A battery health degree evaluation method and system based on an internet-of-things charging pile

Abstract

The application discloses a battery health degree evaluation method and system based on an Internet of Things charging pile, and belongs to the technical field of battery health degree evaluation. The method comprises the following steps: collecting dynamic working condition parameters of a battery to be evaluated in a charging and discharging process through a sensing array of an Internet of Things charging pile, wherein the parameters comprise charging and discharging cycle data, temperature response data and internal resistance response data; constructing multi-physical field characteristic parameters and constructing an electrochemical characteristic map to represent phase change characteristics and interface passivation characteristics of electrode materials; tracking a capacity degradation trajectory and an impedance growth trajectory according to the electrochemical characteristic map, and determining a historical state distribution in a preset state of charge interval through state space reconstruction; performing differential mapping processing on the historical state distribution and an initial state baseline to obtain a decoupling and separation result of reversible degradation components and irreversible degradation components; and calculating and outputting a health degree evaluation value based on the decoupling and separation result. The scheme realizes fine decoupling of battery degradation components and accurate quantification of the health degree.

Description

Technical Field

[0001] This invention relates to the field of battery health assessment technology, specifically to a battery health assessment method and system based on Internet of Things (IoT) charging piles. Background Technology

[0002] The application of IoT charging piles in battery charging and discharging management is becoming increasingly widespread. Real-time monitoring of battery health is crucial for ensuring charging safety and the secondary use of batteries. During use, batteries experience capacity decay and increased impedance. Existing battery health assessment methods typically rely on empirical fitting of a single degradation indicator, such as tracking only changes in capacity retention or DC internal resistance to estimate health. This assessment method ignores the essential difference between active material loss and recoverable polarization effects during battery aging, and fails to reflect the microscopic degradation mechanisms under the coupling of multiple physical fields within the battery. Existing technical solutions often extract statistical features from macroscopic time-series data and directly establish a mapping model between these features and health. The main drawback of this approach is that the obtained health value is a mixture of various degradation effects, unable to distinguish between permanent capacity loss caused by irreversible damage to the electrode structure and transient performance fluctuations caused by recoverable factors such as uneven lithium-ion concentration distribution and hindered interface charge transfer. When the battery experiences different charge / discharge rates or temperature environments, fluctuations in recoverable degradation components can overwhelm the true decay signal of the irreversible structure, leading to significant jumps in assessment results under different operating conditions, resulting in insufficient consistency and accuracy. Furthermore, conventional methods lack the ability to deeply analyze the dynamic behavior of batteries across multiple states of charge, making it difficult to capture subtle signs of early failure, such as shifts in electrochemical characteristic peaks and thickening of the passivation layer. In terms of the in-depth utilization of battery charge-discharge data, it is necessary to address how to extract and visualize multi-dimensional features characterizing phase transitions and interface evolution within the electrochemical system from dynamic operating parameters, and how to decouple the observed macroscopic degradation trajectories to separate the irreversible degradation trend driven by aging mechanisms from the reversible fluctuation components affected by operating condition disturbances. Summary of the Invention

[0003] This invention provides a battery health assessment method and system based on IoT charging piles. The purpose is to construct an electrochemical characteristic spectrum that integrates capacity increment characteristics and polarization internal resistance characteristics from dynamic operating parameters, realize the visual characterization of phase transition and interface passivation behavior of internal electrode materials of the battery, and decouple and separate the macroscopic degradation trajectory of the battery into reversible degradation components and irreversible degradation components by combining phase space reconstruction and differential mapping, thereby obtaining a battery structural health assessment value that is not affected by transient operating condition fluctuations.

[0004] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a battery health assessment method and system based on an IoT charging pile. The method includes: collecting dynamic operating parameters of the battery under evaluation during the charging and discharging process through the sensor array of the IoT charging pile, wherein the dynamic operating parameters include charge-discharge cycle data, temperature response data, and internal resistance response data. Based on the dynamic operating parameters, constructing multi-physics characteristic parameters of the battery under evaluation, and using the multi-physics characteristic parameters to construct an electrochemical characteristic spectrum, wherein the electrochemical characteristic spectrum is used to characterize the phase transition characteristics and interface passivation characteristics of the electrode materials of the battery under evaluation. According to the electrochemical characteristic spectrum, tracking the capacity degradation trajectory and impedance growth trajectory of the battery under evaluation, and determining the historical state distribution of the battery under evaluation within a preset state of charge range by reconstructing the state space of the capacity degradation trajectory and the impedance growth trajectory. Performing differential mapping processing on the historical state distribution and the initial state baseline of the battery under evaluation to obtain the decoupling and separation results of the reversible degradation component and the irreversible degradation component of the battery under evaluation. Based on the decoupling and separation results, the health assessment value of the evaluated battery is calculated and output. This method performs in-depth fusion analysis of multi-dimensional dynamic parameters of the power battery during charging and discharging, constructs a feature map that can sensitively reflect the internal electrochemical state of the battery, realizes high-dimensional state tracking of capacity decay and internal resistance growth processes, and further decouples the degradation process into reversible degradation and irreversible degradation, thereby accurately outputting a comprehensive health value that truly reflects the battery's structural health and performance recovery capability. This solves the problem that relying solely on capacity decay indicators leads to lagging assessment results and an inability to distinguish degradation types.

[0005] As a technical solution of the present invention, when collecting dynamic operating condition parameters, the terminal voltage and charging / discharging current of the battery under evaluation are collected at equal time intervals during the charge / discharge cycle of the battery under evaluation, and the DC internal resistance value of the battery under evaluation at each collection moment is calculated based on the terminal voltage and the charging / discharging current; the surface temperature and internal temperature of the battery under evaluation are collected simultaneously, and the difference between the surface temperature and the internal temperature is used as the temperature response data; the cumulative charge / discharge capacity of the battery under evaluation in each charge / discharge cycle is recorded, and the cumulative charge / discharge capacity is used as the charge / discharge cycle data. This solution utilizes multiple types of sensors integrated into the IoT charging pile to simultaneously acquire electrical, temperature, and cycle data, avoiding the complexity of adding additional collection equipment. At the same time, the introduction of temperature difference can more accurately reflect the dynamic balance between internal heat generation and external heat dissipation of the battery, providing a high-dimensional information foundation for subsequent feature construction.

[0006] As a preferred technical solution of the present invention, when constructing multi-physics characteristic parameters, the capacity increment curve of the battery under evaluation during the constant current charging stage is calculated based on the charge-discharge cycle data and the temperature response data, and the peak position and peak width of the capacity increment curve are extracted; based on the internal resistance response data, the ohmic internal resistance change rate and polarization internal resistance change rate of the battery under evaluation at the charge-discharge switching moment are calculated; the peak position, peak width, ohmic internal resistance change rate, and polarization internal resistance change rate are combined into the multi-physics characteristic parameters. By refining the geometric morphological features and internal resistance dynamic change features of the capacity increment curve, the complex electrochemical evolution process can be parameterized into a quantitatively comparable feature vector. In particular, the synergistic change of peak position and width can effectively indicate the phase transition reversibility loss of the electrode active material, while the internal resistance change rate independently reflects the growth degree of the interface passivation layer.

