycles to obtain a remaining service life evaluation result; wherein the remaining service life evaluation result includes the remaining number of cycles, the remaining cumulative number of charge and discharge times, and the remaining usage time.

2. The BMS testing method according to claim 1, characterized in that: The step of collecting data from the BMS system of the battery pack to obtain collected data, and preprocessing the collected data to obtain preprocessed data includes: The BMS system of the battery pack is subjected to high-resolution data collection through preset sensors to obtain collected data; wherein the preset sensors include electrical sensors and temperature sensors; The Hilbert-Huang transform technique is used to perform time-frequency analysis on the collected data to obtain multi-scale time-frequency analysis data; Performing time series anomaly detection on the multi-scale time-frequency analysis data to obtain an anomaly detection result; If the abnormal detection result is that the multi-scale time-frequency analysis data has abnormal data, the abnormal data is marked to form a marked data set; wherein the abnormal data includes abnormal voltage fluctuations, abnormal current spikes and abnormal temperature changes; Performing spatial interpolation and repair on the labeled data set by using a preset Kriging interpolation algorithm to obtain a repaired data set; The repair data set is normalized to obtain preprocessed data.

3. The BMS testing method according to claim 1, characterized in that: The state of charge of the battery pack is estimated based on the preprocessed data to obtain an estimated value of the state of charge, including: Performing data fusion on the preprocessed data to obtain fused data; Performing nonlinear observation on the state of charge of the battery pack based on the fused data to obtain a preliminary state of charge distribution; Predicting the real-time state of the battery pack based on the preliminary state of charge distribution to obtain a preliminary state of charge estimation value; Performing state calibration on the preliminary state-of-charge estimate to obtain a calibrated state-of-charge estimate; Through a preset multi-task learning algorithm, a multi-dimensional analysis is performed on various states of the battery pack based on the calibrated state of charge estimation value to obtain a multi-dimensional analysis result; wherein the multi-dimensional analysis result includes the impact of the charging and discharging efficiency, power output capacity and single cell consistency on the performance of the entire battery pack; Acquire internal connection relationship data and topological structure of the battery pack, and obtain energy consumption data of the battery pack based on the internal connection relationship data and topological structure and multi-dimensional analysis results; The calibrated state of charge estimate is estimated based on the energy consumption data to obtain a state of charge estimate.

4. The BMS testing method according to claim 3, characterized in that: Performing a health status analysis on the battery pack based on the remaining service life assessment result to obtain a health status report of the battery pack; wherein the health status report includes electrode aging, impedance increase, and battery capacity attenuation, including: Performing curve fitting on the remaining service life assessment result by a preset nonlinear least square method to obtain a battery pack capacity decay curve, wherein the capacity decay curve includes an exponential decay stage and a linear decay stage; Performing electrode aging analysis on the battery pack based on the capacity decay curve to obtain the electrode aging degree, wherein the electrode aging degree includes the degree of structural change of the positive electrode material and the degree of growth of the negative electrode SEI film; Measuring the impedance of the battery pack by electrochemical impedance spectroscopy to obtain a Nyquist diagram of the battery pack, wherein the Nyquist diagram includes a high-frequency region, a medium-frequency region, and a low-frequency region; Performing equivalent circuit modeling on the battery pack based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes ohmic internal resistance, charge transfer impedance and diffusion impedance; Performing parameter identification on the equivalent circuit model to obtain impedance parameters of the battery pack, wherein the impedance parameters include ohmic internal resistance, charge transfer impedance and diffusion impedance; The health status of the battery pack is evaluated based on the electrode aging degree, capacity decay curve and impedance parameters to obtain a health status report of the battery pack, wherein the health status report includes the electrode aging degree, impedance increase amplitude and battery capacity decay rate.

