A backup battery charging and discharging system

Abstract

The application discloses a backup battery charging and discharging system and relates to the technical field of battery charging and discharging. The system comprises a parameter acquisition module, a dynamic response signal generation module, a signal analysis module, a feature fusion module and a control decision module. The dynamic response signal generation module is used for loading a composite micro-disturbance signal to the backup battery in the constant current charging and discharging stage. The signal analysis module is used for performing relaxation time distribution analysis on the dynamic response signal, separating out a first characteristic subset representing the charge transfer process and a second characteristic subset representing the substance diffusion process. The feature fusion module is used for performing normalized weighting on the first characteristic subset in the guidance of the charge transfer rate and performing normalized weighting on the second characteristic subset in the guidance of the diffusion impedance. The application can improve the adaptability of the charging and discharging process to the real state of the battery, effectively optimize the charging and discharging efficiency and the capacity output, and reduce the loss of the battery caused by non-adaptive charging and discharging.

Description

Technical Field

[0001] This invention relates to the field of battery charging and discharging technology, and more particularly to a backup battery charging and discharging system. Background Technology

[0002] During the charging and discharging process of backup batteries, the core requirement is to achieve precise matching between charging and discharging performance and battery state. Existing technologies mostly adopt uniform and fixed charging and discharging parameters or simple linear adjustment strategies, which fail to fully consider the individual electrochemical characteristics differences formed by different batteries under the influence of factors such as production process and usage history, and also cannot dynamically sense the real-time changes in the internal state of the battery during charging and discharging, resulting in a mismatch between the charging and discharging trajectory and the actual state of the battery.

[0003] The reason why existing technologies are difficult to achieve precise adaptation is that they lack the ability to effectively characterize and dynamically analyze the key electrochemical processes inside the battery. The charging and discharging process of a battery involves multiple complex and dynamically changing core processes such as charge transfer and mass diffusion. The characteristics of these processes directly determine the real-time state of the battery.

[0004] However, existing technologies cannot accurately separate and quantify these processes, making it difficult to obtain key characteristic information that reflects the true state of the battery. The lack of a reliable state characterization basis means that the adjustment of charging and discharging parameters can only rely on empirical fixed rules, and cannot be dynamically optimized according to the real-time state of the individual battery. Ultimately, this makes it difficult to achieve the ideal level of charging and discharging performance. Summary of the Invention

[0005] The present invention provides a backup battery charging and discharging system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a backup battery charging and discharging system, comprising a parameter acquisition module, a dynamic response signal generation module, a signal analysis module, a feature fusion module, and a control decision module, wherein:

[0007] The parameter acquisition module is used to acquire the reference electrochemical parameter set of the target backup battery;

[0008] The dynamic response signal generation module is used to apply a composite micro-perturbation signal to the backup battery during the constant current charging and discharging phase, and simultaneously acquire the loop current to obtain the dynamic response signal.

[0009] The signal analysis module is used to perform relaxation time distribution analysis on dynamic response signals, separating the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process.

[0010] The feature fusion module is used to normalize and weight the first feature subset based on charge transfer rate, and at the same time normalize and weight the second feature subset based on diffusion impedance. The two weighted feature sets are then fused into a single vector to obtain the real-time state evaluation vector.

[0011] The control decision module is used to periodically acquire real-time state evaluation vectors. By continuously analyzing the relative magnitude and direction of change of two sets of weights in multiple vectors, it dynamically back-calculates the attenuation slope of the charging current or the rise threshold of the discharge cutoff voltage, forming an individualized charging and discharging trajectory.

[0012] Preferably, when the parameter acquisition module acquires the reference electrochemical parameter set of the target backup battery, it is specifically used for:

[0013] After the backup battery has been left to stand and reach electrochemical equilibrium, its open-circuit voltage and temperature are measured as a steady-state reference.

[0014] Apply DC current excitation to the backup battery and record the time-domain curve of its voltage recovery process after the excitation is removed;

[0015] Based on the decay characteristics of the voltage recovery time-domain curve, the fast relaxation time constant and the slow relaxation time constant are extracted as dynamic reference parameters.

[0016] The reference electrochemical parameter set is composed of steady-state reference parameters and dynamic reference parameters.

[0017] Preferably, the dynamic response signal generation module, when performing the loading of a composite micro-perturbation signal onto the backup battery during the constant current charge / discharge phase, is specifically used for:

[0018] Based on the fast relaxation time constant in the reference electrochemical parameter set, the short-period fluctuation component is determined;

[0019] Based on the lumped slow relaxation time constant of the reference electrochemical parameters, the long-period fluctuation component is determined.