[0007] In the above technical solution, when constructing the electrochemical characteristic spectrum, the state of charge (SOC) of the battery under evaluation is used as the abscissa, and the capacity increment value among the multiphysics characteristic parameters is used as the ordinate to plot a family of capacity increment curves of the battery under evaluation within a preset voltage range; the state of charge (SOC) of the battery under evaluation is used as the abscissa, and the polarization resistance value among the multiphysics characteristic parameters is used as the ordinate to plot a family of polarization resistance curves of the battery under evaluation; the family of capacity increment curves and the family of polarization resistance curves are superimposed on the same coordinate plane to form the electrochemical characteristic spectrum. This spectrum spatially aligns and superimposes the bulk phase evolution information of the electrode material and the dynamic information of the interfacial reaction impedance on a unified state of charge scale, enabling the phase transition peak shift at the electrode level and the polarization growth at the interface level to be correlated and analyzed from the same perspective, providing an intuitive visualization analysis method for degradation pattern recognition.

[0008] As a further improvement of the present invention, when tracking the capacity degradation trajectory and impedance growth trajectory, the change in the family of capacity increment curves between adjacent charge-discharge cycles is extracted from the electrochemical characteristic spectrum, and the integral value of the change in the state of charge dimension is used as the displacement vector of the capacity degradation trajectory; the increment value of the family of polarization resistance curves between adjacent charge-discharge cycles is extracted from the electrochemical characteristic spectrum, and the maximum value of the increment value in the state of charge dimension is used as the amplitude vector of the impedance growth trajectory; the displacement vector and the amplitude vector are arranged in the order of charge-discharge cycles to obtain the capacity degradation trajectory and the impedance growth trajectory. By extracting the cumulative degradation displacement within the full state of charge range through integration, local measurement noise can be smoothed and the overall capacity loss trend can be accurately captured; while extracting the maximum value of the polarization resistance increment can sensitively capture the degree of interface attenuation under the worst operating conditions. Together, they form a multi-dimensional degradation time series description.

[0009] As a preferred technical solution of the present invention, during state space reconstruction, the capacity degradation trajectory and the impedance growth trajectory are mapped to the first and second dimensions in the phase space, respectively, and the state evolution path of the battery under evaluation is plotted in the phase space. All state points within the preset state of charge interval are clustered to obtain multiple state clusters. The geometric center and dispersion radius of each state cluster are calculated, and these are used as high-dimensional characterization parameters of the historical state distribution. This scheme extends the one-dimensional degradation trajectory to a high-dimensional phase space to reproduce the dynamic evolution structure of the system. Cluster analysis can automatically identify different dominant degradation modes and mode switching points experienced by the battery during degradation. The geometric center calibrates the average state under each degradation mode, and the dispersion radius quantifies the fluctuation amplitude of the state within a local range. The resulting historical state distribution provides richer information about the degradation process than traditional single numerical descriptions.

[0010] As a specific implementation of the present invention, during differential mapping processing, the initial state baseline of the battery under factory calibration is obtained, the initial state baseline including the initial capacity increment curve and the initial polarization resistance curve; the capacity increment curve family in the historical state distribution is subtracted point by point from the initial capacity increment curve to obtain the capacity increment residual surface; the polarization resistance curve family in the historical state distribution is subtracted point by point from the initial polarization resistance curve to obtain the polarization resistance residual surface; the capacity increment residual surface is integrated in the time domain to obtain the first quantization index of the irreversible degradation component; the polarization resistance residual surface is decomposed in the frequency domain to extract the low-frequency component as the second quantization index of the irreversible degradation component, and the high-frequency component as the quantization index of the reversible degradation component. Using the factory baseline as a reference for residual calculation can accurately separate the original characteristics. Time-domain integration can accumulate the irreversible capacity loss caused by permanent structural damage to the active material, while frequency-domain decomposition utilizes the difference between irreversible degradation, which is usually characterized by a low-frequency trend, and reversible degradation, which is characterized by high-frequency fluctuations, to naturally separate the two and achieve component decoupling with clear physical meaning.

[0011] In the above scheme, when obtaining the decoupling and separation results, the quantitative index of the reversible degradation component is normalized and correlated with the average temperature of the battery under evaluation in the current charge-discharge cycle to obtain the temperature-compensated reversible degradation component value; the first quantitative index and the second quantitative index of the irreversible degradation component are weighted and fused to obtain the comprehensive irreversible degradation component value; the temperature-compensated reversible degradation component value and the comprehensive irreversible degradation component value are arranged in charge-discharge cycle order to form the decoupling and separation results. Temperature compensation for the reversible component can correct pseudo-reversible fluctuations caused by differences in operating temperature, making the reversible degradation assessment results comparable under different environmental conditions; the weighted fusion of irreversible indices from the capacity and impedance dimensions fully leverages the advantages of multi-source information complementarity, improving the robustness and accuracy of the irreversible degradation assessment.

[0012] As a preferred technical solution of the present invention, when calculating the health assessment value, the reversible degradation component value and the irreversible degradation component value under the current charge-discharge cycle are extracted from the decoupling and separation results; the irreversible degradation component value is mapped to a preset degradation level range to obtain the structural degradation level of the battery being evaluated; the reversible degradation component value is mapped to a preset recoverability level range to obtain the state recoverability level of the battery being evaluated; based on the structural degradation level and the state recoverability level, a preset health assessment mapping table is queried to obtain the health assessment value. This assessment process converts the original degradation amount into a level index with clear engineering meaning, and uses a two-dimensional mapping table to jointly decide on the degree of irreversible structural damage and the state recoverability. The output health assessment value not only reflects the current permanent damage amount of the battery, but also contains the potential recovery space of its performance after adjusting the operating conditions. The assessment results are closer to the guidance needs of actual tiered utilization and retirement decisions.