5. The BMS testing method according to claim 4, characterized in that: Based on the Nyquist diagram, the internal connection relationship data and the topological structure of the battery pack, an equivalent circuit model of the battery pack is performed to obtain an equivalent circuit model of the battery pack, including: Performing frequency domain analysis on the Nyquist diagram by Fourier transform to obtain a frequency response characteristic of the battery pack, wherein the frequency response characteristic includes an amplitude-frequency characteristic and a phase-frequency characteristic; Performing complex impedance calculation on the battery pack based on the frequency response characteristic to obtain a complex impedance value of the battery pack, wherein the complex impedance value includes a real part and an imaginary part; Performing topological analysis on the internal connection relationship data of the battery pack by a preset graph theory algorithm to obtain a topological structure matrix of the battery pack, wherein the topological structure matrix includes node relationships and edge weights, the node relationships are the connection modes between battery cells, and the edge weights are the connection impedances between battery cells; constructing a dynamic equation of the battery pack based on the complex impedance value and the topology matrix by a state space method; Performing circuit model simulation based on the dynamic equation to obtain a state space model of the battery pack, wherein the state space model includes a state equation and an output equation; Estimating circuit parameters of the state space model by using a Kalman filter algorithm to obtain equivalent circuit parameters of the battery pack, wherein the equivalent circuit parameters include resistance, capacitance and inductance values; By using the Bond diagram method, a multi-physical field coupling model of the battery pack is constructed based on the equivalent circuit parameters to obtain an equivalent circuit model of the battery pack, wherein the equivalent circuit model includes coupling relationships among electrical, thermal and chemical reactions.

6. The BMS testing method according to claim 1, characterized in that: Formulating a management strategy for the battery pack based on the health status report, including: Extracting features from the health status report to obtain performance indicators of the battery pack, wherein the performance indicators include internal resistance growth rate, capacity attenuation rate, and charge and discharge efficiency; Performing hierarchical cluster analysis on the battery pack based on the performance indicators to obtain a performance grade classification result of the battery pack, wherein the performance grade classification result includes four grades: excellent, good, fair, and to be replaced; The performance level classification results are quantitatively scored using a fuzzy comprehensive evaluation method to obtain a comprehensive performance score of the battery pack, wherein the comprehensive performance score includes a single cell score and a whole battery pack score; Performing correlation analysis on the comprehensive performance score by grey correlation analysis method to obtain factors affecting battery pack performance, wherein the factors affecting battery pack performance include environmental humidity, discharge depth, charge and discharge cycle and temperature fluctuation; By using a preset Markov chain Monte Carlo method, a probability prediction of the future performance of the battery pack is performed based on the factors affecting the performance of the battery pack to obtain a performance attenuation trend of the battery pack; The battery pack is optimized and analyzed based on the performance degradation trend to specify a corresponding battery pack management strategy; wherein the battery pack management strategy includes a battery pack use strategy and a battery pack maintenance strategy.

7. A BMS testing device, characterized in that: include: The acquisition module is used to acquire data from the BMS system of the battery pack to obtain acquired data, and pre-process the acquired data to obtain pre-processed data; wherein the acquired data includes voltage data, current data and temperature data; An estimation module, used for estimating the state of charge of the battery pack based on the preprocessed data to obtain an estimated value of the state of charge; An evaluation module, configured to construct a time series variation curve based on the charge state estimation value, and to evaluate the remaining service life of the battery pack based on the time series variation curve to obtain a remaining service life evaluation result; An analysis module, configured to perform a health status analysis on the battery pack based on the remaining service life assessment result to obtain a health status report of the battery pack; A formulation module, used for formulating a management strategy for the battery pack based on the health status report; A time series variation curve is constructed based on the estimated state of charge value, and a remaining service life evaluation is performed on the battery pack based on the time series variation curve to obtain a remaining service life evaluation result, including: Decomposing the state of charge estimation value in the time-frequency domain by wavelet transform to obtain multi-scale time-frequency features; wherein the multi-scale time-frequency features include low-frequency trend features and high-frequency fluctuation features; Constructing a state of charge change trend curve based on the multi-scale time-frequency characteristics, and using the state of charge change trend curve as a time series change curve; Performing piecewise linear regression on the time series change curve to obtain piecewise linear fitting parameters; wherein the piecewise linear fitting parameters include a slope and an intercept; Using a support vector regression algorithm, the remaining service life of the battery pack is predicted based on the piecewise linear fitting parameters to obtain a predicted value of the number of remaining cycles; Correcting the predicted value of the number of remaining cycles by using a particle swarm optimization algorithm to obtain a corrected number of remaining cycles; Obtain historical usage data of the battery pack, and comprehensively evaluate the remaining service life of the battery pack based on the historical usage data and the corrected remaining number of cycles to obtain a remaining service life evaluation result; wherein the remaining service life evaluation result includes the remaining number of cycles, the remaining cumulative number of charge and discharge times, and the remaining usage time.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images