Battery pack real-time health state estimation method, fast charging regulation and control method and device

Abstract

The invention provides a battery pack real-time health state estimation method, a fast charging regulation and control method and a fast charging regulation and control device, and the method comprises the steps: collecting a multi-source physical signal of a battery pack in a charging process, the multi-source physical signals comprise the current charge state of the battery pack, electrical signals and temperature signals of the single batteries, mechanical signals of mechanical deformation between the single batteries and electrochemical impedance signals of the battery pack; performing feature extraction on the multi-source physical signal to obtain feature vectors including instantaneous ohmic internal resistance, expansion change rate and electrochemical impedance relative change rate; and inputting the feature vector into a pre-trained health state estimation model to obtain a real-time health factor reflecting the current internal health state of the battery pack. According to the invention, the estimation real-time performance of the health state of the battery is improved, and the response speed of regulating and controlling fast charging subsequently can be improved.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle power battery management technology, specifically to a method for estimating the real-time health status of a battery pack, a fast-charging control method, and a device. Background Technology

[0002] The state of health (SOH) of a power battery is a key indicator for measuring its performance degradation, remaining lifespan, and operational safety. Accurate and real-time estimation of SOH is crucial for electric vehicles. Especially in fast-charging scenarios, the internal electrochemical and thermal stresses of the battery increase dramatically. Real-time monitoring of its health status is a prerequisite for implementing precise power regulation and ensuring long battery life and high safety while improving charging speed.

[0003] Currently, the industry mainly relies on monitoring external electrical parameters of the battery and combining historical data for SOH estimation and charging protection. Representative solutions include: Patent CN113311385A, which adjusts the maximum charging current based on the battery's SOH, but the SOH it relies on is usually based on long-term statistical values ​​of cycle count or capacity decay, resulting in slow updates and failing to reflect instantaneous internal deterioration (such as rapid lithium plating) occurring during fast charging, leading to lagging and inefficient regulation. Patent CN116184217A integrates multiple factors such as capacity, time, and internal resistance for online estimation; however, electrical parameters such as internal resistance are only indirect and slow-moving responses to internal changes. For example, when slight lithium plating has already occurred, the DC internal resistance may not have changed significantly, causing the optimal intervention time to be missed. Patent CN115453448A calculates the Health Index (BHI) by recording data such as voltage, current, and temperature, but it is essentially a post-event analysis and early warning tool, lacking the ability to control fast charging power in real time, thus breaking the control closed loop. Most commercial battery management systems employ dual cutoff protection based on voltage and surface temperature during fast charging, or derating power according to a fixed aging curve calibrated at the factory. These methods completely ignore individual battery differences, usage history, and changes in internal mechanical state, resulting in limited and conservative protection.

[0004] In summary, existing technologies generally cannot directly, in real-time, and online monitor the core internal physicochemical changes that lead to battery life degradation during fast charging (such as lithium dendrite growth, active material shedding, and electrolyte consumption). Their health status estimation relies on indirect and lagging external parameters, causing subsequent fast charging protection strategies to either risk damaging the battery due to information lag or be overly conservative in their safety measures, severely limiting charging efficiency. Therefore, there is an urgent need to provide a real-time battery pack health status estimation method, fast charging control method, and device that can directly sense the battery's internal microscopic health status and achieve second-level response control. Summary of the Invention

[0005] In view of this, it is necessary to provide a method for estimating the real-time health status of a battery pack, a fast charging control method and device, to solve the technical problem that the health status estimation in the prior art depends on indirect and lagging external parameters, which leads to the subsequent fast charging protection strategy either risking damage to the battery due to information lag, or being too conservative in order to ensure safety, which seriously restricts the charging efficiency.