[0013] This invention also provides a battery health assessment system based on IoT charging piles, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the aforementioned assessment method. This system can be embedded in existing IoT charging pile control units, utilizing high-density sensor data acquired by the charging pile on a second-level timescale to complete online assessment of battery health status. It eliminates the need to disassemble the battery or transfer it to dedicated testing equipment. The assessment process is completely transparent to the user's charging behavior, and the assessment results can be generated instantly after charging and pushed to the user's terminal via an interactive interface or network. This significantly reduces the time and space costs of battery health detection and improves the convenience and timeliness of the detection.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: When constructing multi-physics characteristic parameters based on dynamic operating conditions, the peak positions and peak widths of the capacity increment curves characterizing the lithium insertion / extraction behavior of the active material, as well as the rates of change of ohmic internal resistance and polarization internal resistance reflecting charge transfer and mass transport resistance, were simultaneously extracted. By superimposing the family of capacity increment curves and the family of polarization internal resistance curves with the state of charge as the common coordinate axis, an electrochemical characteristic spectrum was created. This spectrum correlates the dual information of electrode bulk phase structure evolution and interfacial impedance growth in the same view. The shift and broadening of the capacity increment peak in the spectrum directly correspond to the phase transition path shift of the electrode material and the degree of heterogeneity of the active components, while the overall rise of the polarization internal resistance curve indicates the accumulation of the solid electrolyte interfacial film and the decay of the charge transfer rate. This joint visualization of multi-dimensional electrochemical characteristics allows for the quantitative observation of microscopic processes such as the loss of active materials, cyclic lithium consumption, and interfacial passivation within the battery in the form of morphological changes in the spectrum. This overcomes the limitation of relying solely on a single electrical parameter, which cannot distinguish the dominant mechanism of capacity decay, and provides a high-resolution information carrier for capturing multi-dimensional signs of early latent degradation. The state space of the capacity degradation trajectory and impedance growth trajectory is reconstructed, mapping the two degradation trajectories to the first and second dimensions of the phase space respectively, and the state evolution path is plotted. The geometric center and dispersion radius of the state cluster within the preset charged state interval are obtained as high-dimensional characterization parameters of the historical state distribution. Based on this, the historical state distribution is differentially mapped to the initial state baseline. The first quantization index of the irreversible degradation component is obtained by time-domain integration of the capacity increment residual surface. The frequency-domain decomposition of the polarization internal resistance residual surface is performed, and the low-frequency component is extracted as the second quantization index of the irreversible degradation component, and the high-frequency component is extracted as the quantization index of the reversible degradation component. This differential mapping process utilizes the characteristic that irreversible degradation accumulates continuously in the time domain and its energy is concentrated at low frequencies in the frequency domain to separate reversible and irreversible degradation in the residual space. The decoupled irreversible degradation component is not affected by concentration polarization recovery caused by short-term charge and discharge conditions or temperature fluctuations. It can stably track the real process of permanent structural damage such as electrode active material peeling and current collector corrosion. The health assessment value calculated accordingly will not be artificially high due to partial performance recovery after battery storage, nor will it be underestimated due to transient increase in polarization internal resistance caused by continuous high-rate charging. This significantly improves the accuracy and robustness of health assessment under operating conditions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1This is a flowchart of a battery health assessment method based on IoT charging piles; Figure 2 This is a schematic diagram of electrochemical characteristic spectra. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0018] See Figure 1 This invention provides a battery health assessment method based on IoT charging piles. This method utilizes the sensor array of the IoT charging pile to collect dynamic operating parameters of the battery during charging and discharging. These dynamic operating parameters include charge-discharge cycle data, temperature response data, and internal resistance response data. After acquiring these parameters, a multi-physics characteristic parameter is constructed based on the parameters, and an electrochemical characteristic spectrum characterizing the phase transition and interface passivation features of the electrode material is constructed using this multi-physics characteristic parameter. Based on the electrochemical characteristic spectrum, the capacity degradation trajectory and impedance growth trajectory are tracked, and the state space of these two trajectories is reconstructed to obtain the historical state distribution of the battery under assessment within a preset state of charge range. The historical state distribution is differentially mapped to the initial state baseline of the battery under assessment to decouple reversible degradation components from irreversible degradation components. Finally, the health assessment value is calculated and output based on the decoupling results.

[0019] Example 1: In practical implementation, the IoT charging pile integrates a sensor array, which includes a current sensor connected in series with the charging circuit, a voltage sensor connected in parallel at the connection between the charging output terminal and the battery, a surface temperature sensor attached to the surface of the battery being evaluated, and an internal temperature sensor built into the battery being evaluated or obtained through the battery management system.

[0020] When charging or discharging the battery under evaluation, voltage and current sensors are synchronously triggered to sample at pre-set equal time intervals. The sampling interval is determined based on the charge / discharge rate and dynamic characteristics of the battery under evaluation, for example, set to a fixed value within the range of 0.1 seconds to 1 second. At each sampling moment, the voltage sensor collects the terminal voltage between the positive and negative terminals of the battery under evaluation, and the current sensor collects the charging / discharging current flowing through the battery. The charging / discharging current is recorded as positive during charging and negative during discharging. Using the terminal voltage and charging / discharging current collected at two adjacent sampling moments, the DC internal resistance value corresponding to each sampling moment is calculated. The DC internal resistance value is obtained according to the following formula: in, This represents the DC internal resistance value calculated at the k-th acquisition time. This represents the terminal voltage value acquired at the k-th acquisition time. This represents the terminal voltage value acquired at the (k-1)th acquisition time. This represents the charging / discharging current value collected at the k-th acquisition time. This represents the charging and discharging current value collected at the (k-1)th acquisition time. This represents the difference in terminal voltage between two adjacent data acquisition times. This represents the difference between the charging and discharging currents at two adjacent acquisition times. When the current difference is not zero and the sign is stable, the DC internal resistance value is calculated using the above formula; if the current difference approaches zero, the DC internal resistance value at the previous valid calculation time is used as the DC internal resistance value at the current time. Following this method, time-series data consisting of the DC internal resistance values ​​corresponding to each acquisition time is obtained throughout the entire charging and discharging process, serving as the internal resistance response data.

[0021] While simultaneously acquiring voltage and current, the surface temperature sensor and the internal temperature sensor also sample the temperature at equal time intervals. The surface temperature sensor acquires the temperature value at a designated location on the surface of the battery being evaluated, which is recorded as the surface temperature; the internal temperature sensor acquires the temperature value at the center of the battery cell or a preset internal location, which is recorded as the internal temperature. At each acquisition moment, the difference between the surface temperature and the internal temperature is used as the temperature response data. That is, the temperature response data consists of a series of differences, reflecting the dynamic changes in the temperature difference between the inside and outside of the battery being evaluated.

[0022] For each complete charge-discharge cycle experienced by the battery under evaluation, the cumulative charge-discharge capacity of the battery during the entire charge-discharge process is recorded. The cumulative charge-discharge capacity is calculated by integrating the charge-discharge current over time. Specifically, starting from the beginning of the charge-discharge cycle, the real-time collected charge-discharge current is integrated in ampere-hours until the end of the charge-discharge cycle, and the resulting integral value is used as the charge-discharge cycle data. The cumulative charge-discharge capacity represents the actual capacity charged into the battery under evaluation during the charging cycle and the actual capacity discharged into the battery under evaluation during the discharging cycle.

[0023] The DC internal resistance sequence, temperature response data sequence, and cumulative charge / discharge capacity generated in each charge / discharge cycle together constitute the dynamic operating parameters of the battery being evaluated during that charge / discharge process. These dynamic operating parameters are stored in real time and used to construct subsequent multiphysics characteristic parameters.