[0020] By superimposing the short-period fluctuation component with the long-period fluctuation component, a composite micro-perturbation signal is obtained.

[0021] Preferably, when performing relaxation time distribution analysis on the dynamic response signal, the signal analysis module is specifically used for:

[0022] Based on the correspondence between the dynamic response signal and the composite micro-perturbation signal, the frequency domain impedance response of the backup battery in the current state is established;

[0023] The frequency domain impedance response is converted into a continuous distribution spectrum with the relaxation time constant as the abscissa and the characteristic intensity as the ordinate, thus obtaining the relaxation time distribution spectrum.

[0024] Preferably, when the signal analysis module separates the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process, it is specifically used for:

[0025] On the relaxation time distribution spectrum, determine the relative distances between the time constants of each characteristic distribution and the fast relaxation time constant and the slow relaxation time constant, respectively;

[0026] Among all feature distributions, those that are relatively closer to the fast relaxation time constant are assigned to the first feature subset.

[0027] Among all the feature distributions, those that are relatively closer to the slow relaxation time constant are assigned to the second feature subset.

[0028] Preferably, when the feature fusion module performs normalization weighting of the first feature subset based on charge transfer rate, it is specifically used for:

[0029] Obtain the change in the area of ​​each feature distribution in the first feature subset over adjacent analysis periods;

[0030] The weight value is determined based on the proportion of the change in each characteristic distribution to the total change in the subset.

[0031] Preferably, when the feature fusion module performs normalization weighting of the second feature subset based primarily on diffusion impedance, it is specifically used for:

[0032] The trend of the peak intensity of each feature distribution in the second feature subset changes with the battery's discharged or charged capacity, thus obtaining the feature distribution trend;

[0033] The weight values ​​are determined based on the monotonicity and steepness of the characteristic distribution trend.

[0034] Preferably, when the feature fusion module performs the operation of fusing the two weighted sets of features into a single vector to obtain the real-time state evaluation vector, it is specifically used for:

[0035] Using the weight values ​​determined for the first feature subset, the corresponding feature distribution areas are weighted and summed to obtain the first fusion value;

[0036] Using the weight values ​​determined for the second feature subset, the corresponding feature distribution areas are weighted and summed to obtain the second fusion value;

[0037] The real-time state evaluation vector is composed of the first fusion value and the second fusion value.

[0038] Preferably, when the control decision module performs dynamic reverse calculation of the charging current attenuation slope, it is specifically used for:

[0039] Arrange multiple real-time state evaluation vectors in chronological order to obtain a real-time state evaluation vector sequence;

[0040] The first fusion sequence is obtained by extracting a subsequence of the first fusion value changing over time from the real-time state evaluation vector sequence;

[0041] The first fused value sequence is subjected to continuous difference processing to obtain the change in its instantaneous rate of decrease;

[0042] Identify periods in which the instantaneous rate of descent continuously increases as the acceleration phase of descent;

[0043] Based on the duration and rate of increase of the descent acceleration phase, the rate of change of the constant current charging current with time is determined as the individualized charging current decay slope.

[0044] Preferably, when the information extraction module performs the reverse discharge cutoff voltage rise threshold calculation, it is specifically used for:

[0045] The second fusion sequence is obtained by extracting a subsequence of the second fusion value changing over time from the real-time state evaluation vector sequence;

[0046] By differentiating the second fusion sequence, the variation curve of the second fusion sequence is obtained;

[0047] Locate the inflection point on the curve where the voltage transitions from a stable phase to an increasing phase, and determine the inflection point as the threshold for raising the discharge cutoff voltage.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. By adopting a feature fusion strategy guided by charge transfer rate and dominated by diffusion impedance, the features of key electrochemical processes inside the battery are accurately integrated into a real-time state assessment vector. Based on this vector, key charging and discharging parameters are dynamically deduced, which can deeply match the dynamic state changes of individual batteries, significantly improve the adaptability of the charging and discharging process to the actual state of the battery, effectively optimize charging and discharging efficiency and capacity output, and reduce the battery loss caused by mismatched charging and discharging.

[0050] 2. By leveraging the synergistic support of benchmark electrochemical parameters in both steady-state and dynamic states, and through precise matching of composite micro-perturbation signals and relaxation time distribution analysis, efficient separation and quantitative evaluation of feature subsets are achieved. This further enhances the characterization accuracy and reliability of real-time state evaluation vectors, making the dynamic adjustment of individualized charge-discharge trajectories more scientific and targeted, thereby further ensuring the stability of battery charge-discharge and extending battery cycle life. Attached Figure Description

[0051] Figure 1This is a system architecture diagram of a backup battery charging and discharging system provided in an embodiment of the present invention;

[0052] Figure 2 This is a schematic diagram of the execution flow of a signal analysis module according to an embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of the execution flow of the feature fusion module provided in an embodiment of the present invention.