[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a method for estimating the real-time health status of a battery pack, wherein the battery pack comprises multiple individual battery cells, and the method includes:

[0007] The battery pack is collected from multiple sources during the charging process. These multiple sources include: the current state of charge of the battery pack, the electrical and temperature signals of individual battery cells, the mechanical signals of mechanical deformation between individual battery cells, and the electrochemical impedance signal of the battery pack. Feature extraction is performed on the multi-source physical signal to obtain a feature vector including instantaneous ohmic internal resistance, expansion rate, and relative electrochemical impedance rate. The feature vector is input into a pre-trained health status estimation model to obtain a real-time health factor that reflects the current internal health status of the battery pack.

[0008] In one possible implementation, prior to feature extraction from the multi-source physical signal, the method further includes: The electrical and mechanical signals are subjected to high-frequency noise filtering to obtain optimized electrical and mechanical signals. The temperature signal is subjected to delay compensation processing to obtain a temperature compensation signal; The optimized electrical signal, the optimized mechanical signal, the temperature compensation signal, the current state of charge, and the electrochemical impedance signal are time-aligned.

[0009] In one possible implementation, the battery cell includes tabs and a casing, then the electrical signals include the terminal voltage and charging current of the battery cell, and the temperature signals include a first temperature signal at the location of the tabs and a second temperature signal at the center location of the casing.

[0010] In one possible implementation, the health status estimation model includes a depthwise separable convolutional unit and a gated recurrent unit. The depthwise separable convolutional unit is used to extract local features from the feature vector to obtain local features, and the gated recurrent unit is used to capture the temporal evolution features of the local features to obtain the real-time health factor.

[0011] In one possible implementation, the real-time health factors include an instantaneous degradation index characterizing the battery pack’s ability to withstand electrochemical and thermal stresses under current charging conditions, and a risk confidence level characterizing the reliability of the estimate.

[0012] Secondly, the present invention also provides a method for controlling fast charging of a battery pack, comprising: Obtain real-time health factors for the battery pack; Based on the real-time health factors, the temperature of the battery pack, and the current state of charge, a charging control command including charging control parameters is generated in real time. In response to the charging control command, the battery pack is controlled to perform fast charging; The real-time health factor is determined based on the battery pack real-time health status estimation method described in any of the above possible implementations.

[0013] In one possible implementation, the charging control parameters include a current scaling factor, a voltage offset, and a current ramp rate limit.

[0014] In one possible implementation, the method further includes: Obtain the battery indicator values ​​after executing the charging control command; When the battery indicator value exceeds the expected range, a safety fault tolerance strategy is triggered to perform derating charging or stop charging, and the battery indicator value is recorded.

[0015] Thirdly, the present invention also provides a real-time health status estimation device for a battery pack, the battery pack comprising multiple battery cells, the device comprising: A multi-source physical signal acquisition unit is used to acquire multi-source physical signals of the battery pack during the charging process. The multi-source physical signals include the current state of charge of the battery pack, the electrical signals of the individual battery cells, the temperature signals of the individual battery cells, the mechanical signals between the individual battery cells, and the electrochemical impedance signals of the individual battery cells. The feature extraction unit is used to extract features from the multi-source physical signal to obtain the instantaneous ohmic internal resistance, the rate of change of expansion and the relative rate of change of electrochemical impedance, and to form a feature vector based on the instantaneous ohmic resistance. The real-time health factor determination unit is used to input the feature vector into a pre-trained health status estimation model and output real-time health factors that reflect the current health status of the battery.

[0016] Fourthly, the present invention also provides a battery pack fast charging control device, comprising: Real-time health factor acquisition unit, used to acquire the real-time health factors of the battery; A charging control command generation unit is used to generate charging control commands, including charging control parameters, in real time based on the real-time health factors, battery temperature, and current state of charge. Fast charging control unit, used to control the battery pack to perform fast charging in response to the charging control command; The real-time health factor is determined based on the battery pack real-time health status estimation method described in any of the above possible implementations.

