Lithium battery charging and discharging management method and system based on data analysis

Abstract

This invention provides a lithium battery charging and discharging management method and system based on data analysis. It generates a temperature-compensated internal resistance change rate sequence by performing nonlinear compensation processing on the original internal resistance data sequence of the lithium battery according to the dynamic change sequence of the temperature coefficient. Then, an adaptive state estimation algorithm is used to process the temperature-compensated internal resistance change rate sequence to generate a corrected state of charge estimation sequence. Finally, based on the corrected state of charge estimation sequence and the predicted health state level sequence of the lithium battery, a charging and discharging speed decision matrix is ​​determined. This enables the formulation of optimal charging and discharging speed decisions based on the actual state of the battery, resulting in the optimal battery charging and discharging control strategy. It achieves intelligent control of charging and discharging speed by combining multi-dimensional control parameters and conducting in-depth analysis of the interrelationships between parameters, thus protecting the lithium battery while pursuing rapid charging.

Description

Technical Field

[0001] This invention relates to the field of lithium battery charging and discharging technology, and in particular to a lithium battery charging and discharging management method and system based on data analysis. Background Technology

[0002] Lithium-ion batteries, as a core component of modern energy storage technology, play a decisive role in key applications such as electric vehicles, energy storage systems, and portable electronic devices. With the increasing urgency of energy transition, lithium-ion battery charge and discharge management technology has become a significant technological bottleneck affecting the development of the entire industry. Current lithium-ion battery charge and discharge management methods generally suffer from insufficient control precision; most systems rely on simple threshold control using only one or a few parameters, failing to adapt to complex and ever-changing actual operating conditions.

[0003] These traditional methods often exhibit shortcomings in response lag and inaccurate control when faced with dynamic changes in battery state, making it difficult to find the optimal balance between battery performance and lifespan. The fundamental challenge in lithium-ion battery charge and discharge management lies in the complex coupling relationships among multiple key control parameters. Changes in the battery temperature coefficient directly affect the accurate measurement of the rate of change of internal resistance. When the battery operates under different temperature environments, its internal electrochemical reaction rate changes, leading to non-linear changes in internal resistance. Fluctuations in the rate of change of internal resistance further affect the accurate assessment of state of charge (SOC) and state of health (SOH), as SOC is a crucial indicator of the battery's internal state, and its measurement error is amplified during state estimation. Inaccurate SOC and SOH assessments ultimately lead to deviations in charge and discharge rate control decisions. The system cannot formulate the optimal current control strategy based on the battery's true state, potentially damaging battery life while pursuing rapid charging, or sacrificing charging efficiency while protecting the battery.

[0004] Therefore, how to construct a precise management system that can comprehensively consider multi-dimensional control parameters such as battery temperature coefficient, internal resistance change rate, state of charge, and health status, and achieve intelligent regulation of charging and discharging speed through in-depth analysis of the interrelationships between parameters, has become a key issue in the development of lithium battery charging and discharging management technology. Summary of the Invention

[0005] This invention provides a lithium battery charge and discharge management method and system based on data analysis to solve the problem that in the prior art, pursuing fast charging may damage battery life or sacrifice charging efficiency in the process of protecting the battery. It realizes the formulation of the optimal current control strategy based on the actual state of the battery, combined with multi-dimensional control parameters, and achieves intelligent regulation of charge and discharge speed through in-depth analysis of the interrelationship between parameters.

[0006] This invention provides a data analysis-based lithium battery charge and discharge management method, executed by a computer, comprising:

[0007] Based on real-time temperature data of lithium batteries, a dynamic change sequence of temperature coefficient is determined, wherein the dynamic change sequence of temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data.

[0008] Based on the internal resistance reference correction parameter table corresponding to the temperature coefficient dynamic change sequence, the original internal resistance data sequence of the lithium battery is subjected to nonlinear compensation processing to generate a temperature-compensated internal resistance change rate sequence.

[0009] Based on the temperature-compensated internal resistance change rate sequence, an adaptive state estimation algorithm is used to generate a corrected state of charge estimation sequence.

[0010] The corrected state of charge estimation sequence and the preset battery cycle count data are input into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset.

[0011] Based on the corrected state of charge estimation sequence and the health status level sequence, a charge / discharge rate decision matrix is ​​generated;

[0012] Based on the charge / discharge rate decision matrix, the required charging mode for the lithium battery is determined.

[0013] The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

[0014] The lithium battery charge and discharge management method based on data analysis provided by the present invention further includes:

[0015] A real-time feedback control mechanism is adopted to monitor real-time current changes and temperature response data during the charging and discharging process;

[0016] Based on the current change data and the temperature response data, the charging speed setpoint and discharging speed setpoint in the charging and discharging speed decision matrix are dynamically adjusted to generate an optimized control parameter sequence.

[0017] Based on the optimized control parameter sequence, the operating status of the charging and discharging equipment is updated in real time, and the updated operating status data is determined.

[0018] If the operating status data does not meet the preset stability conditions, a new optimized control parameter sequence is determined based on the current change data and the temperature response data.

[0019] The new optimized control parameter sequence is used to update the charge / discharge speed decision matrix.

[0020] The lithium battery charge and discharge management method based on data analysis provided by the present invention further includes:

[0021] Based on the dynamic change sequence of the temperature coefficient, the change rate sequence of the temperature-compensated internal resistance, the corrected state of charge estimation sequence, and the health status level sequence, a weighted fusion algorithm is applied to determine the comprehensive evaluation index sequence of the battery status.

[0022] Based on the comprehensive evaluation index sequence and the preset average performance life threshold of the lithium battery, a personalized charge and discharge management scheme sequence is generated.

[0023] Based on the charge and discharge management scheme sequence, the control strategy parameters are updated, wherein the control strategy parameters include at least the charge and discharge speed decision matrix, the optimized control parameter sequence, the charging mode, the battery cycle count data, or the original internal resistance data sequence.

[0024] According to a data analysis-based lithium battery charging and discharging management method provided by the present invention, the method involves performing nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the internal resistance reference correction parameter table corresponding to the dynamic temperature coefficient change sequence, to generate a temperature-compensated internal resistance change rate sequence, including:

[0025] The original internal resistance data sequence of the lithium battery is nonlinearly compensated using the internal resistance reference correction parameter table to obtain a preliminary compensated data sequence.

[0026] If the influence of temperature data in the preliminary compensation data sequence exceeds a preset threshold, a temperature compensation data sequence is generated based on the preliminary compensation data sequence and the ambient temperature data.

[0027] Based on the temperature compensation data sequence, the internal resistance change rate of the lithium battery is calculated, and the internal resistance change rate sequence is determined.

[0028] Key feature points are extracted from the internal resistance change rate sequence, and the support vector regression algorithm is used to predict the internal resistance change trend of the lithium battery, thus obtaining the internal resistance change trend sequence.