[0054] The realization of the objective, functional characteristics, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0055] 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 belong to some, but not all, embodiments of the present invention. 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.

[0056] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0057] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0058] In practice, the server-side equipment deployed in a backup battery charging and discharging system may consist of one or more devices. The aforementioned backup battery charging and discharging system can be implemented as: a business instance, a virtual machine, or hardware devices. For example, this backup battery charging and discharging system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, this backup battery charging and discharging system can be understood as software deployed on a cloud node, used to provide a backup battery charging and discharging system for each user terminal. Alternatively, this backup battery charging and discharging system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing each user terminal. Alternatively, this backup battery charging and discharging system can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide a backup battery charging and discharging system for each user terminal.

[0059] In terms of implementation, the backup battery charging and discharging system and the user terminal are mutually compatible. That is, if the backup battery charging and discharging system is implemented as an application installed on a cloud service platform, then the user terminal is implemented as a client that establishes a communication connection with the application; or if the backup battery charging and discharging system is implemented as a website, then the user terminal is implemented as a webpage; or if the backup battery charging and discharging system is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0060] Example 1, as Figure 1 The diagram shown is a system architecture diagram of a backup battery charging and discharging system provided in an embodiment of the present invention.

[0061] This invention discloses a backup battery charging and discharging system that can be hosted on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, a backup battery charging and discharging system may include a parameter acquisition module, a dynamic response signal generation module, a signal analysis module, a feature fusion module, and a control decision module. The modules of this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0062] In this embodiment of the invention, in a backup battery charging and discharging system, each of the above-mentioned modules can be implemented independently and can call other modules. Here, "calling" can be understood as a module connecting to multiple modules of another type and providing corresponding services to those connected modules. In the backup battery charging and discharging system provided by this embodiment of the invention, without modifying the program code, the applicability of a backup battery charging and discharging system architecture can be adjusted by adding modules and directly calling them, achieving cluster-based horizontal expansion to quickly and flexibly expand the backup battery charging and discharging system. In practical applications, the above-mentioned modules can be set in the same device or different devices, or they can be set in a virtual device, such as a service instance in a cloud server.

[0063] The following describes, with reference to specific embodiments, each component of a backup battery charging and discharging system and its specific workflow:

[0064] The parameter acquisition module is used to acquire the reference electrochemical parameter set of the target backup battery;

[0065] The dynamic response signal generation module is used to apply a composite micro-perturbation signal to the backup battery during the constant current charging and discharging phase, and simultaneously acquire the loop current to obtain the dynamic response signal.

[0066] The signal analysis module is used to perform relaxation time distribution analysis on dynamic response signals, separating the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process.

[0067] The feature fusion module is used to normalize and weight the first feature subset based on charge transfer rate, and at the same time normalize and weight the second feature subset based on diffusion impedance. The two weighted feature sets are then fused into a single vector to obtain the real-time state evaluation vector.

[0068] The control decision module is used to periodically acquire real-time state evaluation vectors. By continuously analyzing the relative magnitude and direction of change of two sets of weights in multiple vectors, it dynamically back-calculates the attenuation slope of the charging current or the rise threshold of the discharge cutoff voltage, forming an individualized charging and discharging trajectory.

[0069] In this embodiment, when the parameter acquisition module acquires the reference electrochemical parameter set of the target backup battery, it is specifically used for:

[0070] After the backup battery has been left to stand and reach electrochemical equilibrium, its open-circuit voltage and temperature are measured as a steady-state reference.

[0071] Apply DC current excitation to the backup battery and record the time-domain curve of its voltage recovery process after the excitation is removed;

[0072] Based on the decay characteristics of the voltage recovery time-domain curve, the fast relaxation time constant and the slow relaxation time constant are extracted as dynamic reference parameters.

[0073] The reference electrochemical parameter set is composed of steady-state reference parameters and dynamic reference parameters.

[0074] Specifically, the target backup battery is placed in an environment with no current input or output for a continuous period of time until the battery voltage and temperature remain constant and no longer fluctuate over time. At this point, the battery is determined to have reached an electrochemical equilibrium state. A voltage measuring device is used to directly collect the battery open-circuit voltage under this state, and a temperature sensing device is used to closely attach to the battery casing to collect the real-time temperature. The collected open-circuit voltage and temperature together form a steady-state reference.