[0017] The beneficial effects of this invention are as follows: The real-time health status estimation method for battery packs provided by this invention collects the current state of charge of the battery pack, electrical and temperature signals of individual battery cells, mechanical signals of mechanical deformation between battery cells, and electrochemical impedance signals of the battery pack. The current state of charge can characterize the energy state of the battery pack, the electrical signals can reflect the conductivity and polarization behavior of individual battery cells, the mechanical signals can reflect the changing trend of lithium intercalation behavior inside individual battery cells and structural stress changes caused by abnormal side reactions or lithium plating, and the electrochemical impedance signals can reflect the battery interface state and transmission characteristics. Moreover, the mechanical and electrochemical impedance signals are real-time sensing signals. Based on the multi-source physical signals composed of the above-mentioned signals representing different dimensions, the health status estimation can be performed so that the health status estimation no longer depends on the lagging changes in macroscopic parameters. It is directly rooted in the core physicochemical processes that lead to performance degradation, realizing the real-time estimation of battery health status, and thus improving the response speed of subsequent fast charging regulation.

[0018] Furthermore, this invention integrates multi-source physical signals for real-time health status estimation, overcoming the shortcomings of noise interference or incomplete characterization that may exist with a single signal source, and significantly improving the accuracy and robustness of health status estimation. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A schematic flowchart of an embodiment of the battery pack real-time health status estimation method provided by the present invention; Figure 2 A schematic diagram of an embodiment of the health status estimation model provided by the present invention; Figure 3 This is a schematic flowchart of an embodiment of the battery pack fast charging control method provided by the present invention; Figure 4 A schematic diagram of an embodiment of the triggering and execution of the security fault tolerance strategy provided by the present invention; Figure 5 A schematic diagram of an embodiment of the battery pack real-time health status estimation device provided by the present invention; Figure 6 This is a schematic diagram of an embodiment of the battery pack fast charging control device provided by the present invention. Detailed Implementation

[0021] 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 a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0022] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this invention illustrate operations implemented according to some embodiments of the invention. It should be understood that the operations in the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed or performed simultaneously. Furthermore, those skilled in the art, guided by the content of this invention, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0023] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0024] This invention provides a method for estimating the real-time health status of a battery pack, a fast-charging control method, and a device, which are described in detail below.

[0025] Before demonstrating specific embodiments, the structure of the battery pack will be described. Specifically, the battery pack includes multiple battery cells, which are the smallest independently functioning electrochemical units constituting the battery pack. The battery cells include: Electrode assembly: Composed of positive electrode, negative electrode and separator placed between them by winding or stacking, it is the core area where lithium-ion insertion / extraction reaction occurs.

[0026] Electrolyte: The medium that provides lithium-ion transport, which is immersed in the electrode assembly and the separator.

[0027] Housing: Used to encapsulate electrode components and electrolyte, providing mechanical protection, electrical insulation and maintaining the internal environment.

[0028] Tabs: These are metal conductive plates extending from the electrode assembly, divided into positive and negative tabs, used to lead the current from the battery cell to the external circuit. Tabs are the key interface for power transmission and connection with the outside world, and are also the parts where current density is concentrated and significant ohmic heat is generated during fast charging.

[0029] Figure 1 This is a schematic flowchart of an embodiment of the battery pack real-time health status estimation method provided by the present invention, as shown below. Figure 1 As shown, the real-time health status estimation method for battery packs includes: S101. Collect multi-source physical signals of the battery pack during the charging process. The multi-source physical signals include: the current state of charge of the battery pack, the electrical and temperature signals of the individual battery cells, the mechanical signals of mechanical deformation between the individual battery cells, and the electrochemical impedance signal of the battery pack.

[0030] Specifically, the current state of charge is calculated by the state of charge estimation module inside the battery management system. The battery management system is an integrated and mature system in the vehicle. It can calculate and output the current state of charge in real time based on the common method of ampere-hour integration combined with periodic open-circuit voltage correction, which will not be elaborated here.