[0029] For the internal resistance change trend sequence, the sliding window method is used to calculate the mean local change rate of the lithium battery, and a smooth internal resistance change rate sequence is obtained.

[0030] Based on the smoothed internal resistance change rate sequence, a time series analysis method is used to detect and remove abnormal change points to generate the temperature-compensated internal resistance change rate sequence.

[0031] According to the present invention, a lithium battery charge and discharge management method based on data analysis is provided, wherein the step of generating a corrected state of charge estimation sequence based on the temperature-compensated internal resistance change rate sequence and using an adaptive state estimation algorithm includes:

[0032] Acquire temperature sensor data and historical voltage and current data sequences, and based on the temperature sensor data and the historical voltage and current data sequences, use the Kalman filter algorithm to smooth the temperature-compensated internal resistance change rate sequence to obtain a smoothed internal resistance change rate sequence.

[0033] By using a preset temperature compensation model, the smoothed internal resistance change rate sequence is temperature compensated to obtain a temperature-compensated internal resistance change rate sequence.

[0034] Using a particle filter algorithm, combined with the temperature-compensated internal resistance change rate sequence and the voltage and current historical data sequence, the battery state of the lithium battery is obtained, and the initial state of charge sequence is obtained.

[0035] If the rate of change of the initial state of charge sequence exceeds a preset threshold, the initial state of charge sequence is adjusted by an error compensation algorithm to generate a corrected state of charge estimation sequence.

[0036] According to a data analysis-based lithium battery charge and discharge management method provided by the present invention, the step of inputting the corrected state of charge estimation sequence and preset battery cycle number data into a health status assessment model to determine the health status level sequence of the lithium battery includes:

[0037] The corrected state of charge sequence and the battery cycle count data are integrated into a unified dataset to obtain a pre-processed battery state dataset.

[0038] Based on the pre-processed battery state dataset, a support vector machine model is used to extract features from it to determine the preliminary feature value sequence of capacity decay, wherein the preliminary feature value sequence of capacity decay is used to characterize the correlation between the state of charge and the number of cycles of the lithium battery.

[0039] Based on the preliminary characteristic value sequence of capacity decay, combined with the pre-established aging index calculation rules, the dynamic change sequence of internal aging index is calculated, and the quantitative result of aging index is obtained.

[0040] Based on the quantitative results of the aging index, anomalies are marked to determine the abnormal distribution of the aging index.

[0041] Based on the abnormal distribution of the aging indicators and the preliminary characteristic value sequence of the capacity decay, a health status level sequence is obtained by comprehensive calculation using a preset health status assessment model.

[0042] According to a data analysis-based lithium battery charge and discharge management method provided by the present invention, the step of generating a charge and discharge rate decision matrix based on the corrected state of charge estimation sequence and the health state level sequence includes:

[0043] The corrected state of charge estimation sequence and health status level sequence are input into the lithium battery, and the real-time state data of the lithium battery is extracted to obtain state sequence data;

[0044] Based on the state sequence data, an intelligent current control algorithm is used to obtain the state of charge and health status of the lithium battery.

[0045] The state of charge and the state of health are classified using the support vector machine algorithm to obtain a classification feature vector.

[0046] Based on the classification feature vector, combined with the preset state of charge threshold and health state threshold, an initial velocity decision matrix is ​​generated through a logistic regression model.

[0047] If the variance of the candidate values ​​in the initial velocity decision matrix is ​​lower than a preset threshold, the initial velocity decision matrix is ​​smoothed by a linear interpolation method to generate the charge / discharge velocity decision matrix.

[0048] This invention also provides a data analysis-based lithium battery charge and discharge management system, comprising:

[0049] The first analysis module is used to determine the dynamic change sequence of the temperature coefficient based on the real-time temperature data of the lithium battery, wherein the dynamic change sequence of the temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data.

[0050] The second analysis module is used to perform nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the internal resistance benchmark correction parameter table corresponding to the dynamic change sequence of the temperature coefficient, and generate a temperature-compensated internal resistance change rate sequence.

[0051] The third analysis module is used to generate a corrected state of charge estimation sequence based on the temperature-compensated internal resistance change rate sequence and an adaptive state estimation algorithm.

[0052] The fourth analysis module is used to input the corrected state of charge estimation sequence and the preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset.

[0053] The fifth analysis module is used to generate a charge / discharge rate decision matrix based on the corrected state of charge estimation sequence and the health status level sequence.

[0054] The sixth analysis module is used to determine the required charging mode for the lithium battery based on the charging and discharging speed decision matrix.

[0055] The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

[0056] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the lithium battery charge and discharge management method based on data analysis as described in any of the preceding claims.

[0057] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the lithium battery charge and discharge management method based on data analysis as described in any of the preceding claims.

[0058] This invention provides a lithium battery charging and discharging management method and system based on data analysis. It generates a temperature-compensated internal resistance change rate sequence by performing nonlinear compensation processing on the original internal resistance data sequence of the lithium battery according to the dynamic change sequence of the temperature coefficient. Then, an adaptive state estimation algorithm is used to process the temperature-compensated internal resistance change rate sequence to generate a corrected state of charge estimation sequence. Finally, based on the corrected state of charge estimation sequence and the predicted health state level sequence of the lithium battery, a charging and discharging speed decision matrix is ​​determined. This enables the formulation of optimal charging and discharging speed decisions based on the actual state of the battery, resulting in the optimal battery charging and discharging control strategy. It achieves intelligent control of charging and discharging speed by combining multi-dimensional control parameters and conducting in-depth analysis of the interrelationships between parameters, thus protecting the lithium battery while pursuing rapid charging. Attached Figure Description

[0059] Figure 1 This is one of the flowcharts of the lithium battery charge and discharge management method based on data analysis provided in the embodiments of the present invention;

[0060] Figure 2 This is the second flowchart of the lithium battery charge and discharge management method based on data analysis provided in the embodiments of the present invention;

[0061] Figure 3 This is the third flowchart of the lithium battery charge and discharge management method based on data analysis provided in this embodiment of the invention;

[0062] Figure 4This is the fourth flowchart of the lithium battery charge and discharge management method based on data analysis provided in this embodiment of the invention;

[0063] Figure 5 This is the fifth flowchart of the lithium battery charge and discharge management method based on data analysis provided in the embodiments of the present invention;

[0064] Figure 6 This is the sixth flowchart of the lithium battery charge and discharge management method based on data analysis provided in this embodiment of the invention;

[0065] Figure 7 This is the seventh flowchart of the lithium battery charge and discharge management method based on data analysis provided in this embodiment of the invention;

[0066] Figure 8 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Reference Figure 1 This invention provides a lithium battery charge and discharge management method based on data analysis, comprising the following steps:

[0069] Step 100: Based on the real-time temperature data of the lithium battery, determine the dynamic change sequence of the temperature coefficient, wherein the dynamic change sequence of the temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data.