[0024] Example 2: In specific implementation, please refer to Figure 2 Capacity increment curves are constructed based on charge-discharge cycle data and temperature response data. The charge-discharge cycle data contains a sequence of the cumulative charge-discharge capacity of the evaluated battery during a complete charge-discharge cycle as a function of sampling time. The temperature response data contains a sequence of the differences between surface and internal temperatures within the same cycle. The construction of the capacity increment curves employs the voltage window differential method: during the constant current charging stage, the terminal voltage is divided into several equally spaced voltage intervals, using the terminal voltage as the sorting criterion. The interval between adjacent voltage intervals is set to a fixed voltage step size, selected based on the electrochemical characteristics of the evaluated battery. For example, for lithium iron phosphate batteries, the voltage step size is 2mV to 5mV; for ternary lithium batteries, the voltage step size is 5mV to 10mV. For each voltage interval, the change in cumulative charge-discharge capacity within the interval and the corresponding change in terminal voltage are calculated. The ratio of these two values ​​is used as the capacity increment value for that voltage interval. The capacity increment value is calculated using the following formula: in, This represents the capacity increment value corresponding to the p-th voltage interval; This represents the change in cumulative charge and discharge capacity within the p-th voltage interval, which is obtained by subtracting the cumulative charge and discharge capacity at the beginning of the interval from the cumulative charge and discharge capacity at the end of the interval. The change in terminal voltage for the p-th voltage interval is represented by subtracting the starting voltage from the ending voltage of the interval. The median terminal voltage for each voltage interval is converted into a state of charge (SCC) value. The SCC value is plotted on the x-axis, and the capacity increment for each voltage interval is plotted on the y-axis to plot the capacity increment curve of the battery under evaluation in one charge-discharge cycle.

[0025] Peak position and peak width are extracted from the capacity increment curve. An extreme value search method is used to identify local maxima on the capacity increment curve where the peak height exceeds a preset threshold. The peak height threshold is set at 10% to 20% of the maximum value of the capacity increment curve within the analysis interval. Among all identified local maxima, the local maximum with the largest ordinate value is designated as the principal characteristic peak, and the corresponding state of charge value is recorded as the peak position. On the capacity increment curve, a straight line parallel to the horizontal axis is drawn at half the height of the principal characteristic peak. This line intersects the principal characteristic peak on both sides of the curve, and the difference in the abscissas of the two intersection points is recorded as the peak width, reflecting the broadening degree of the principal characteristic peak.

[0026] The rate of change of ohmic internal resistance and the rate of change of polarization internal resistance are calculated from the internal resistance response data in the dynamic operating condition parameters. The internal resistance response data is a sequence of DC internal resistance values, exhibiting characteristic changes near the charge / discharge switching moment. The method for identifying the charge / discharge switching moment is to detect the sign change of the charge / discharge current sequence; the moment when the current sign changes from positive to negative or from negative to positive is determined as the charge / discharge switching moment. Taking the charge / discharge switching moment as the center, a subsequence of DC internal resistance values ​​is extracted within a first preset time period before the switching moment and a second preset time period after the switching moment. The first preset time period is 0.5 seconds to 1 second before the switching moment, and the second preset time period is 1 second to 2 seconds after the switching moment. Within the first preset time period before the switching moment, the average value of the DC internal resistance is calculated as the ohmic internal resistance value before the switching moment; within the second preset time period after the switching moment, the average value of the DC internal resistance is calculated as the ohmic internal resistance value after the switching moment. The rate of change of ohmic internal resistance is obtained by subtracting the ohmic internal resistance value before switching from the value after switching, and then dividing by the switching time interval, which is the time difference between the midpoint of the first preset duration and the second preset duration. The rate of change of polarization internal resistance is calculated by linearly fitting the DC internal resistance value sequence within the second preset duration after the switching moment, and using the slope of the fitted line as the rate of change of polarization internal resistance.

[0027] The extracted peak position, peak width, rate of change of ohmic internal resistance, and rate of change of polarization internal resistance are combined into a four-dimensional vector, which constitutes the multiphysics characteristic parameters.

[0028] The process of constructing electrochemical characteristic spectra using multiphysics field parameter mapping involves plotting two families of curves. First, the state of charge (SOC) of the battery under evaluation is plotted on the x-axis, and the capacity increment is plotted on the y-axis. The capacity increment curves obtained under different charge-discharge cycles are then plotted on the same coordinate system, forming a family of capacity increment curves. The SOC value is calculated using the ampere-hour integration method: SOC equals the initial SOC plus the integral of the charge / discharge current over time, divided by the rated capacity of the battery under evaluation. The polarization resistance is obtained as follows: at each sampling time, the DC internal resistance value is subtracted from the ohmic internal resistance value determined for that charge-discharge cycle. The ohmic internal resistance value is taken as the average DC internal resistance over a first preset time period before the charge / discharge switching time. The difference is used as the polarization resistance value at that sampling time. The polarization resistance curves obtained under different charge-discharge cycles are then plotted on the same coordinate system, with the SOC as the x-axis and the polarization resistance value as the y-axis, forming a family of polarization resistance curves.

[0029] The capacity increment curves and polarization resistance curves are superimposed on the same coordinate plane. A common horizontal axis is set as the state of charge axis, the left vertical axis is labeled as the capacity increment axis, and the right vertical axis is labeled as the polarization resistance axis. The capacity increment curves are plotted with reference to the scale of the left vertical axis, and the polarization resistance curves are plotted with reference to the scale of the right vertical axis. The two types of curves are superimposed within the same state of charge range to form an electrochemical characteristic spectrum. The peak evolution of the capacity increment curves and the overall rise or deformation of the polarization resistance curves in the electrochemical characteristic spectrum correspond to changes in the phase transition characteristics and interface passivation characteristics of the electrode material, respectively.

[0030] Example 3: In practice, the displacement vector of the capacity degradation trajectory is extracted from the electrochemical characteristic spectrum. The family of capacity increment curves in the electrochemical characteristic spectrum consists of multiple capacity increment curves corresponding to different charge-discharge cycles. Each capacity increment curve represents the relationship between the capacity increment value and the state of charge (SOC) of the battery under evaluation in a single charge-discharge cycle. Using two adjacent charge-discharge cycles as the basic analysis unit, the capacity increment curve of the previous charge-discharge cycle is denoted as the first capacity increment curve, and the capacity increment curve of the subsequent charge-discharge cycle is denoted as the second capacity increment curve. On the horizontal axis of the electrochemical characteristic spectrum, i.e., the SOC axis, a series of discrete SOC points are selected at preset fixed intervals. The interval of the SOC points is consistent with the SOC resolution obtained by mapping the voltage range when constructing the capacity increment curves. At each discrete SOC point, the corresponding capacity increment value on the first capacity increment curve is obtained and denoted as the first capacity increment value; the corresponding capacity increment value on the second capacity increment curve is obtained and denoted as the second capacity increment value. Subtracting the first capacity increment value from the second capacity increment value yields the change in capacity increment at that SOC point. After performing subtraction operations at all discrete charge state points, a sequence of capacity increment changes is obtained, where each element corresponds to the capacity increment change at a charge state point. This sequence of capacity increment changes is then numerically integrated along the charge state dimension using either the composite trapezoidal rule or the composite Simpson's rule. The integration interval is the lower to upper limit of a preset charge state interval. The integration result is treated as a scalar value, which represents the displacement vector of the capacity degradation trajectory corresponding to two adjacent charge-discharge cycles. This process is repeated for each adjacent charge-discharge cycle pair in the electrochemical characteristic spectrum, yielding a series of displacement vectors.