[0075] Furthermore, a DC power supply circuit is connected to the target backup battery that has reached electrochemical equilibrium, so that a DC current of constant amplitude is formed in the circuit and continues to act for a preset time. Then, the DC power supply circuit is disconnected to terminate the current excitation. From the moment the circuit is disconnected, the battery voltage data is continuously collected at equal time intervals. The voltage values ​​corresponding to each time point are arranged one by one in chronological order to form a time-domain curve of the voltage recovery process.

[0076] Furthermore, observing the decay characteristics of the voltage recovery time-domain curve, the curve shows a pattern of rapid voltage recovery in the early stage and a gradual slowdown in the later stage. The rapid recovery stage in the early stage corresponds to the fast relaxation process. The time interval in this stage during which the voltage changes from the initial recovery value to a fixed proportion of the stable recovery value is identified, and the characteristic time parameter of this time interval is extracted as the fast relaxation time constant. The slow recovery stage in the later stage corresponds to the slow relaxation process. The characteristic time interval in this stage during which the voltage approaches the final stable value is identified, and the characteristic time parameter of this time interval is extracted as the slow relaxation time constant. The above two time constants together constitute the dynamic reference parameter.

[0077] Furthermore, the open-circuit voltage and temperature parameters included in the steady-state reference are integrated and grouped with the fast relaxation time constant and slow relaxation time constant parameters included in the dynamic reference to form a complete set covering the four types of parameters. This set is the reference electrochemical parameter set for the target backup battery.

[0078] In summary, the parameter acquisition module is the core foundation for the entire charging and discharging system to achieve accurate adaptation and dynamic optimization. First, by collecting steady-state and dynamic benchmarks, this module constructs a complete parameter system covering the static equilibrium state and dynamic response characteristics of the battery. It accurately captures the individualized electrochemical characteristics of different batteries due to differences in manufacturing processes and usage history, solving the pain point of existing technologies neglecting individual battery differences and providing a unique data profile for the subsequent formulation of individualized charging and discharging strategies.

[0079] Secondly, the fast and slow relaxation time constants in the reference parameter set provide a key design basis for the dynamic response signal generation module, enabling it to generate composite micro-perturbation signals that match the battery's own dynamic characteristics. This avoids response distortion caused by incompatibility between the perturbation signal and the battery's internal processes, and ensures that the dynamic response signal can effectively reflect the battery's true state, providing high-quality input for subsequent signal analysis.

[0080] Then, this parameter set provides a core reference standard for the relaxation time distribution analysis of the signal analysis module. By comparing the feature distribution with the fast and slow relaxation time constants, the accurate separation of the feature subsets of charge transfer and mass diffusion processes is achieved, solving the problem that existing technologies cannot quantitatively evaluate the key internal electrochemical processes, and laying a solid data foundation for subsequent feature fusion and state assessment.

[0081] Finally, the synergistic support of steady-state and dynamic parameters allows the system to perceive the battery state in a way that is both stable and dynamic. It not only clarifies the initial baseline state of the battery but also captures the time-scale characteristics of its internal processes. This enables subsequent adjustments to charge and discharge parameters to not only conform to the individual baseline characteristics of the battery but also dynamically adapt to state changes during the charge and discharge process, ultimately improving charge and discharge efficiency and capacity output while reducing battery loss.

[0082] In this embodiment of the invention, the dynamic response signal generation module applies a composite micro-perturbation signal to the backup battery during the constant current charge and discharge phase, specifically for:

[0083] Based on the fast relaxation time constant in the reference electrochemical parameter set, the short-period fluctuation component is determined;

[0084] Based on the lumped slow relaxation time constant of the reference electrochemical parameters, the long-period fluctuation component is determined.

[0085] By superimposing the short-period fluctuation component with the long-period fluctuation component, a composite micro-perturbation signal is obtained.

[0086] The formula for calculating composite micro-perturbation signals is as follows:

[0087]

[0088] Individualized time constant, This represents the scaling factor, used to match the disturbance period with... , The magnitude.

[0089] This represents the amplitude of the short-period component. This represents the amplitude of the long-period component.

[0090] Composite micro-perturbation signals customized based on individual battery dynamic characteristics.