[0031] Specifically, electrical signals include the terminal voltage and charging current of individual battery cells, while mechanical signals are mechanical expansion signals or micro-strain. These electrical, temperature, and mechanical signals are obtained based on corresponding sensors; for example, terminal voltage is acquired using a voltage sensor, charging current using a current sensor, and mechanical signals using a mechanical sensor.

[0032] The acquisition frequency of electrical signals and temperature signals shall not be less than 100Hz, and the acquisition frequency of mechanical signals shall not be less than 10Hz.

[0033] Preferably, a thin-film micro-pressure sensor array is pre-embedded between adjacent battery cells to acquire mechanical signals.

[0034] The electrochemical impedance spectroscopy (EIS) signal acquisition process is as follows: Before each DC fast charge begins, or during a pause (>30 seconds) in the charging process due to vehicle pausing, the battery management system applies a multi-frequency AC signal ranging from 0.1Hz to 1000Hz to the battery pack via a dedicated excitation circuit, and acquires the voltage and current responses of the battery pack. By processing the response signal, an EIS signal reflecting the battery interface state and transmission characteristics is obtained. The real and imaginary parts of the EIS spectrum are extremely sensitive to the state of the SEI film on the electrode surface and the ionic conductivity of the electrolyte.

[0035] S102. Perform feature extraction on the multi-source physical signals to obtain feature vectors including instantaneous ohmic internal resistance, expansion rate, and relative change rate of electrochemical impedance.

[0036] In a specific embodiment of the present invention, the instantaneous ohmic internal resistance Through terminal voltage and charging current The calculations show that, specifically: ,in, The instantaneous rate of change of the terminal voltage. This represents the instantaneous rate of change of the charging current.

[0037] When the mechanical signal is mechanical expansion F, the rate of expansion change is... When the mechanical signal is micro-strain At that time, the collision change rate .

[0038] Among them, the relative change rate of electrochemical impedance It refers to the rate of change of the current electrochemical impedance relative to the initial electrochemical impedance of a brand-new battery pack.

[0039] Then the eigenvector is X = [ , , [, T, SOC], where T is the temperature signal and SOC is the current state of charge.

[0040] S103. Input the feature vector into the pre-trained health status estimation model to obtain the real-time health factor reflecting the current internal health status of the battery pack.

[0041] In a specific embodiment of the present invention, the real-time health factors include an instantaneous degradation index characterizing the battery pack’s ability to withstand electrochemical and thermal stresses under current charging conditions, and a risk confidence level characterizing the reliability of the estimate.

[0042] Specifically, the instantaneous decline index is a scalar between 0 and 1, where 1 represents excellent health and 0 represents severe decline. The risk confidence level is also a scalar between 0 and 1.

[0043] To improve the efficiency of real-time health factor estimation, in a specific embodiment of the present invention, the health status estimation model is a lightweight neural network model, and combined with technologies such as knowledge distillation, it is ensured that it can run at millisecond speeds on automotive-grade chips.

[0044] It should be understood that the real-time health status estimation method for the battery pack in this embodiment of the invention can be implemented in any device based on the real-time health status estimation method for the battery pack, such as a controller or battery management system integrated in a vehicle. Specifically, the real-time health status estimation method for the battery pack is stored in the aforementioned device as a pre-programmed program. When the device is started, the program is invoked, and the real-time health status estimation method for the battery pack is implemented.

[0045] Compared with existing technologies, the real-time health status estimation method for battery packs provided in this invention collects the current state of charge of the battery pack, electrical and temperature signals of individual battery cells, mechanical signals of mechanical deformation between battery cells, and electrochemical impedance signals of the battery pack. The current state of charge can characterize the energy state of the battery pack, the electrical signals can reflect the conductivity and polarization behavior of individual battery cells, the mechanical signals can reflect the changing trend of lithium intercalation behavior inside individual battery cells and structural stress changes caused by abnormal side reactions or lithium plating, and the electrochemical impedance signals can reflect the battery interface state and transmission characteristics. Moreover, the mechanical and electrochemical impedance signals are real-time sensing signals. Based on the multi-source physical signals composed of the above-mentioned signals representing different dimensions, the health status estimation can be performed so that the health status estimation no longer depends on the lagging changes in macroscopic parameters. It is directly rooted in the core physicochemical processes that lead to performance degradation, realizing the real-time estimation of battery health status, and thus improving the response speed of subsequent fast charging regulation.