[0031] The amplitude vector of the impedance growth trajectory is extracted from the electrochemical characteristic spectrum. The family of polarization resistance curves in the electrochemical characteristic spectrum consists of multiple polarization resistance curves corresponding to different charge-discharge cycles. Each polarization resistance curve represents the relationship between the polarization resistance value and the state of charge of the evaluated battery in a single charge-discharge cycle. Similarly, using two adjacent charge-discharge cycles as the basic analysis unit, the polarization resistance curve of the previous charge-discharge cycle is denoted as the first polarization resistance curve, and the polarization resistance curve of the subsequent charge-discharge cycle is denoted as the second polarization resistance curve. At the same discrete state of charge points used in the capacity increment analysis, the corresponding polarization resistance value on the first polarization resistance curve is obtained and denoted as the first polarization resistance value; the corresponding polarization resistance value on the second polarization resistance curve is obtained and denoted as the second polarization resistance value. The second polarization resistance value is subtracted from the first polarization resistance value to obtain the polarization resistance increment value. After performing the subtraction operation at all discrete state of charge points, a sequence of polarization resistance increment values ​​is obtained. In the sequence of polarization resistance increments, the element with the largest absolute value is searched, and this largest absolute value is taken as the amplitude vector of the impedance growth trajectory corresponding to two adjacent charge-discharge cycles. A series of amplitude vectors are obtained by calculating each adjacent charge-discharge cycle pair in the electrochemical characteristic spectrum.

[0032] Arranging the displacement vectors of the capacity degradation trajectory in the order of charge-discharge cycles creates a one-dimensional sequence, which represents the capacity degradation trajectory. The displacement vectors are arranged as follows: the displacement vectors corresponding to the first and second cycles from the first calculation are placed first; the displacement vectors corresponding to the second and third cycles from the second calculation are placed second; and so on. Similarly, arranging the magnitude vectors of the impedance growth trajectory in the order of charge-discharge cycles creates another one-dimensional sequence, which represents the impedance growth trajectory.

[0033] After constructing the capacity degradation trajectory and impedance growth trajectory, state space reconstruction is performed. A phase space is defined, spanned by a first dimension and a second dimension. Each displacement vector in the capacity degradation trajectory is used as the coordinate value of the first dimension, and the magnitude vector corresponding to the same cycle order in the impedance growth trajectory is used as the coordinate value of the second dimension. Thus, each pair of displacement vectors and magnitude vectors determines a state point in the phase space. Adjacent state points are connected by line segments according to the charge / discharge cycle order, forming a broken line in the phase space. This broken line represents the state evolution path of the battery being evaluated. The state evolution path depicts the process of the battery being evaluated evolving from an early cycle state to a late cycle state in the phase space.

[0034] Throughout the entire usage of the battery being evaluated, a predefined state of charge (SOC) range is established. This SOC range represents a specific area of ​​interest for battery operation, such as the range from 30% to 80% SOC. The portions of the state evolution path that meet certain conditions are extracted: when the SOC range of a given state point covers both the preceding and subsequent charge-discharge cycles, that state point is included in the subsequent analysis. All such state points included in the analysis along the state evolution path are then grouped together to form a state point set.

[0035] The state points in the set are clustered to form multiple state clusters. The clustering process uses a density-based clustering algorithm, such as the DBSCAN algorithm. The neighborhood radius parameter is set to the 80th percentile of the Euclidean distance between state points in phase space divided by a preset constant factor, which is 2. The minimum number of neighborhood points parameter is set to 3. After clustering, each state point in the set is assigned a state cluster identifier. All state points assigned the same identifier form a state cluster, resulting in several state clusters.

[0036] Calculate the geometric center and spread radius for each state cluster. For the u-th state cluster, the calculation of the geometric center and spread radius is as follows: in, Denotes the geometric center of the u-th state cluster. It is a two-dimensional vector containing first-dimensional coordinate components and second-dimensional coordinate components; This represents the total number of state points contained in the u-th state cluster; This represents the coordinate vector of the r-th state point within the u-th state cluster. It is also a two-dimensional vector; This represents the summation vector obtained by summing the coordinate vectors of all state points within the u-th state cluster along their respective dimensions. The calculation process for the scatter radius is as follows: first, calculate the scatter radius for each state point within the u-th state cluster... With geometric center The Euclidean distance between the states is used to obtain a series of distance values, and then the maximum value among these distance values ​​is taken as the scatter radius; or the root mean square value of these distance values ​​is taken as the scatter radius. The geometric center of each state cluster is used as the scatter radius. The combination of the dispersion radius and the high-dimensional characterization parameter of the state cluster is used as the high-dimensional characterization parameter of the historical state distribution of the battery under evaluation within the preset state of charge interval, thereby determining the historical state distribution.

[0037] Example 4: In practice, the initial state baseline of the battery under evaluation is obtained under factory calibration conditions. Before leaving the factory, the battery under evaluation completes one calibration cycle on the same model of IoT charging pile according to the standard charge-discharge test procedure. The standard charge-discharge test procedure includes constant current charging to the upper limit cutoff voltage, switching to constant voltage charging until the current drops to the cutoff current, resting for a preset resting time, and constant current discharging to the lower limit cutoff voltage. During the constant current charging phase of the calibration cycle, the voltage step size and differential calculation method are used in the same way as the capacity increment curve family construction process in the historical state distribution to generate the initial capacity increment curve. The initial capacity increment curve consists of a series of state-of-charge sampling points and corresponding initial capacity increment values. During the charge and discharge process of the calibration cycle, the DC internal resistance value sequence is collected synchronously, and the ohmic internal resistance value determined by the charge-discharge switching time is subtracted from the DC internal resistance value sequence in the same way as the polarization internal resistance value calculation process to obtain the initial polarization internal resistance curve. The initial polarization internal resistance curve consists of a series of state-of-charge sampling points and corresponding initial polarization internal resistance values. The initial capacity increment curve and the initial polarization internal resistance curve are stored in the memory of the IoT charging pile as the initial state baseline.