[0070] Real-time temperature data and historical temperature curves are acquired from lithium battery sensors. Based on these data, peak temperature, average temperature, and temperature change rate sequences are extracted. Time series analysis is used to determine the acquisition frequency and completeness of the real-time temperature data, resulting in a standardized temperature dataset. A peak temperature detection algorithm is then used to extract the peak temperature sequence from the standardized temperature dataset. The average temperature within a preset time window is calculated using the standardized temperature dataset to obtain the average temperature sequence. Finally, the ratio of the temperature difference between adjacent time points to the time interval is calculated from the peak temperature sequence and the average temperature sequence to obtain the temperature change rate sequence.

[0071] The temperature peak sequence, average sequence, and rate of change sequence are used to construct the temperature matrix to be processed. Principal component analysis (PCA) is then applied to this matrix to extract key information from the multidimensional temperature feature sequence, identifying the main eigenvectors and obtaining a multi-parameter coupled feature set. Based on this multi-parameter coupled feature set, the dynamic change sequence of the temperature coefficient is determined. Alternatively, the multi-parameter coupled feature set can be directly used as the dynamic change sequence of the temperature coefficient. Specifically, if the deviation of the multi-parameter coupled feature set from a preset threshold exceeds a specified range, a Kalman filter algorithm is used to smooth the multi-parameter coupled feature set to obtain the dynamic change sequence of the temperature coefficient.

[0072] Step 200: Based on the internal resistance reference correction parameter table corresponding to the temperature coefficient dynamic change sequence, perform nonlinear compensation processing on the original internal resistance data sequence of the lithium battery to generate a temperature-compensated internal resistance change rate sequence.

[0073] Among them, the internal resistance reference correction parameter table is determined based on the dynamic change sequence of the temperature coefficient, and the temperature-compensated internal resistance change rate sequence records the law of battery internal resistance changing with time under different temperature conditions.

[0074] The determination method for the internal resistance benchmark correction parameter table is as follows: First, the reaction rate constant is calculated based on the absolute temperature of the dynamic temperature coefficient sequence and a preset gas constant to obtain the reaction rate sequence. The calculation of the reaction rate sequence can refer to the Arrhenius equation. Temperature feature values ​​are extracted from the reaction rate sequence, and the correlation with the temperature coefficient sequence is calculated to obtain the internal resistance correction factor. The temperature feature values ​​in the reaction rate sequence include temperature sensitivity features, reaction kinetic state features, safety warning features, or aging correlation features. If the correlation is lower than a preset threshold, the gas constant is adjusted, and the reaction rate sequence is recalculated. Then, based on the internal resistance correction factor and the battery internal resistance value, an initial internal resistance benchmark parameter table is generated. The initial internal resistance benchmark parameter table contains the mapping relationship between the temperature coefficient and the internal resistance correction factor. Parameter pairs are extracted from the initial internal resistance benchmark parameter table, and the least squares method is used to optimize the parameter pairs to obtain optimized internal resistance correction parameters. Based on the optimized internal resistance correction parameters, the initial internal resistance benchmark parameter table is updated, generating the internal resistance benchmark correction parameter table.

[0075] A high-frequency pulse injection method was employed to control external interference and humidity, and the raw data sequence of the internal resistance of a lithium battery was measured. Specifically, a pulse signal was generated using the high-frequency pulse injection method, and a stable frequency pulse signal was output through a preset signal generator to obtain initial pulse data. Then, the initial pulse data was preprocessed using a digital filter to suppress the effects of external interference and humidity, resulting in filtered pulse data. Based on the filtered pulse data, if the signal amplitude exceeded a preset threshold, the signal frequency characteristics were extracted using a Fast Fourier Transform algorithm. These frequency characteristics characterized the frequency response of the battery's internal resistance. Based on the signal frequency characteristics, the raw value of the battery's internal resistance was calculated using the impedance calculation formula. From the raw value of the battery's internal resistance, a sliding window method was used to smooth the raw value, generating a continuous raw internal resistance data sequence. Using the reference correction parameters in the internal resistance reference correction parameter table and combined with ambient temperature data, nonlinear compensation was applied to the raw internal resistance data sequence to generate a temperature-compensated internal resistance change rate sequence.

[0076] Step 300: Based on the temperature-compensated internal resistance change rate sequence, an adaptive state estimation algorithm is used to generate a corrected state of charge estimation sequence.

[0077] The adaptive state estimation algorithm is used to correct the battery's state of charge (SOC) in real time based on the temperature-compensated internal resistance change rate sequence, significantly improving the accuracy of battery state monitoring. Specifically, the core principle of the adaptive state estimation algorithm is to dynamically fuse internal resistance change data, temperature-compensated internal resistance data, and SOC data, dynamically analyze the trend of internal resistance change after temperature compensation, and construct a nonlinear SOC correction model. That is, the adaptive state estimation algorithm can construct a nonlinear dynamic relationship between internal resistance change and charge state to correct the battery's SOC, thus improving the estimation accuracy. Therefore, the adaptive state estimation algorithm establishes a dynamic correlation between battery internal resistance and lithium battery SOC, fully considering the impact of temperature changes on this relationship. It then uses the temperature-compensated internal resistance change rate sequence to correct the lithium battery's SOC under that internal resistance change rate, obtaining a corrected SOC estimation sequence. This corrected SOC estimation sequence characterizes the percentage of remaining charge in the lithium battery, reflecting the proportion of usable energy at a given moment relative to its total energy in a fully charged state. The proposed method for correcting the state of charge using an adaptive state estimation algorithm in this embodiment can dynamically analyze the trend of internal resistance change after temperature compensation and achieve noise adaptation, which significantly improves the robustness of lithium battery state estimation.

[0078] Step 400: Input the corrected state of charge estimation sequence and the preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset.

[0079] Based on the corrected state of charge (SCC) sequence, a sliding window method is used to analyze its changing trend, yielding the battery SCC trend. Using a pre-defined battery health model, combined with the battery SCC trend and historical voltage and current sequences, the battery health is calculated, resulting in a health index. Time series analysis is then employed to predict the future battery SCC based on the health index and the corrected SCC sequence, yielding a predicted SCC sequence.

[0080] By acquiring the corrected state of charge (SOC) sequence from the monitoring system and pre-stored battery cycle count data, a unified dataset is integrated to obtain a pre-processed battery SOC dataset. Based on this dataset, a support vector machine (SVM) model is used to extract features, analyzing the correlation between SOC and cycle count to determine the preliminary feature value sequence of capacity decay and the quantification results of aging indicators. Subsequently, the preliminary feature value sequence of capacity decay and the quantification results of aging indicators are input into a health status assessment model. This model performs comprehensive calculations to obtain a sequence of lithium battery health status levels. The health status assessment model is trained using training sets from the SOC dataset and the charge / discharge cycle count dataset.