[0046] Furthermore, the embodiments of the present invention integrate multi-source physical signals for real-time health status estimation, overcoming the shortcomings of noise interference or incomplete characterization that may exist in a single signal source, and significantly improving the accuracy and robustness of health status estimation.

[0047] When multi-source physical signals are directly used for feature extraction, the inherent noise in the original signals, the delay errors caused by sensor thermal inertia, and the time asynchrony between different signal acquisition systems can severely affect the accuracy of subsequent feature extraction and the input quality of the fusion estimation model. Specifically, high-frequency switching noise in electrical signals, vibration interference in mechanical signals, response lag in temperature signals, and small deviations in the sampling time of each signal, if not properly addressed, will lead to distorted and unreliable extracted features. Ultimately, this results in large fluctuations and low reliability in the health status estimation results, making it impossible to provide stable input for high-precision control.

[0048] To solve this technical problem, in some embodiments of the present invention, before step S102, the following step is further included: High-frequency noise filtering is performed on electrical and mechanical signals to obtain optimized electrical and mechanical signals. The temperature signal is processed by delay compensation to obtain a temperature compensation signal; Time alignment is performed on the optimized electrical signals, optimized mechanical signals, temperature compensation signals, current state of charge, and electrochemical impedance signals.

[0049] Specifically, time alignment can be achieved through timestamp-based synchronization or resampling, aligning various signals on the same time reference.

[0050] This invention effectively filters out irrelevant noise and interference from electrical and mechanical signals by performing high-frequency noise filtering, extracting the effective components that truly reflect changes in the battery's internal state. Simultaneously, it compensates for temperature signals, correcting the measurement lag caused by sensor thermal inertia, making thermal state characterization more timely and accurate. Furthermore, it aligns various physical signals in time, ensuring strict alignment of signals from different physical dimensions and sampling rates on the time axis. This provides a time-consistent data foundation for subsequent multi-feature fusion modeling, resolving model misjudgments caused by signal asynchrony, and significantly improving the stability, consistency, and risk confidence of the final output of health state estimation.

[0051] To further improve the accuracy of real-time health status estimation results for battery packs, in some embodiments of the present invention, the temperature signal includes a first temperature signal at the tab location and a second temperature signal at the center of the casing.

[0052] Then the eigenvector is X = [ , , [, T1, T2, SOC], where T1 is the first temperature signal and T2 is the second temperature signal.

[0053] Specifically, the first temperature signal at the tab position can directly reflect the ohmic heat accumulation at the current collection point, which is the position most prone to overheating during fast charging, while the second temperature signal at the center of the casing is closer to the overall average thermal state of the battery cell.

[0054] By simultaneously monitoring the temperature at the tab location and the center of the casing, this invention can capture local overheating risks and assess the overall thermal management level. It provides key information for health status estimation models to distinguish between local hot spots and overall temperature rise, greatly enhancing the ability to provide early warning of thermal runaway.

[0055] It should be noted that when the internal materials of a battery cell undergo minute and rapid fractures or deformations (such as lithium dendrites piercing the separator, active material particles developing cracks, or electrode sheets undergoing micro-peeling), energy is released instantaneously, generating high-frequency stress waves.

[0056] Therefore, in some embodiments of the present invention, the multi-source physical signals may further include stress wave signals. Specifically, stress wave signals are acquired by attaching a high-frequency acoustic sensor to the surface of the battery cell casing.

[0057] It should also be noted that when a battery cell is overcharged or overheated, a series of side reactions occur (such as electrolyte decomposition and SEI film thickening), many of which produce gases (such as carbon dioxide, ethylene, and hydrogen). These gases accumulate inside the battery, leading to an increase in internal pressure.