[0091] Specifically, the fast relaxation time constant is extracted from the set of reference electrochemical parameters. Multiple constant current charge-discharge comparative tests were conducted on similar backup batteries, and the proportional coefficient was adjusted in each test group. The value of the voltage response is selected and the fluctuation range and stability of the battery voltage response are continuously observed. The selected value ensures that the voltage response fluctuation range is within a preset reasonable range and does not interfere with the normal electrochemical process of the battery. The numerical value serves as a fixed proportionality coefficient, used to match the period of the short-period fluctuation component with the fast relaxation time constant. The magnitude.

[0092] Simultaneously, loading tests with micro-perturbations of different amplitudes were conducted on backup batteries of the same type. The charging and discharging efficiency and the maintenance of electrochemical equilibrium under different amplitudes were observed. A fixed amplitude that would not affect the charging and discharging performance of the battery was selected as the amplitude of the short-period fluctuation component. Based on the fast relaxation time constant With a fixed proportional coefficient Determine the period of the short-period fluctuation component so that it matches the timescale of the battery's fast relaxation process, and then combine it with a fixed amplitude. A current signal that changes continuously with time is generated according to a sinusoidal variation law.

[0093] It should be noted that this current signal fluctuates periodically over time, with the fluctuation amplitude remaining constant within each period. Furthermore, it will not attenuate, and this current signal is the short-period fluctuation component.

[0094] Furthermore, the slow relaxation time constant was extracted from the reference electrochemical parameter set. Multiple sets of long-term constant current charge-discharge observation tests were conducted on similar backup batteries, and the proportional coefficient was adjusted in each set of tests. The value of [value] is selected and the long-term electrochemical stability and capacity decay of the battery are continuously tracked. The selected value ensures that the battery's long-term charge-discharge performance remains stable and the fluctuations meet preset requirements. The numerical value serves as a fixed proportionality coefficient, used to match the period of the long-period fluctuation component with the slow relaxation time constant. The magnitude.

[0095] Simultaneously, loading tests with long-cycle micro-perturbations of different amplitudes were conducted on backup batteries of the same type. The cycle life and electrochemical impedance changes of the batteries under different amplitudes were observed. A fixed amplitude that would not accelerate battery aging and would not affect the battery's energy storage performance was selected as the amplitude of the long-cycle fluctuation component. Based on the slow relaxation time constant With a fixed proportional coefficient Determine the period of the long-period fluctuation component so that it matches the timescale of the battery's slow relaxation process, and then combine it with a fixed amplitude. A current signal that changes continuously with time is generated according to a sinusoidal variation law.

[0096] It should be noted that this current signal fluctuates periodically over time, with the fluctuation amplitude remaining constant within each period. Furthermore, it will not attenuate, and this current signal is a long-period fluctuation component.

[0097] Furthermore, the generated short-period fluctuation component and long-period fluctuation component are placed on the same time axis. At each same time node, the current value corresponding to the short-period fluctuation component and the current value corresponding to the long-period fluctuation component are algebraically added to obtain the composite micro-perturbation current value at that time node.

[0098] Furthermore, the composite micro-perturbation current values ​​corresponding to all time points are arranged sequentially according to time order to form a continuous current signal that changes dynamically with time.

[0099] It should be noted that the trend of the current signal here is that the rapid fluctuations of the short-period fluctuation component are superimposed on the slow fluctuations of the long-period fluctuation component, exhibiting an overall periodic composite fluctuation characteristic, and the fluctuation amplitude remains stable. This current signal is a composite micro-perturbation signal customized based on the individual battery dynamic characteristics. .

[0100] In summary, the composite micro-perturbation signal in this embodiment is customized based on the fast and slow relaxation time constants of the reference electrochemical parameter set. The short-period and long-period fluctuation components are precisely matched with the time scale characteristics of charge transfer and material diffusion inside the battery. This avoids the defect that a single perturbation signal cannot fully excite the dynamic response inside the battery, and does not interfere with the main process of constant current charging and discharging, thus achieving the effect of "perturbation without power disturbance".

[0101] In summary, this embodiment synchronously acquires loop current to obtain dynamic response signals, ensuring that the time axis of disturbance input and response output is aligned. This provides a data foundation with both synchronization and reliability for subsequent signal analysis modules to establish frequency domain impedance response and invert relaxation time distribution spectrum, avoiding feature extraction distortion caused by timing misalignment.

[0102] In summary, this embodiment enables the system to capture the internal state changes of the battery in real time during the charging and discharging process through dynamic perturbation and response acquisition. This provides a core data source for subsequent separation of key feature subsets and generation of real-time state evaluation vectors. It is a key link in realizing dynamic optimization of charging and discharging trajectories and accurate adaptation to the actual state of the battery, indirectly ensuring improved charging and discharging efficiency and reduced battery loss.