[0081] Step 500: Based on the corrected state of charge estimation sequence and the health state level sequence, generate a charge / discharge rate decision matrix;

[0082] In lithium battery management systems, the generation of the charge / discharge rate decision matrix is ​​an intelligent, dynamic decision-making process based on two key input parameters: a corrected state-of-charge (SOC) estimation sequence and a battery health state level sequence. This decision matrix is ​​essentially a multi-dimensional lookup table, providing the system with guidance on the optimal charge / discharge rate under different combinations of SOC and health state conditions.

[0083] First, the corrected state-of-charge (POC) estimation sequence is analyzed. This sequence has undergone temperature compensation and dynamic calibration, accurately reflecting the battery's current true charge state. Simultaneously, the health status level sequence provides an assessment of the battery's long-term health; for example, battery life can be divided into multiple levels (e.g., level A represents good health, level E represents severe aging). By cross-referencing these two key parameters, a two-dimensional decision space is established, where the horizontal axis represents the POC estimation interval and the vertical axis represents the health status level. Therefore, the corrected POC estimation sequence and the lithium battery health status level sequence are input into this two-dimensional decision space to predict charging and discharging strategy decisions, generating a charging and discharging rate decision matrix.

[0084] In each unit where state of charge (SOC) and state of health (SCH) levels intersect, the system presets corresponding charge / discharge rate strategies. For example, when the battery is in a medium SOC range (30%-70%) and the SCH rating is good (Grade A or B), the system allows for a higher charge / discharge rate; however, when the SOC is close to fully charged or discharged, or when the SCH rating indicates significant battery aging, the system automatically limits the charge / discharge rate to protect the battery. This decision matrix also considers dynamic adjustments based on real-time operating conditions, such as automatically lowering the charge / discharge rate thresholds for all units in high-temperature environments.

[0085] Step 600: Based on the charge / discharge rate decision matrix, determine the required charging mode for the lithium battery;

[0086] The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

[0087] The system includes two modes: a fast charging mode parameter sequence and a standard charging mode parameter sequence. Based on the charge / discharge rate decision matrix, it determines whether the state of charge (SOC) is higher than a preset high SOC threshold. If so, the fast charging mode is selected, and a fast charging mode parameter sequence is generated. This sequence is used to charge and discharge the lithium battery according to the fast charging mode. If the SOC is lower than the preset high SOC threshold, the standard charging mode is selected, and a standard charging mode parameter sequence is generated using a parameter mapping table. This sequence is used to charge and discharge the lithium battery according to the standard charging mode. The battery management system transmits either the fast charging mode parameter sequence or the standard charging mode parameter sequence to the charging controller, allowing the controller to adjust the charging and discharging parameters accordingly.

[0088] This invention provides a data analysis-based lithium battery charge and discharge management method. It performs nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the dynamic change sequence of the temperature coefficient, generating a temperature-compensated internal resistance change rate sequence. An adaptive state estimation algorithm is then used to process this temperature-compensated internal resistance change rate sequence, generating a corrected state of charge (SOC) estimation sequence. Subsequently, based on the corrected SOC estimation sequence and the predicted health state level sequence of the lithium battery, a charge and discharge speed decision matrix is ​​determined. This enables the formulation of optimal charge and discharge speed decisions based on the battery's actual state, resulting in an optimal battery charge and discharge control strategy. Furthermore, it achieves intelligent control of the charge and discharge speed by combining multi-dimensional control parameters and conducting in-depth analysis of the relationships between these parameters, thus protecting the lithium battery while pursuing rapid charging.

[0089] In one embodiment, please refer to Figure 2 It also includes:

[0090] Step 700: A real-time feedback control mechanism is adopted to monitor the real-time current change data and temperature response data during the charging and discharging process;

[0091] Step 800: Based on the current change data and the temperature response data, dynamically adjust the charging speed setpoint and discharging speed setpoint in the charging and discharging speed decision matrix to generate an optimized control parameter sequence;

[0092] Step 900: Based on the optimized control parameter sequence, update the operating status of the charging and discharging equipment in real time, and determine the updated operating status data;

[0093] Step 1000: If the operating status data does not meet the preset stability conditions, then a new optimized control parameter sequence is determined based on the current change data and the temperature response data.

[0094] The new optimized control parameter sequence is used to update the charge / discharge speed decision matrix.

[0095] The charging and discharging process utilizes sensor devices to collect real-time current and temperature response data, resulting in an initial monitoring dataset. This dataset includes both current and temperature response data. Based on this dataset, anomaly detection is performed on the current and temperature response data within a preset threshold range. If the initial monitoring dataset exceeds the threshold range, it is marked as an abnormal data point, and the distribution characteristics of the abnormal data are determined. According to these distribution characteristics, abnormal state information corresponding to the abnormal data in the charging and discharging process is obtained. A support vector machine algorithm is used to classify this abnormal state information and determine if there are potential risks in the current charging and discharging process. If potential risks are identified, the charging and discharging speed setpoints in the decision matrix are dynamically adjusted to obtain an optimized control parameter sequence. This optimized control parameter sequence is used to update the operating status of the charging and discharging equipment in real-time, obtaining updated operating data and determining if the operating status meets preset stability conditions. If the operating status does not meet the preset stability conditions, a new optimized control parameter sequence is generated based on the current and temperature response data. This new sequence is then used to update the charging and discharging setpoints in the charging and discharging speed decision matrix.

[0096] Specifically, optimizing the charging and discharging process through a real-time feedback control mechanism requires collecting current and temperature data, dynamically adjusting the charging and discharging speed setpoints in the decision matrix, and generating an optimized control parameter sequence. First, high-precision sensors are used to collect current (amperes, range 0-50A) and temperature (degrees Celsius, range 20-60℃) data during the battery charging and discharging process at a frequency of 100Hz. For example, the current value sequence is [10.5, 11.2, 10.8]A, and the temperature sequence is [25.3, 26.1, 27.0]℃. Next, the rate of change of current (ΔI / Δt) and the rate of temperature rise (ΔT / Δt) are analyzed. If ΔI / Δt > 2A / s or ΔT / Δt > 0.5℃ / s, it indicates that the battery is under high load or at risk of overheating, and the charging and discharging speed needs to be reduced. Based on this, a decision matrix is ​​constructed with a dimension of 3×3, containing charging speed (low: 1C, medium: 2C, high: 3C) and discharging speed (low: 0.5C, medium: 1C, high: 2C), with the initial setting being medium speed (charging 2C, discharging 1C). A fuzzy logic control algorithm is used to adjust the matrix based on the rate of change of current and the rate of temperature rise. For example, when ΔT / Δt = 0.6℃ / s, the fuzzy rule reduces the charging speed from 2C to 1C while maintaining the discharging speed at 1C, outputting the adjusted control parameter sequence: [1C, 1C]. Here, C represents the battery's capacity under standard conditions of full discharge; in the charge / discharge speed calculation, "C" represents the ratio of current to battery capacity, indicating the rate of charge / discharge.