[0058] Leveraging this characteristic, in some embodiments of the present invention, the multi-source physical signals may further include the internal pressure signal of the battery. Specifically, by adding a pressure sensor within the battery cell, changes in internal pressure are monitored in real time.

[0059] In specific embodiments of the present invention, such as Figure 2 As shown, the health status estimation model includes a depthwise separable convolutional (DSC) unit and a gated recurrent unit (GRU). The DSC unit is used to extract local features from the feature vector to obtain local features, and the GRU unit is used to capture the temporal evolution of local features to obtain real-time health factors.

[0060] In addition to the network structure described above, in some embodiments of the present invention, a support vector machine regression model can also be used to estimate the battery health status.

[0061] Once the actual health factors of the battery pack can be quickly and accurately estimated, the lag in the fast charging process can be eliminated. However, existing fast charging technologies still have the following technical problems: charging strategies are usually based on fixed charging curves, battery factory-calibrated aging models, or long-term health status values ​​with low update frequency. This presents a contradiction between safety and charging speed. Fixed or conservative strategies sacrifice charging efficiency to ensure safety, especially when the battery is in good condition, it cannot fully realize its fast charging potential.

[0062] To address this technical problem, embodiments of the present invention also provide a method for controlling fast charging of a battery pack, such as... Figure 3 As shown, the battery pack fast charging control method includes: S301. Obtain the real-time health factors of the battery pack; S302. Based on real-time health factors, battery pack temperature and current state of charge, generate charging control commands including charging control parameters in real time.

[0063] The temperature of the battery pack can be the average temperature of all individual battery cells.

[0064] Specifically, there are two ways to generate charging control commands: The first method involves pre-calibrating a three-dimensional dynamic lookup table. The three input axes of the three-dimensional dynamic lookup table are the instantaneous degradation index, the battery pack temperature, and the current state of charge, with the output being a charging control command. In other words, the charging control command is obtained by matching the instantaneous degradation index, battery pack temperature, and current state of charge within the three-dimensional dynamic lookup table.

[0065] The second method involves constructing a rule-based empirical formula. By inputting the instantaneous degradation index, the battery pack temperature, and the current state of charge into the empirical formula, charging control commands can be obtained.

[0066] S303: Responds to charging control commands to control the battery pack for fast charging; The real-time health factor is determined based on the battery pack real-time health status estimation method in any of the above embodiments.

[0067] This invention dynamically calculates real-time health factors, battery temperature, and state of charge to obtain charging control commands, transforming the charging process from a fixed mode into a dynamic process that fluctuates smoothly in real time according to the actual health condition of the battery. This not only precisely constrains the charging process within the battery's absolute safety zone at every moment to prevent cumulative damage, but also actively increases power to shorten charging time when conditions permit. Thus, while ensuring long battery life and high safety, it maximizes charging efficiency throughout the entire life cycle, achieving a dynamic optimal balance between safety, lifespan, and user experience.

[0068] In a specific embodiment of the present invention, the charging control parameters include current scaling factor, voltage offset, and current ramp rate limit.

[0069] Specifically, a current scaling factor of 0.3-1.0 is multiplied by the maximum permissible current requested by the charging station or the maximum permissible current specified in the standard protocol to obtain a new, lower maximum charging current limit. The charging station must limit the real-time output current below this value.

[0070] Among them, the current scaling factor can directly reduce the electrochemical reaction rate and heat generation inside the battery pack, and is the most effective means to prevent overcurrent and suppress lithium plating.

[0071] The voltage offset is used to fine-tune the charging termination voltage, for example, reducing the charging termination voltage by 10mV when the health status is poor.

[0072] Among them, the voltage offset setting can prevent the battery pack from entering the stress zone of excessively high voltage after aging, reduce side reactions, and delay capacity decay.