[0038] The family of capacity increment curves in the historical state distribution of the battery under evaluation is subtracted point by point from the initial capacity increment curve. The family of capacity increment curves in the historical state distribution contains several capacity increment curves generated by the battery under evaluation during multiple charge-discharge cycles. Each capacity increment curve consists of a set of capacity increment values ​​at discrete state-of-charge (SOC) sampling points. For the m-th capacity increment curve in the family of capacity increment curves, at each SOC sampling point, the capacity increment value on the m-th capacity increment curve is obtained and recorded as the historical capacity increment value; the initial capacity increment value corresponding to the same SOC sampling point on the initial capacity increment curve is used as the reference value. The historical capacity increment value is subtracted from the initial capacity increment value to obtain a difference, which is the capacity increment residual value at that SOC sampling point. The above subtraction operation is performed on all SOC sampling points on the m-th capacity increment curve to obtain a capacity increment residual curve corresponding to the m-th capacity increment curve. Each point in the capacity increment residual curve contains the coordinates of the SOC sampling point and the corresponding capacity increment residual value. The above operation is performed sequentially on all capacity increment curves in the family of capacity increment curves, aligning all capacity increment residual curves along the state of charge dimension. These are then extended with the charge-discharge cycle sequence as the third dimension, forming a three-dimensional array with the state of charge as the first axis, the charge-discharge cycle sequence as the second axis, and the capacity increment residual value as the third axis. This three-dimensional array constitutes the capacity increment residual surface. Different sections along the charge-discharge cycle sequence direction on the capacity increment residual surface correspond to the capacity increment residual curves under different charge-discharge cycles. These sections along the state of charge direction reflect the evolution characteristics of the capacity increment residual value at different state of charge sampling points with the charge-discharge cycle sequence.

[0039] The family of polarization resistance curves in the historical state distribution of the battery under evaluation is subtracted point by point from the initial polarization resistance curve. The family of polarization resistance curves contains several polarization resistance curves generated by the battery under evaluation during multiple charge-discharge cycles. Each polarization resistance curve consists of polarization resistance values ​​at a set of discrete state-of-charge (SOC) sampling points. For the m-th polarization resistance curve in the family of polarization resistance curves, at each SOC sampling point, the polarization resistance value on the m-th polarization resistance curve is obtained and recorded as the historical polarization resistance value. The initial polarization resistance value corresponding to the same SOC sampling point on the initial polarization resistance curve is used as the reference value. The historical polarization resistance value is subtracted from the initial polarization resistance value to obtain a difference value, which is the polarization resistance residual value at that SOC sampling point. The above subtraction operation is performed on all SOC sampling points on the m-th polarization resistance curve to obtain a polarization resistance residual curve corresponding to the m-th polarization resistance curve. The above operation is performed sequentially on all polarization internal resistance curves in the family of polarization internal resistance curves. All polarization internal resistance residual curves are aligned and arranged in the dimension of state of charge, and extended with the charge-discharge cycle order as the third dimension to form a three-dimensional array with state of charge as the first axis, charge-discharge cycle order as the second axis, and polarization internal resistance residual value as the third axis. This three-dimensional array constitutes the polarization internal resistance residual surface.

[0040] Time-domain integration is performed on the capacity increment residual surface to obtain the first quantization index of the irreversible degradation component. Time-domain integration is performed along the charge-discharge cycle sequence dimension. For each fixed state-of-charge (SOC) sampling point in the capacity increment residual surface, the capacity increment residual values ​​corresponding to all charge-discharge cycles are extracted along the charge-discharge cycle sequence dimension, forming a one-dimensional sequence that varies with the charge-discharge cycle sequence. Numerical integration is performed on this one-dimensional sequence, with the integration interval starting from the capacity increment residual value corresponding to the first charge-discharge cycle and ending at the capacity increment residual value corresponding to the most recent charge-discharge cycle. Numerical integration is implemented using a summation method, where all elements in the one-dimensional sequence are summed sequentially according to the charge-discharge cycle sequence. The summation result is the time-domain integral value corresponding to that SOC sampling point. This time-domain integration operation is performed on all SOC sampling points to obtain a set of time-domain integral values. This set of time-domain integral values ​​is then summed again along the SOC dimension to obtain a comprehensive scalar value, which is the first quantization index of the irreversible degradation component. In one implementation, a weighted summation is used when summing the states of charge dimensions, with the weighting coefficients being the ratio of the width of the voltage interval represented by each state of charge sampling point to the width of the total voltage interval.

[0041] Frequency domain decomposition is performed on the polarization resistance residual surface. Low-frequency components are extracted as the second quantization index of irreversible degradation components, and high-frequency components are extracted as the quantization index of reversible degradation components. The frequency domain decomposition process involves taking the polarization resistance residual values ​​corresponding to all charge-discharge cycles along the charge-discharge cycle sequence for each fixed state-of-charge sampling point in the polarization resistance residual surface, forming a one-dimensional sequence that varies with the charge-discharge cycle sequence. A discrete Fourier transform is performed on this one-dimensional sequence to obtain the frequency domain amplitude spectrum and frequency domain phase spectrum. A cutoff frequency threshold is set on the frequency domain amplitude spectrum. The cutoff frequency threshold is the reciprocal of the total number of charge-discharge cycles multiplied by a cutoff factor, with the cutoff factor ranging from 2 to 5. The portion of the frequency domain amplitude spectrum with frequencies below the cutoff frequency threshold is considered the low-frequency component, and the portion with frequencies above or equal to the cutoff frequency threshold is considered the high-frequency component. Performing an inverse discrete Fourier transform on the frequency domain amplitude and phase spectra corresponding to the low-frequency components reconstructs a time-domain sequence of low-frequency components. Each element in the low-frequency component sequence corresponds to the irreversible polarization resistance residual value at the corresponding state of charge sampling point under the charge-discharge cycle. Performing an inverse discrete Fourier transform on the frequency domain amplitude and phase spectra corresponding to the high-frequency components reconstructs a time-domain sequence of high-frequency components. Each element in the high-frequency component sequence corresponds to the reversible polarization resistance residual value at the corresponding state of charge sampling point under the charge-discharge cycle. The above frequency domain decomposition operation is performed on all state-of-charge (SOC) sampling points. The low-frequency component sequences corresponding to each SOC sampling point are recombined on a two-dimensional plane in terms of SOC dimension and charge / discharge cycle order dimension. The root mean square (RMS) value of the low-frequency component sequences of all SOC sampling points under each charge / discharge cycle is obtained. This RMS value is used as the second quantization index of the irreversible degradation component corresponding to the charge / discharge cycle. The high-frequency component sequences corresponding to each SOC sampling point are recombined on a two-dimensional plane in terms of SOC dimension and charge / discharge cycle order dimension. The RMS value of the high-frequency component sequences of all SOC sampling points under each charge / discharge cycle is obtained. This RMS value is used as the quantization index of the reversible degradation component corresponding to the charge / discharge cycle.

[0042] The quantification index of the reversible degradation component is normalized and correlated with the average temperature of the battery under evaluation during the current charge-discharge cycle. The average temperature of the battery under evaluation during the current charge-discharge cycle is obtained by taking the arithmetic mean of the surface temperature sequences collected by the surface temperature sensor throughout the entire charge-discharge cycle. The expression for the normalized correlation between the quantification index of the reversible degradation component and the average temperature is as follows: in, This represents the reversible degradation component value after temperature compensation during the nth charge-discharge cycle. This represents a quantitative index of the reversible degradation component extracted from the frequency domain decomposition of the polarization internal resistance residual surface during the nth charge-discharge cycle. The temperature sensitivity coefficient is calibrated through reversible capacity recovery experiments on multiple batches of the same model of battery under evaluation at multiple temperature points. The temperature sensitivity coefficient ranges from 0.01 to 0.05. The arithmetic mean of the temperature sequence collected by the surface temperature sensor during the nth charge-discharge cycle is used as the average temperature of the nth charge-discharge cycle. The reference temperature is the arithmetic average of the surface temperatures of the battery being evaluated when it has completed the calibration cycle under factory calibration conditions. If the temperature data under factory calibration conditions is not available, the reference temperature is set to 25°C. This represents an exponential function with the natural constant e as its base. The result of the above normalized correlation processing is the temperature-compensated reversible degradation component value. .