[0097] In one embodiment, please refer to Figure 3 It also includes:

[0098] Step 1100: Based on the temperature coefficient dynamic change sequence, the temperature-compensated internal resistance change rate sequence, the corrected state of charge estimation sequence, and the health status level sequence, a weighted fusion algorithm is applied to determine the comprehensive evaluation index sequence of the battery status.

[0099] Step 1200: Based on the comprehensive evaluation index sequence and the preset average performance life threshold of the lithium battery, generate a personalized charge and discharge management scheme sequence.

[0100] Step 1300: Based on the charge / discharge management scheme sequence, update the control strategy parameters, wherein the control strategy parameters include at least the charge / discharge speed decision matrix, the optimized control parameter sequence, the charging mode, the battery cycle count data, or the original internal resistance data sequence.

[0101] First, four key parameter sequences are collected: a dynamic temperature coefficient change sequence, a temperature-compensated internal resistance change rate sequence, a corrected state of charge estimation sequence, and a health status level sequence. These parameters reflect the battery's environmental adaptability, internal characteristics, real-time state of charge, and long-term health, respectively. The system employs a weighted fusion algorithm to dynamically adjust the weight ratio of each parameter according to the current operating conditions, ultimately generating a comprehensive evaluation index sequence between 0 and 1. This index comprehensively characterizes the battery's immediate state; a higher value indicates a better state. After generating the comprehensive evaluation index sequence, a personalized charge / discharge management scheme sequence is formulated based on the user-preset average performance-lifetime threshold. The system divides the battery state into multiple level intervals, each corresponding to a different management strategy. For example, when the comprehensive evaluation index is high, a performance-priority strategy can be adopted, allowing for a larger charge / discharge current; when the index is low, a lifetime-priority strategy is switched to limit the charge / discharge range to extend battery life. These strategies form an ordered scheme sequence to guide subsequent charge / discharge operations. Then, specific strategy updates are executed. The system dynamically adjusts multiple parameters of the control system according to the charge / discharge management scheme. These parameters include the charge / discharge rate decision matrix, the optimized control parameter sequence, the charging mode selection, battery cycle count data, and the raw internal resistance data sequence. By continuously updating these parameters, the system can achieve adaptive control, ensuring that the battery receives optimal management under various operating conditions.

[0102] Take the battery management system of a certain electric vehicle as an example. One afternoon, the vehicle was preparing for fast charging in a 35℃ environment. The system detected a current temperature coefficient of 0.8 (significantly lower than the standard temperature), a 20% increase in temperature-compensated internal resistance, a corrected state of charge (SOC) of 65%, and a battery health rating of B (85% health). Through weighted calculation, the current comprehensive evaluation index was found to be 0.78, indicating a good state. Based on the owner's preset "balanced mode" threshold, the system decided to adopt a medium-intensity charging strategy: using a 0.5x pulse charging method, setting the SOC upper limit to 85%, and closely monitoring temperature changes. If the temperature exceeds 40℃, the charging current will be automatically reduced. The system updated the following control parameters: changing the charging mode from standard constant current to pulse charging; adjusting the charge / discharge speed decision matrix to adopt a more conservative current setting under high-temperature conditions; recording the initial internal resistance data for subsequent health analysis; and updating the battery cycle count counter.

[0103] This application's embodiments comprehensively analyze various indicators and apply a weighted fusion algorithm to calculate a comprehensive battery state evaluation index. Based on this, a personalized charge and discharge management scheme is generated, forming a closed-loop adaptive management process. Through the coordinated work of the above steps, a faster charging speed is ensured while effectively controlling the rise in battery temperature under high-temperature conditions. This effectively improves the lifespan of lithium batteries, achieving a balance between performance and lifespan, optimizing charge and discharge efficiency, and providing a new technical solution for intelligent management of lithium batteries.

[0104] In one embodiment, please refer to Figure 4 The method of performing nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the internal resistance reference correction parameter table corresponding to the dynamic change sequence of the temperature coefficient to generate a temperature-compensated internal resistance change rate sequence includes:

[0105] Step 201: Using the internal resistance reference correction parameter table, perform nonlinear compensation on the original internal resistance data sequence of the lithium battery to obtain a preliminary compensation data sequence;

[0106] Step 202: If the influence of temperature data in the preliminary compensation data sequence exceeds a preset threshold, then a temperature compensation data sequence is generated based on the preliminary compensation data sequence and the ambient temperature data.

[0107] Step 203: Based on the temperature compensation data sequence, calculate the internal resistance change rate of the lithium battery and determine the internal resistance change rate sequence;

[0108] Step 204: Extract key feature points from the internal resistance change rate sequence, and use the support vector regression algorithm to predict the internal resistance change trend of the lithium battery to obtain the internal resistance change trend sequence.

[0109] Step 205: For the internal resistance change trend sequence, the sliding window method is used to calculate the mean local change rate of the lithium battery to obtain a smooth internal resistance change rate sequence.

[0110] Step 206: Based on the smoothed internal resistance change rate sequence, time series analysis is used to detect and remove abnormal change points to generate the temperature-compensated internal resistance change rate sequence.

[0111] The original internal resistance data sequence of the lithium battery is nonlinearly compensated using the reference correction parameters in the preset internal resistance reference correction parameter table to obtain a preliminary compensated data sequence. If the temperature influence of the preliminary compensated data sequence exceeds a preset threshold, a temperature influence compensation model is used, combined with ambient temperature data, to generate a temperature compensated data sequence.

[0112] Based on the internal resistance values ​​from the previous and current times in the temperature-compensated data sequence, the rate of change of internal resistance is calculated, generating an internal resistance rate of change sequence. Key feature points are extracted from this sequence, and a support vector regression algorithm is used to predict the trend of internal resistance change, resulting in an internal resistance trend sequence. For this trend sequence, a sliding window method is applied to calculate the mean of local rates of change, generating a smoothed internal resistance rate of change sequence. Using this smoothed sequence, time series analysis is employed to detect abnormal changes, generating a temperature-compensated internal resistance rate of change sequence. The formula for calculating the rate of change of internal resistance r is as follows:

[0113]

[0114] Where r is the rate of change of internal resistance in the internal resistance change rate sequence, R k R is the internal resistance value after temperature compensation at the k-th time point, used to represent the internal resistance value at the previous time point in the temperature-compensated data sequence. k―1 Let t be the internal resistance value after temperature compensation at time point k-1, used to represent the internal resistance value at the current time in the temperature-compensated data sequence. k Let t be the timestamp of the k-th time point. k―1 This is the timestamp of the (k-1)th time point.