[0103] Example 2, as Figure 2 The diagram shown is a schematic diagram of the execution flow of the signal analysis module provided in an embodiment of the present invention.

[0104] In this embodiment of the invention, when the signal analysis module performs relaxation time distribution analysis on the dynamic response signal, it is specifically used for:

[0105] Based on the correspondence between the dynamic response signal and the composite micro-perturbation signal, the frequency domain impedance response of the backup battery in the current state is established;

[0106] The frequency domain impedance response is converted into a continuous distribution spectrum with the relaxation time constant as the abscissa and the characteristic intensity as the ordinate, thus obtaining the relaxation time distribution spectrum.

[0107] The relaxation time distribution spectrum is obtained by inversion using the following formula:

[0108]

[0109] This represents the frequency domain impedance response of the battery in its current state, obtained by analyzing the dynamic response signal and the composite micro-perturbation signal.

[0110] Represents the ohmic resistance at the high-frequency limit. Represents angular frequency. It represents the imaginary unit.

[0111] This represents the relaxation time distribution spectrum obtained by inversion using the above formula.

[0112] In this embodiment of the invention, when the signal analysis module separates the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process, it is specifically used for:

[0113] On the relaxation time distribution spectrum, determine the relative distances between the time constants of each characteristic distribution and the fast relaxation time constant and the slow relaxation time constant, respectively;

[0114] Among all feature distributions, those that are relatively closer to the fast relaxation time constant are assigned to the first feature subset.

[0115] Among all the feature distributions, those that are relatively closer to the slow relaxation time constant are assigned to the second feature subset.

[0116] Specifically, the dynamic response signal output by the backup battery after being subjected to a composite micro-perturbation signal is acquired. This dynamic response signal is then synchronized with the applied composite micro-perturbation signal on the time axis. Frequency domain transformation is performed on the two synchronized signals to extract the amplitude and phase relationships at different angular frequencies. The battery impedance value at the corresponding angular frequency is calculated using the basic definition of impedance. The impedance values ​​corresponding to all angular frequencies are arranged in angular frequency order, and the resulting set represents the frequency domain impedance response of the backup battery in the current state. .

[0117] Specifically, angular frequency It is inversely proportional to the period of the composite micro-perturbation signal. It characterizes the impedance features corresponding to different frequency disturbances under the current battery state. As the angular frequency increases, the impedance value gradually approaches a stable limit value.

[0118] Furthermore, high-frequency micro-perturbation tests were conducted on backup batteries of the same type. The frequency of the micro-perturbation signal was continuously increased until the battery's charge transfer and mass diffusion processes could no longer respond to the perturbation. The battery impedance value measured at this point is the ohmic resistance under the high-frequency limit. frequency domain impedance response Impedance values ​​corresponding to different angular frequencies and their determination Substitute the values ​​into the formula to perform the inversion calculation.

[0119] Specifically, the relaxation time constant is used in the inversion process. As variables, through continuous fitting of different The corresponding feature intensity makes the calculated result of the formula consistent with the measured result. A perfect match is obtained, ultimately yielding a relaxation time constant. The continuous distribution curve with the x-axis as the horizontal axis and the characteristic intensity as the vertical axis is the relaxation time distribution spectrum. .

[0120] Specifically, by establishing the correlation between frequency domain impedance and relaxation time distribution, when The smaller the value, the better it reflects the battery response under high-frequency disturbances. The larger the value, the better it reflects the battery response under low-frequency disturbances.

[0121] Furthermore, from the relaxation time distribution spectrum The system identifies all characteristic distributions with distinct intensity peaks, records the relaxation time constant corresponding to each characteristic distribution, and calculates the absolute value of the difference between the relaxation time constant of each characteristic distribution and the fast relaxation time constant in the reference electrochemical parameter set, as well as the absolute value of the difference between the fast and slow relaxation time constants. The relationship between the magnitudes of the two absolute values ​​is the relative distance between the time constant of each characteristic distribution and the fast and slow relaxation time constants, respectively.

[0122] Furthermore, all feature distributions in which the absolute value of the difference between the fast relaxation time constant and the slow relaxation time constant is less than the absolute value of the difference between the fast relaxation time constant and the slow relaxation time constant are selected. The relaxation processes corresponding to these feature distributions are directly related to the charge transfer process inside the battery. All the selected feature distributions together form the first feature subset characterizing the charge transfer process.