[0073] The current ramp-up rate limit is used to adjust the power ramp-up rate. Specifically, the current ramp-up rate limit is used to avoid causing severe current surges to the battery pack, allowing for a smoother transition of the electrochemical state and temperature field inside the battery.

[0074] If the actual response of the battery pack after the execution of the charging control command deviates significantly from the expected range due to model prediction errors, sudden changes in individual battery differences, or external disturbances, the battery pack may be in an overstressed state without being noticed. This could accelerate battery degradation or, in severe cases, lead to serious safety accidents such as thermal runaway.

[0075] To address this technical problem, in some embodiments of the present invention, such as... Figure 4 As shown, the battery pack fast charging control method also includes: S401. Obtain the battery indicator values ​​after executing the charging control command.

[0076] Among them, battery performance indicators include, but are not limited to, the rate of temperature rise and / or voltage response of the battery pack.

[0077] S402. When the battery indicator value exceeds the expected range, the safety fault tolerance strategy is triggered to perform derating charging or stop charging, and the battery indicator value is recorded.

[0078] It should be noted that the recorded battery index values ​​can be used to optimize and iterate the health status estimation model to make the health status estimation model more accurate.

[0079] This invention introduces a safety fault-tolerant closed loop based on real-time feedback, enabling continuous comparison of battery performance values ​​with expected safe ranges after charging parameters are adjusted. Upon detecting any abnormal deviation, preset fault-tolerant strategies such as derating or stopping charging are immediately triggered, cutting off risk transmission paths within milliseconds to seconds and nipping potential dangers in the bud. Simultaneously, all abnormal events triggering fault-tolerant strategies and their corresponding data are recorded and uploaded, providing samples for subsequent health status estimation models. This allows the health status estimation model to continuously learn and evolve.

[0080] On the other hand, embodiments of the present invention also provide a battery pack real-time health status estimation device, such as... Figure 5 As shown, the battery pack real-time health status estimation device 500 includes: The multi-source physical signal acquisition unit 501 is used to acquire multi-source physical signals of the battery pack during the charging process. The multi-source physical signals include the current state of charge of the battery pack, the electrical signals of the individual battery cells, the temperature signals of the individual battery cells, the mechanical signals between the individual battery cells, and the electrochemical impedance signals of the individual battery cells. The feature extraction unit 502 is used to extract features from multi-source physical signals to obtain instantaneous ohmic internal resistance, expansion rate of change and relative rate of change of electrochemical impedance, and to form a feature vector based on instantaneous ohmic resistance. The real-time health factor determination unit 503 is used to input the feature vector into the pre-trained health status estimation model and output real-time health factors that reflect the current health status of the battery.

[0081] The battery pack real-time health status estimation device 500 provided in the above embodiments can realize the technical solutions described in the above battery pack real-time health status estimation method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above battery pack real-time health status estimation method embodiments, and will not be repeated here.

[0082] On the other hand, embodiments of the present invention also provide a battery pack fast charging control device, such as... Figure 6 As shown, the battery pack fast charging control device 600 includes: The real-time health factor acquisition unit 601 is used to acquire the real-time health factors of the battery. The charging control command generation unit 602 is used to generate charging control commands, including charging control parameters, in real time based on real-time health factors, battery temperature and current state of charge. The fast charging control unit 603 is used to control the battery pack to perform fast charging in response to charging control commands. The real-time health factor is determined based on the battery pack real-time health status estimation method in any of the above embodiments.

[0083] The battery pack fast charging control device 600 provided in the above embodiments can realize the technical solutions described in the above battery pack fast charging control method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above battery pack real-time health status estimation method embodiments, which will not be repeated here.

[0084] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0085] The present invention provides a detailed description of a method for estimating the real-time health status of a battery pack, a fast-charging control method, and a device. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for estimating the real-time health status of a battery pack, characterized in that, The battery pack includes multiple battery cells, and the method includes: The battery pack is collected from multiple sources of physical signals during the charging process. These multiple sources of physical signals include: the current state of charge of the battery pack, the electrical and temperature signals of individual battery cells, the mechanical signals of mechanical deformation between individual battery cells, and the electrochemical impedance signal of the battery pack. Feature extraction is performed on the multi-source physical signal to obtain a feature vector including instantaneous ohmic internal resistance, expansion rate, and relative electrochemical impedance rate. The feature vector is input into a pre-trained health status estimation model to obtain a real-time health factor that reflects the current internal health status of the battery pack.