[0115] This embodiment establishes a multi-level linkage compensation mechanism, significantly improving data quality through the synergistic effect of internal resistance correction, temperature compensation, and anomaly removal. Secondly, it organically combines traditionally separate compensation processing and trend prediction, forming a complete analysis chain from data cleaning to state prediction. Furthermore, it employs various adaptive processing methods, including dynamic thresholds and variable windows, enabling the system to intelligently respond to different operating conditions. Finally, the combination of support vector regression and time series analysis ensures both prediction accuracy and enhances the algorithm's robustness. Thus, through the established intelligent processing flow for lithium battery internal resistance data, multi-level compensation and trend analysis achieve precise characterization and reliable prediction of internal resistance characteristics.

[0116] In one embodiment, please refer to Figure 5 The step of generating a corrected state-of-charge estimation sequence based on the temperature-compensated internal resistance change rate sequence using an adaptive state estimation algorithm includes:

[0117] Step 301: Obtain temperature sensor data and historical voltage and current data sequences, and based on the temperature sensor data and the historical voltage and current data sequences, use the Kalman filter algorithm to smooth the temperature-compensated internal resistance change rate sequence to obtain a smoothed internal resistance change rate sequence.

[0118] Step 302: Using a preset temperature compensation model, perform temperature compensation on the smoothed internal resistance change rate sequence to obtain a temperature-compensated internal resistance change rate sequence.

[0119] Step 303: Using a particle filter algorithm, combined with the temperature-compensated internal resistance change rate sequence and the voltage and current historical data sequence, the battery state of the lithium battery is obtained to obtain the initial state of charge sequence.

[0120] Step 304: If the rate of change of the initial state of charge sequence exceeds a preset threshold, the initial state of charge sequence is adjusted by an error compensation algorithm to generate a corrected state of charge estimation sequence.

[0121] Temperature sensor data and voltage / current acquisition data are acquired. A Kalman filter algorithm is used to smooth the internal resistance change rate sequence, generating a smoothed internal resistance change rate sequence. A pre-defined temperature compensation model is then used to perform temperature compensation on the smoothed internal resistance change rate sequence, resulting in a temperature-compensated internal resistance change rate sequence. A particle filter algorithm is then used, combined with the temperature-compensated internal resistance change rate sequence and historical voltage / current data sequences, to estimate the battery state, obtaining an initial state of charge (SOC) sequence. If the rate of change of the initial SOC sequence exceeds a preset threshold, an error compensation algorithm is used to adjust the estimation parameters, resulting in a corrected SOC sequence.

[0122] Specifically, to analyze battery state and generate a corrected state-of-charge (SOC) estimation sequence by processing the temperature-compensated internal resistance change rate sequence and historical voltage and current data sequences using an adaptive state estimation algorithm, the following implementation method can be adopted. Assume the battery management system collects the internal resistance change rate sequence [0.002, 0.003, 0.0025, 0.004] Ω / s at a temperature of 25℃, the voltage sequence [3.7, 3.68, 3.65, 3.6] V, and the current sequence [1.0, 1.2, 1.1, 0.9] A. First, temperature compensation is applied to the internal resistance change rate, resulting in a compensated internal resistance change rate sequence of [0.00203, 0.003045, 0.0025375, 0.00406] Ω / s. Then, the Kalman filter algorithm is used to process the historical voltage and current data to construct a state-space model. Through iterative updates, the estimated state of charge (SOC) sequence is [0.85, 0.849, 0.848, 0.847]. Analyzing the rate of change of internal resistance, with a preset threshold range of [0.001, 0.0035] Ω / s, a value of 0.00406 is found to exceed the threshold, triggering the error compensation mechanism. Error compensation employs the least squares method to optimize the battery SOC estimation. Combined with the voltage residual, the SOC is adjusted to minimize the residual, resulting in a corrected SOC sequence of [0.851, 0.8495, 0.8482, 0.8473]. Finally, by mapping the battery SOC sequence to the battery health status, the battery condition is assessed, and the calculated battery health status is 99.71%, indicating a good battery condition.

[0123] This embodiment proposes an advanced lithium battery state estimation process. Through multi-stage data processing and intelligent algorithm fusion, it achieves high-precision dynamic estimation of battery internal resistance and state of charge. The innovation of this series of steps is reflected in three aspects: First, it creatively combines Kalman filtering and particle filtering, two algorithms with distinct advantages. The former ensures the quality of data preprocessing, while the latter provides powerful state estimation capabilities. Second, the temperature compensation model is no longer a simple linear correction but considers the performance evolution throughout the battery's entire life cycle, achieving true adaptive compensation. Finally, dynamic threshold design and error compensation mechanisms provide dual protection, enabling the system to automatically identify and correct abnormal estimates, significantly improving the robustness of state estimation.

[0124] In one embodiment, please refer to Figure 6 The step of inputting the corrected state of charge estimation sequence and preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery includes:

[0125] Step 401: Integrate the corrected state of charge sequence and the battery cycle count data into a unified dataset to obtain a pre-processed battery state dataset.

[0126] Step 402: Based on the pre-processed battery state dataset, a support vector machine model is used to extract features from it to determine the preliminary feature value sequence of capacity decay, wherein the preliminary feature value sequence of capacity decay is used to characterize the correlation between the state of charge and the number of cycles of the lithium battery.

[0127] Step 403: Based on the preliminary characteristic value sequence of the capacity decay, and combined with the pre-established aging index calculation rules, calculate the dynamic change sequence of the internal aging index to obtain the quantitative result of the aging index.

[0128] Step 404: Based on the quantification results of the aging index, anomalies are marked to determine the abnormal distribution of the aging index.

[0129] Step 405: Based on the abnormal distribution of the aging indicators and the preliminary characteristic value sequence of the capacity decay, a preset health status assessment model is used to perform comprehensive calculations to obtain a health status level sequence.

[0130] By acquiring the corrected state of charge (SOC) sequence from the monitoring system and pre-stored battery cycle count data, a unified dataset is integrated to obtain a pre-processed battery SOC dataset. Based on this dataset, a support vector machine (SVM) model is used to extract features, analyzing the correlation between SOC and cycle count to determine a preliminary feature value sequence for capacity decay. Using this preliminary feature value sequence, combined with pre-established aging index calculation rules, the dynamic change sequence of internal aging indices is calculated, yielding quantified results. For the quantified results, if an aging index exceeds a preset threshold range, it is marked as an anomaly, and the corresponding cycle count interval is recorded to determine the abnormal distribution of the aging index.

[0131] Based on the abnormal distribution of aging indicators and combined with the preliminary characteristic value sequence of capacity decay, a pre-set health status assessment model is used for comprehensive calculation to obtain a battery health status level sequence. Furthermore, the battery health status level sequence can be compared and analyzed using historical cycle count data. If the trend deviates from the preset normal range, the battery health status level sequence is further corrected to determine the final battery health status level sequence. The health status assessment model is trained using training sets of state-of-charge datasets and charge / discharge cycle count datasets.