[0043] The first quantization index of the irreversible degradation component is weighted and fused with the second quantization index of the irreversible degradation component. Let the first quantization index of the irreversible degradation component in the nth charge-discharge cycle be denoted as... The second quantitative indicator of the irreversible degradation component is The formula for weighted fusion is: in, This represents the combined irreversible degradation component value during the nth charge-discharge cycle; This represents the weighted fusion coefficient of the first quantitative indicator. The value is 0.6; This represents the weighted fusion coefficient of the second quantitative indicator. The value is 0.4. Weighted fusion coefficient. and The values ​​are determined based on the following: the first quantization index comes from the time-domain integral of the capacity increment residual surface. The capacity increment value is more sensitive to irreversible structural degradation, therefore it is given a larger weight. The second quantization index comes from the low-frequency component of the polarization resistance residual surface. The polarization resistance value is affected by both irreversible degradation and reversible fluctuations, therefore it is given a smaller weight. and The sum is 1.

[0044] The temperature-compensated reversible degradation component value and the comprehensive irreversible degradation component value Arranged from front to back according to the charge-discharge cycle sequence, the decoupling separation result is a two-dimensional data table. The first column of the two-dimensional data table is the charge-discharge cycle sequence number, the second column is the reversible degradation component value after temperature compensation, and the third column is the comprehensive irreversible degradation component value. Each row corresponds to the decoupling separation information of the reversible degradation component and the irreversible degradation component of one charge-discharge cycle.

[0045] Example 5: In practice, the reversible and irreversible degradation component values ​​under the current charge-discharge cycle are extracted from the decoupling results. The decoupling results are presented as a two-dimensional data table. The first column of the two-dimensional data table is the charge-discharge cycle sequence number, the second column is the temperature-compensated reversible degradation component value, and the third column is the comprehensive irreversible degradation component value. Based on the number of charge-discharge cycles completed by the evaluated battery as recorded in real time by the IoT charging pile, the charge-discharge cycle sequence number corresponding to the current charge-discharge cycle is determined. The data is then located in the two-dimensional data table based on the charge-discharge cycle sequence number, and the temperature-compensated reversible degradation component value stored in the second column of that row and the comprehensive irreversible degradation component value stored in the third column of that row are read.

[0046] The comprehensive irreversible degradation component value is mapped to a preset degradation level range. The preset degradation level range is constructed as follows: The statistical distribution of the comprehensive irreversible degradation component values ​​of multiple battery samples of the same model as the evaluated battery at the end of their lifespan is obtained. The end of lifespan is defined as the moment when the usable capacity of a battery sample decays to 80% of its rated capacity. The 5th percentile of the comprehensive irreversible degradation component values ​​of all battery samples at the end of their lifespan is taken as the lower limit of the irreversible degradation component, and the 95th percentile of the comprehensive irreversible degradation component values ​​of all battery samples at the end of their lifespan is taken as the upper limit of the irreversible degradation component. Four intermediate thresholds are inserted at equal intervals between the lower and upper limits of the irreversible degradation component to divide the degradation level into five levels. The comprehensive irreversible degradation component value is converted into a structural degradation level index using the following formula: in, This indicates the level of structural degradation of the battery being evaluated. It is an integer, ranging from 1 to 5; This represents the combined irreversible degradation component value extracted from the decoupling and separation results for the nth charge-discharge cycle. It represents the lower limit of the irreversible degradation component, and is taken as the 5th percentile of the comprehensive irreversible degradation component value of multiple battery samples of the same model at the end of their lifespan. This represents the upper limit of the irreversible degradation component, and its value is the 95th percentile of the combined irreversible degradation component value of multiple battery samples of the same model at the end of their lifespan. This represents the floor function. When the calculated result... When it is less than 1, Corrected to 1; when When it is greater than 5, Revised to 5. (Revised version) The value is used as the structural degradation level of the battery being evaluated.

[0047] The temperature-compensated reversible degradation component values ​​are mapped to a preset recoverability level range. This preset recoverability level range is constructed as follows: The statistical distribution of reversible degradation component values ​​is obtained from multiple battery samples of the same model as the battery being evaluated after undergoing one complete charge-discharge cycle and being left to recover for a preset time. The 5th percentile of the reversible degradation component values ​​of all battery samples is taken as the lower limit of the reversible degradation component, and the 95th percentile is taken as the upper limit. Two intermediate thresholds are inserted at equal intervals between the lower and upper limits of the reversible degradation component to divide the recoverability into three levels. Using an interpolation mapping method, the temperature-compensated reversible degradation component values ​​are compared with the lower, upper, and intermediate thresholds. Based on the threshold range in which the temperature-compensated reversible degradation component values ​​fall, the recoverability level of the evaluated battery is determined. The recoverability level is 1, 2, or 3, where 1 indicates low recoverability and 3 indicates high recoverability.

[0048] Based on the structural degradation level and recoverability level of the battery being evaluated, a preset health assessment mapping table is consulted to obtain the health assessment value. The health assessment mapping table is a two-dimensional matrix stored in the memory of the IoT charging pile. The row index of the two-dimensional matrix represents the structural degradation level, with values ​​ranging from 1 to 5, corresponding to structural degradation levels 1 to 5. The column index represents the recoverability level, with values ​​ranging from 1 to 3, corresponding to recoverability levels 1 to 3. The element in the a-th row and b-th column of the two-dimensional matrix represents the health assessment value when the structural degradation level is a and the recoverability level is b. The health assessment value ranges from 0% to 100%, where 0% indicates the battery is completely failed and 100% indicates the battery is in an initially intact state. The assignment of values ​​to each element in the two-dimensional matrix is ​​completed through offline calibration. Specifically, multiple battery samples of the same model are subjected to cycle life tests under different combinations of structural degradation levels and recoverable state levels. The probability-weighted average of the battery sample maintaining usable capacity in the next service interval under each combination of conditions is recorded. This probability-weighted average is used as the health assessment value of the corresponding matrix element and written into the health assessment mapping table. During querying, the structural degradation level of the battery being evaluated is used as the row index, and the recoverable state level of the battery being evaluated is used as the column index. The matrix is ​​then searched to retrieve the value of the corresponding element, which is used as the health assessment value of the battery being evaluated.

[0049] The processor in the IoT charging pile outputs the health assessment value to the charging pile display interface or the host computer management platform through the communication module, thus completing the output of the health assessment value.