2. The method for estimating the real-time health status of a battery pack according to claim 1, characterized in that, Before performing feature extraction on the multi-source physical signal, the method further includes: The electrical signals and the mechanical signals are subjected to high-frequency noise filtering to obtain optimized electrical signals and optimized mechanical signals. The temperature signal is subjected to delay compensation processing to obtain a temperature compensation signal; The optimized electrical signal, the optimized mechanical signal, the temperature compensation signal, the current state of charge, and the electrochemical impedance signal are time-aligned.

3. The method for estimating the real-time health status of a battery pack according to claim 1 or 2, characterized in that, The battery cell includes a tab and a casing. The electrical signals include the terminal voltage and charging current of the battery cell. The temperature signals include a first temperature signal at the location of the tab and a second temperature signal at the center of the casing.

4. The method for estimating the real-time health status of a battery pack according to claim 1, characterized in that, The health status estimation model includes a depthwise separable convolutional unit and a gated recurrent unit. The depthwise separable convolutional unit is used to extract local features from the feature vector to obtain local features, and the gated recurrent unit is used to capture the temporal evolution features of the local features to obtain the real-time health factor.

5. The method for estimating the real-time health status of a battery pack according to claim 1, characterized in that, The real-time health factors include an instantaneous degradation index characterizing the battery pack’s ability to withstand electrochemical and thermal stresses under current charging conditions, and a risk confidence level characterizing the reliability of the estimate.

6. A method for controlling fast charging of a battery pack, characterized in that, include: Obtain real-time health factors for the battery pack; Based on the real-time health factors, the temperature of the battery pack, and the current state of charge, a charging control command including charging control parameters is generated in real time. The battery pack is controlled to perform fast charging in response to the charging control command. The real-time health factor is determined based on the battery pack real-time health status estimation method according to any one of claims 1-5.

7. The battery pack fast charging control method according to claim 6, characterized in that, The charging control parameters include current scaling factor, voltage offset, and current ramp rate limit.

8. The battery pack fast charging control method according to claim 7, characterized in that, The method further includes: Obtain the battery indicator values ​​after executing the charging control command; When the battery indicator value exceeds the expected range, a safety fault tolerance strategy is triggered to perform derating charging or stop charging, and the battery indicator value is recorded.

9. A device for estimating the real-time health status of a battery pack, characterized in that, The battery pack includes multiple battery cells, and the device includes: A multi-source physical signal acquisition unit is used to acquire multi-source physical signals of the battery pack during the charging process. The multi-source physical signals include the current state of charge of the battery pack, the electrical signals of the individual battery cells, the temperature signals of the individual battery cells, the mechanical signals between the individual battery cells, and the electrochemical impedance signals of the individual battery cells. The feature extraction unit is used to extract features from the multi-source physical signal to obtain the instantaneous ohmic internal resistance, the rate of change of expansion and the relative rate of change of electrochemical impedance, and to form a feature vector based on the instantaneous ohmic resistance. The real-time health factor determination unit is used to input the feature vector into a pre-trained health status estimation model and output real-time health factors that reflect the current health status of the battery.

10. A battery pack fast charging control device, characterized in that, include: Real-time health factor acquisition unit, used to acquire the real-time health factors of the battery; A charging control command generation unit is used to generate charging control commands, including charging control parameters, in real time based on the real-time health factors, battery temperature, and current state of charge. Fast charging control unit, used to control the battery pack to perform fast charging in response to the charging control command; The real-time health factor is determined based on the battery pack real-time health status estimation method according to any one of claims 1-5.

Images