[0132] This embodiment proposes a complete lithium battery health status assessment process, achieving accurate assessment and anomaly detection of battery aging status through multi-source data fusion and intelligent analysis. This embodiment correlates state of charge evolution data with cycle count, overcoming the limitations of traditionally viewing these two types of data separately and discovering many new aging characteristics. Secondly, a dynamically configurable aging index calculation rule engine was developed, supporting flexible adjustment of assessment strategies based on different battery chemistry systems (such as NMC and LFP). Furthermore, a multi-level anomaly detection system was established, capable of simultaneously identifying sudden failures and gradual degradation.

[0133] In one embodiment, please refer to Figure 7 The step of generating a charge / discharge rate decision matrix based on the corrected state of charge estimation sequence and the health state level sequence includes:

[0134] Step 501: Input the corrected state of charge estimation sequence and health status level sequence into the lithium battery, and extract the real-time state data of the lithium battery to obtain state sequence data;

[0135] Step 502: Based on the state sequence data, an intelligent current control algorithm is used to obtain the state of charge and health status of the lithium battery.

[0136] Step 503: Use the support vector machine algorithm to classify the state of charge and the state of health to obtain a classification feature vector;

[0137] Step 504: Based on the classification feature vector, and combined with the preset state of charge threshold and health state threshold, an initial velocity decision matrix is ​​generated through a logistic regression model.

[0138] Step 505: If the variance of the candidate values ​​in the initial velocity decision matrix is ​​lower than a preset threshold, the initial velocity decision matrix is ​​smoothed by linear interpolation to obtain the charge / discharge velocity decision matrix.

[0139] The process involves acquiring a state of charge (SOC) estimation sequence and a state of health (SQH) level sequence, then inputting these revised SOC and SQH levels into the lithium battery. Real-time SQH data is extracted from the battery management system (BMS) to obtain the SQH sequence data. If the SQH sequence data is complete and conforms to a preset format, an intelligent current control algorithm is used to predict the SOCH and SQH of the lithium battery. The calculation principle of the intelligent current control algorithm is as follows: In the lithium battery management system, the intelligent current control algorithm dynamically adjusts the charging and discharging strategy by analyzing the battery's SQH sequence data in real time to accurately estimate the SOCH and SQH of the lithium battery. This allows for maximizing charging and discharging efficiency while ensuring battery safety and lifespan. The SQH sequence data can include dynamic parameters such as voltage, current, temperature, and internal resistance. For example, the SQH sequence data can be electrical parameters, temperature data, internal resistance data, and historical cycle data. The intelligent current control algorithm is a multi-objective optimization algorithm that optimizes various target parameters such as battery charging time, battery damage index (lifespan), and energy efficiency by inputting multi-dimensional time series data to predict the optimal SOCH and SQH.

[0140] Next, the state of charge and health status are classified using a support vector machine algorithm to obtain classification feature vectors. Based on the classification feature vectors and combined with preset state of charge and health status thresholds, candidate values ​​for charging and discharging speeds are calculated using a logistic regression model to generate an initial speed decision matrix. If the variance of the candidate values ​​in the initial speed decision matrix is ​​lower than a preset threshold, the matrix is ​​smoothed using a linear interpolation method to obtain the charging and discharging speed decision matrix.

[0141] For example, in the testing of a certain type of power battery, after the system identified that the battery had entered a "high charge state - rapid aging" state, it automatically adjusted the charging current from 1.5A to 1.2A, reducing the capacity decay rate in this state by 40%, while only increasing the charging time by 8%. This intelligent trade-off decision fully demonstrates the system's superior performance in multi-objective optimization, providing an innovative solution for the safe and efficient use of lithium batteries.

[0142] This embodiment constructs a complete intelligent decision-making system for lithium battery charging and discharging speed. Through multi-algorithm collaboration and dynamic optimization, it achieves precise control and adaptive adjustment of the charging and discharging strategy. By establishing a closed-loop linkage mechanism between state evaluation and control strategy generation, the system can continuously learn and improve from actual control effects. Secondly, a hybrid decision-making architecture integrating data-driven and model-driven approaches is developed, possessing both the development capabilities of machine learning and the interpretability of physical models. Furthermore, a matrix optimization algorithm considering the battery's dynamic response is designed, avoiding the strategy jump problem common in traditional methods.

[0143] The data analysis-based lithium battery charge and discharge management system provided by the present invention will be described below. The data analysis-based lithium battery charge and discharge management system described below can be referred to in correspondence with the data analysis-based lithium battery charge and discharge management method described above.

[0144] This invention also provides a data analysis-based lithium battery charge and discharge management system, comprising:

[0145] The first analysis module is used to determine the dynamic change sequence of the temperature coefficient based on the real-time temperature data of the lithium battery, wherein the dynamic change sequence of the temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data.

[0146] The second analysis module is used to perform nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the internal resistance benchmark correction parameter table corresponding to the dynamic change sequence of the temperature coefficient, and generate a temperature-compensated internal resistance change rate sequence.

[0147] The third analysis module is used to generate a corrected state of charge estimation sequence based on the temperature-compensated internal resistance change rate sequence and an adaptive state estimation algorithm.

[0148] The fourth analysis module is used to input the corrected state of charge estimation sequence and the preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset.

[0149] The fifth analysis module is used to generate a charge / discharge rate decision matrix based on the corrected state of charge estimation sequence and the health status level sequence.

[0150] The sixth analysis module is used to determine the required charging mode for the lithium battery based on the charging and discharging speed decision matrix.

[0151] The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

[0152] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840. The processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logic instructions from the memory 830 to execute a lithium battery charge / discharge management method based on data analysis.

[0153] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0154] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the data analysis-based lithium battery charge and discharge management method provided by the above methods.

[0155] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0156] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A lithium battery charge-discharge management method based on data analysis, characterized by, Executed by a computer, including: Based on real-time temperature data of lithium batteries, a dynamic change sequence of temperature coefficient is determined, wherein the dynamic change sequence of temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data. Based on the internal resistance reference correction parameter table corresponding to the temperature coefficient dynamic change sequence, the original internal resistance data sequence of the lithium battery is subjected to nonlinear compensation processing to generate a temperature-compensated internal resistance change rate sequence. Based on the temperature-compensated internal resistance change rate sequence, an adaptive state estimation algorithm is used to generate a corrected state of charge estimation sequence. The corrected state of charge estimation sequence and the preset battery cycle count data are input into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset. Based on the corrected state of charge estimation sequence and the health status level sequence, a charge / discharge rate decision matrix is ​​generated; Based on the charge / discharge rate decision matrix, the required charging mode for the lithium battery is determined. The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

2. The data analysis based lithium battery charge-discharge management method according to claim 1, wherein, Also includes: A real-time feedback control mechanism is adopted to monitor real-time current changes and temperature response data during the charging and discharging process; Based on the current change data and the temperature response data, the charging speed setpoint and discharging speed setpoint in the charging and discharging speed decision matrix are dynamically adjusted to generate an optimized control parameter sequence. Based on the optimized control parameter sequence, the operating status of the charging and discharging equipment is updated in real time, and the updated operating status data is determined. If the operating status data does not meet the preset stability conditions, a new optimized control parameter sequence is determined based on the current change data and the temperature response data. The new optimized control parameter sequence is used to update the charge / discharge speed decision matrix.