[0123] Furthermore, all feature distributions in which the absolute value of the difference between the feature distribution and the slow relaxation time constant is less than the absolute value of the difference between the feature distribution and the fast relaxation time constant are selected. The relaxation processes corresponding to these feature distributions are directly related to the material diffusion process inside the battery. All the selected feature distributions together form the second feature subset characterizing the material diffusion process.

[0124] In summary, this embodiment transforms the dynamic response signal into a continuous distribution spectrum with the relaxation time constant as the abscissa and the characteristic intensity as the ordinate through relaxation time distribution analysis. This successfully visualizes and quantifies the abstract internal electrochemical process of the battery, solving the pain point that traditional technologies cannot effectively separate the core processes of charge transfer and mass diffusion, and providing a new technical path for accurately assessing battery status.

[0125] In summary, this embodiment uses the fast and slow relaxation time constants in the benchmark electrochemical parameter set as a reference to achieve precise separation of feature subsets. Features closer to the fast relaxation time constant are assigned to the first subset representing charge transfer, while those closer to the slow relaxation time constant are assigned to the second subset representing material diffusion. This ensures the purity of the two types of core process features and avoids evaluation distortion caused by mutual interference between different process features.

[0126] In summary, this embodiment correlates the dynamic response signal with the composite micro-perturbation signal, establishes the frequency domain impedance response, and inverts the relaxation time distribution spectrum, thereby achieving a deep binding between the signal and the internal dynamic characteristics of the battery. This enables the extracted feature subset to truly reflect the key process characteristics of the battery in real time, providing high-purity and highly correlated core data support for subsequent feature fusion.

[0127] In summary, the analysis results in this embodiment directly determine the representation accuracy of the real-time state evaluation vector. The separated feature subsets provide a clear object for the targeted weighting of the feature fusion module, enabling subsequent adjustments to charging and discharging parameters, such as the charging current decay slope and the discharge cutoff voltage rise threshold, to accurately match changes in the battery's internal processes. This indirectly ensures the achievement of the system goals of optimizing charging and discharging efficiency, improving capacity output, and reducing battery losses.

[0128] Example 3, as Figure 3 The diagram shown is a schematic diagram of the execution flow of the feature fusion module provided in an embodiment of the present invention.

[0129] In this embodiment of the invention, when the feature fusion module performs normalization weighting of the first feature subset based on charge transfer rate, it is specifically used for:

[0130] Obtain the change in the area of ​​each feature distribution in the first feature subset over adjacent analysis periods;

[0131] The weight value is determined based on the proportion of the change in each characteristic distribution to the total change in the subset.

[0132] Specifically, each feature distribution in the first feature subset is extracted. For each feature distribution, its corresponding relaxation time constant interval is determined on the relaxation time distribution spectrum. The boundary of this interval is the two relaxation time constant points where the feature intensity of the feature distribution drops to zero. The area of ​​each feature distribution in the interval within the current analysis period is calculated by integration. At the same time, the area data of the same feature distribution in the corresponding interval within the previous adjacent analysis period is retrieved. The area of ​​the current analysis period is subtracted from the area of ​​the previous adjacent analysis period. The result is the change in the area of ​​the feature distribution within the adjacent analysis period.

[0133] Furthermore, the area changes of all feature distributions in the first feature subset are summarized, and the sum of the absolute values ​​of these changes is calculated as the total change of the subset. For each feature distribution, the ratio of the absolute value of its area change to the total change of the subset is calculated.

[0134] The ratio here is the weight value corresponding to the feature distribution. The magnitude of the weight value directly reflects the degree of contribution of the change in the area of ​​the corresponding feature distribution to the overall change of the first feature subset. The higher the degree of contribution, the larger the weight value.

[0135] In this invention, when the feature fusion module performs normalization weighting on the second feature subset based primarily on diffusion impedance, it is specifically used for:

[0136] The trend of the peak intensity of each feature distribution in the second feature subset changes with the battery's discharged or charged capacity, thus obtaining the feature distribution trend;

[0137] The weight values ​​are determined based on the monotonicity and steepness of the characteristic distribution trend.

[0138] Specifically, the capacity data of the backup battery that has been discharged or charged during the charging and discharging process is continuously recorded. At the same time, the peak intensity of each feature distribution in the second feature subset is recorded synchronously in chronological order. The peak intensity data of the same feature distribution at different capacity nodes are sequentially associated and arranged in order of increasing or decreasing capacity. By comparing the direction and magnitude of the change in peak intensity between adjacent capacity nodes segment by segment, the law of change of the peak intensity of the feature distribution with the capacity discharged or charged by the battery is determined. This law is the feature distribution trend.