[0050] A battery health assessment system based on IoT charging piles includes a memory, a processor, and a computer program stored in the memory and running on the processor. When executing the computer program, the processor collects dynamic operating parameters through the sensor array of the IoT charging pile, constructs multi-physics characteristic parameters and generates electrochemical characteristic spectra, tracks capacity degradation trajectories and impedance growth trajectories and performs state-space reconstruction, performs differential mapping processing to obtain decoupling results, and calculates and outputs a health assessment value based on the decoupling results.

[0051] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A battery health assessment method based on IoT charging piles, characterized in that, include: S1: Collect dynamic operating parameters of the battery being evaluated during the charging and discharging process through the sensor array of the IoT charging pile. The dynamic operating parameters include charge and discharge cycle data, temperature response data, and internal resistance response data. S2: Based on the dynamic operating parameters, construct the multi-physics field characteristic parameters of the battery under evaluation, and use the multi-physics field characteristic parameters to construct an electrochemical characteristic spectrum. The electrochemical characteristic spectrum is used to characterize the phase transition characteristics of the electrode materials and the interface passivation characteristics of the battery under evaluation. S3: Based on the electrochemical characteristic spectrum, track the capacity degradation trajectory and impedance growth trajectory of the battery under evaluation, and determine the historical state distribution of the battery under evaluation within a preset state of charge range by reconstructing the state space of the capacity degradation trajectory and the impedance growth trajectory. S4: Perform differential mapping processing on the historical state distribution and the initial state baseline of the battery under evaluation to obtain the decoupling and separation results of the reversible degradation component and the irreversible degradation component of the battery under evaluation. S5: Based on the decoupling and separation results, calculate the health assessment value of the battery being evaluated, and output the health assessment value.

2. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The dynamic operating parameters of the battery being evaluated during the charging and discharging process are collected via the sensor array of the IoT charging pile, including: During the charge-discharge cycle of the battery under evaluation, the terminal voltage and charge-discharge current of the battery under evaluation are collected at equal time intervals, and the DC internal resistance of the battery under evaluation at each collection moment is calculated based on the terminal voltage and the charge-discharge current. The surface temperature and internal temperature of the battery being evaluated are collected simultaneously, and the difference between the surface temperature and the internal temperature is used as the temperature response data. Record the cumulative charge-discharge capacity of the battery under evaluation in each charge-discharge cycle, and use the cumulative charge-discharge capacity as the charge-discharge cycle data.

3. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The construction of the multiphysics characteristic parameters of the evaluated battery based on the dynamic operating condition parameters includes: Based on the charge-discharge cycle data and the temperature response data, the capacity increment curve of the evaluated battery during the constant current charging stage is calculated, and the peak position and peak width of the capacity increment curve are extracted. Based on the internal resistance response data, calculate the rate of change of ohmic internal resistance and the rate of change of polarization internal resistance of the battery under evaluation at the charging and discharging switching moment; The peak position, the peak width, the rate of change of ohmic internal resistance, and the rate of change of polarization internal resistance are combined to form the multiphysics characteristic parameters.

4. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The construction of electrochemical characteristic maps using the multi-physics field characteristic parameters includes: Using the state of charge of the battery under evaluation as the horizontal axis and the capacity increment value in the multiphysics characteristic parameters as the vertical axis, a family of capacity increment curves of the battery under evaluation within a preset voltage range is plotted. A family of polarization resistance curves of the battery under evaluation is plotted with the state of charge of the battery under evaluation as the x-axis and the polarization resistance value among the multi-physics characteristic parameters as the y-axis. The family of capacity increment curves and the family of polarization resistance curves are superimposed on the same coordinate plane to form the electrochemical characteristic spectrum.

5. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The step of tracing the capacity degradation trajectory and impedance growth trajectory of the evaluated battery based on the electrochemical characteristic spectrum includes: Extract the change in the family of capacity increment curves between adjacent charge-discharge cycles from the electrochemical feature spectrum, and use the integral value of the change in the state of charge dimension as the displacement vector of the capacity degradation trajectory. The incremental values ​​of the family of polarization internal resistance curves between adjacent charge-discharge cycles are extracted from the electrochemical characteristic spectrum, and the maximum value of the incremental value in the state of charge dimension is taken as the amplitude vector of the impedance growth trajectory. Arrange the displacement vector and the amplitude vector in the order of charge and discharge cycles to obtain the capacity degradation trajectory and the impedance growth trajectory.

6. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The step of determining the historical state distribution of the evaluated battery within a preset state of charge range by reconstructing the state space of the capacity degradation trajectory and the impedance growth trajectory includes: The capacity degradation trajectory and the impedance growth trajectory are respectively mapped to the first and second dimensions in the phase space, and the state evolution path of the evaluated battery is plotted in the phase space. The state evolution path is clustered into multiple state clusters by clustering all state points within the preset charged state interval; Calculate the geometric center and scatter radius of each state cluster, and use each geometric center and scatter radius as high-dimensional characterization parameters of the historical state distribution.

7. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The step of performing differential mapping processing between the historical state distribution and the initial state baseline of the battery being evaluated includes: Obtain the initial state baseline of the battery under factory calibration, the initial state baseline including the initial capacity increment curve and the initial polarization internal resistance curve; The capacity increment residual surface is obtained by subtracting the family of capacity increment curves in the historical state distribution from the initial capacity increment curve point by point. The polarization internal resistance curve family in the historical state distribution is subtracted from the initial polarization internal resistance curve point by point to obtain the polarization internal resistance residual surface. The first quantization index of the irreversible degradation component is obtained by performing time-domain integration on the capacity increment residual surface. The polarization internal resistance residual surface is decomposed in the frequency domain, and the low-frequency component is extracted as the second quantization index of the irreversible degradation component, and the high-frequency component is extracted as the quantization index of the reversible degradation component.

8. The battery health assessment method based on an IoT charging pile according to claim 7, characterized in that, The decoupling and separation results of the reversible degradation component and the irreversible degradation component of the evaluated battery include: The quantitative index of the reversible degradation component is normalized and correlated with the average temperature of the battery under evaluation in the current charge-discharge cycle to obtain the temperature-compensated reversible degradation component value. The first quantitative index and the second quantitative index of the irreversible degradation component are weighted and fused to obtain the comprehensive irreversible degradation component value. The temperature-compensated reversible degradation component value and the comprehensive irreversible degradation component value are arranged in charge-discharge cycle order to form the decoupling and separation result.

9. A battery health assessment method based on an IoT charging pile according to claim 1, characterized in that, The step of calculating the health assessment value of the battery being evaluated based on the decoupling and separation results includes: Extract the reversible degradation component value and the irreversible degradation component value under the current charge-discharge cycle from the decoupling and separation results; The irreversible degradation component value is mapped to a preset degradation level range to obtain the structural degradation level of the evaluated battery; The reversible degradation component value is mapped to a preset recoverability level range to obtain the recoverability level of the evaluated battery. Based on the structural degradation level and the recoverable state level, a preset health assessment mapping table is queried to obtain the health assessment value.

10. A battery health assessment system based on an Internet of Things (IoT) charging pile, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the battery health assessment method based on IoT charging piles as described in any one of claims 1 to 9.

Images