3. The lithium battery charge and discharge management method based on data analysis according to claim 2, characterized in that, Also includes: Based on the dynamic change sequence of the temperature coefficient, the change rate sequence of the temperature-compensated internal resistance, the corrected state of charge estimation sequence, and the health status level sequence, a weighted fusion algorithm is applied to determine the comprehensive evaluation index sequence of the battery status. Based on the comprehensive evaluation index sequence and the preset average performance life threshold of the lithium battery, a personalized charge and discharge management scheme sequence is generated. Based on the charge and discharge management scheme sequence, the control strategy parameters are updated, wherein the control strategy parameters include at least the charge and discharge speed decision matrix, the optimized control parameter sequence, the charging mode, the battery cycle count data, or the original internal resistance data sequence.

4. The lithium battery charge and discharge management method based on data analysis according to claim 1, characterized in that, The internal resistance reference correction parameter table based on the dynamic change sequence of the temperature coefficient is used to perform nonlinear compensation processing on the original internal resistance data sequence of the lithium battery to generate a temperature-compensated internal resistance change rate sequence, including: The original internal resistance data sequence of the lithium battery is nonlinearly compensated using the internal resistance reference correction parameter table to obtain a preliminary compensated data sequence. If the influence of temperature data in the preliminary compensation data sequence exceeds a preset threshold, a temperature compensation data sequence is generated based on the preliminary compensation data sequence and the ambient temperature data. Based on the temperature compensation data sequence, the internal resistance change rate of the lithium battery is calculated, and the internal resistance change rate sequence is determined. Key feature points are extracted from the internal resistance change rate sequence, and the support vector regression algorithm is used to predict the internal resistance change trend of the lithium battery, thus obtaining the internal resistance change trend sequence. For the internal resistance change trend sequence, the sliding window method is used to calculate the mean local change rate of the lithium battery, and a smooth internal resistance change rate sequence is obtained. Based on the smoothed internal resistance change rate sequence, a time series analysis method is used to detect and remove abnormal change points to generate the temperature-compensated internal resistance change rate sequence.

5. The data analysis based lithium battery charge-discharge management method according to claim 1, wherein, The step of generating a corrected state-of-charge estimation sequence based on the temperature-compensated internal resistance change rate sequence using an adaptive state estimation algorithm includes: Acquire temperature sensor data and historical voltage and current data sequences, and based on the temperature sensor data and the historical voltage and current data sequences, use the Kalman filter algorithm to smooth the temperature-compensated internal resistance change rate sequence to obtain a smoothed internal resistance change rate sequence. By using a preset temperature compensation model, the smoothed internal resistance change rate sequence is temperature compensated to obtain a temperature-compensated internal resistance change rate sequence. Using a particle filter algorithm, combined with the temperature-compensated internal resistance change rate sequence and the voltage and current historical data sequence, the battery state of the lithium battery is obtained, and the initial state of charge sequence is obtained. If the rate of change of the initial state of charge sequence exceeds a preset threshold, the initial state of charge sequence is adjusted by an error compensation algorithm to generate a corrected state of charge estimation sequence.

6. The lithium battery charge and discharge management method based on data analysis according to claim 1, characterized in that, The step of inputting the corrected state of charge estimation sequence and preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery includes: The corrected state of charge sequence and the battery cycle count data are integrated into a unified dataset to obtain a pre-processed battery state dataset. Based on the pre-processed battery state dataset, a support vector machine model is used to extract features from it to determine the preliminary feature value sequence of capacity decay, wherein the preliminary feature value sequence of capacity decay is used to characterize the correlation between the state of charge and the number of cycles of the lithium battery. Based on the preliminary characteristic value sequence of capacity decay, combined with the pre-established aging index calculation rules, the dynamic change sequence of internal aging index is calculated, and the quantitative result of aging index is obtained. Based on the quantitative results of the aging index, anomalies are marked to determine the abnormal distribution of the aging index. Based on the abnormal distribution of the aging indicators and the preliminary characteristic value sequence of the capacity decay, a health status level sequence is obtained by comprehensive calculation using a preset health status assessment model.

7. The data analysis based lithium battery charge-discharge management method according to claim 1, wherein, The step of generating a charge / discharge rate decision matrix based on the corrected state of charge estimation sequence and the health state level sequence includes: The corrected state of charge estimation sequence and health status level sequence are input into the lithium battery, and the real-time state data of the lithium battery is extracted to obtain state sequence data; Based on the state sequence data, an intelligent current control algorithm is used to obtain the state of charge and health status of the lithium battery. The state of charge and the state of health are classified using the support vector machine algorithm to obtain a classification feature vector. Based on the classification feature vector, combined with the preset state of charge threshold and health state threshold, an initial velocity decision matrix is ​​generated through a logistic regression model. If the variance of the candidate values ​​in the initial velocity decision matrix is ​​lower than a preset threshold, the initial velocity decision matrix is ​​smoothed by a linear interpolation method to generate the charge / discharge velocity decision matrix.

8. A lithium battery charge-discharge management system based on data analysis, characterized by, include: The first analysis module is used to determine the dynamic change sequence of the temperature coefficient based on the real-time temperature data of the lithium battery, wherein the dynamic change sequence of the temperature coefficient is obtained based on the temperature peak, temperature average and temperature change rate sequence of the real-time temperature data. The second analysis module is used to perform nonlinear compensation processing on the original internal resistance data sequence of the lithium battery based on the internal resistance benchmark correction parameter table corresponding to the dynamic change sequence of the temperature coefficient, and generate a temperature-compensated internal resistance change rate sequence. The third analysis module is used to generate a corrected state of charge estimation sequence based on the temperature-compensated internal resistance change rate sequence and an adaptive state estimation algorithm. The fourth analysis module is used to input the corrected state of charge estimation sequence and the preset battery cycle count data into the health status assessment model to determine the health status level sequence of the lithium battery. The health status assessment model is trained based on the training set of the state of charge dataset and the charge-discharge cycle count dataset. The fifth analysis module is used to generate a charge / discharge rate decision matrix based on the corrected state of charge estimation sequence and the health status level sequence. The sixth analysis module is used to determine the required charging mode for the lithium battery based on the charging and discharging speed decision matrix. The charging mode is used to regulate the charging and discharging speed of the lithium battery, and the charging mode includes a fast charging mode and a standard charging mode.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the lithium battery charge and discharge management method based on data analysis as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the lithium battery charge and discharge management method based on data analysis as described in any one of claims 1 to 7.

Images