[0139] Furthermore, the characteristic distribution trend of each characteristic distribution in the second characteristic subset is analyzed to determine the monotonicity of the trend. If the peak intensity always increases or always decreases with the change in capacity, it is determined to be a monotonic trend; otherwise, it is determined to be a non-monotonic trend.

[0140] Simultaneously, the average value of the change in peak intensity between adjacent capacity nodes in the trend is calculated. This average value reflects the steepness of the trend; the larger the average value, the steeper the trend.

[0141] Furthermore, the weights of monotonic trends are assigned higher priority than those of non-monotonic trends. Under the same monotonicity, the steeper the feature distribution, the greater the weight value is assigned. The weight value corresponding to each feature distribution is determined in this way. The magnitude of the weight value is positively correlated with the sensitivity of the feature distribution trend to the change in diffusion impedance.

[0142] In this embodiment of the invention, when the feature fusion module fuses the two weighted sets of features into a single vector to obtain the real-time state evaluation vector, it is specifically used for:

[0143] Using the weight values ​​determined for the first feature subset, the corresponding feature distribution areas are weighted and summed to obtain the first fusion value;

[0144] Using the weight values ​​determined for the second feature subset, the corresponding feature distribution areas are weighted and summed to obtain the second fusion value;

[0145] The real-time state evaluation vector is composed of the first fusion value and the second fusion value.

[0146] Specifically, the weight values ​​determined for each feature distribution in the first feature subset are retrieved, and each weight value is multiplied by the area of ​​the corresponding feature distribution in the current analysis period to obtain the weighted area of ​​each feature distribution. The weighted areas of all feature distributions are accumulated, and the sum is the first fusion value. This value comprehensively reflects the comprehensive characterization effect of each feature distribution in the first feature subset on the charge transfer process.

[0147] Furthermore, the weight values ​​determined for each feature distribution in the second feature subset are retrieved, and each weight value is multiplied by the area of ​​the corresponding feature distribution in the current analysis period to obtain the weighted area of ​​each feature distribution. The weighted areas of all feature distributions are accumulated, and the sum is the second fusion value. This value comprehensively reflects the comprehensive characterization effect of each feature distribution in the second feature subset on the material diffusion process.

[0148] Finally, the order of the first fusion value and the second fusion value is determined and kept fixed. The first fusion value is taken as the first element and the second fusion value is taken as the second element. The two values ​​are combined in this order to form an ordered data combination containing two elements. This ordered data combination is the real-time state evaluation vector, which can take into account the characteristic information of the charge transfer process and the mass diffusion process at the same time, and realize a comprehensive characterization of the current state of the backup battery.

[0149] In summary, this embodiment employs a targeted weighting strategy to precisely amplify the value of features. For the first feature subset, weights are assigned based on the contribution of each feature's distribution area change, guided by charge transfer rate, thus giving higher weights to features more sensitive to changes in the charge transfer process. For the second feature subset, weights are assigned based on the monotonicity and steepness of the feature distribution trend, highlighting features more representative of changes in diffusion impedance. This differentiated weighting overcomes the limitations of indiscriminate fusion, highlighting key changes in the core process and improving the relevance of feature representation.

[0150] In summary, in this embodiment, the normalization process eliminates the differences in dimensions and scales of different features, ensuring that the two sets of features are comparable during fusion. This avoids the problem that a single feature may obscure other key information due to its numerical magnitude advantage, making the feature weight allocation for charge transfer and matter diffusion processes more scientific and providing a guarantee for the objectivity of the subsequent fusion vector.

[0151] In summary, this embodiment integrates the weighted features of two core processes into a single real-time state evaluation vector. This not only consolidates comprehensive information on key electrochemical processes within the battery but also simplifies the feature dimensions, addressing the pain point of disorganized, complex features that are difficult for subsequent modules to efficiently analyze. This vector simultaneously carries the dynamic changes in charge transfer and mass diffusion, enabling a concentrated and comprehensive characterization of the battery's real-time state.

[0152] In summary, this embodiment provides high-quality analytical basis for the control decision module. The real-time state evaluation vector generated by the module accurately reflects the dynamic balance relationship of the internal process of the battery, making the subsequent calculation of the charging current decay slope and the discharge cutoff voltage rise threshold more data-supported. This ensures a deep fit between the individualized charging and discharging trajectory and the actual state of the battery, laying a key foundation for the system to optimize charging and discharging efficiency, improve capacity output, and reduce battery loss.

[0153] In this embodiment of the invention, when the control decision module performs dynamic reverse calculation of the charging current attenuation slope, it is specifically used for:

[0154] Arrange multiple real-time state evaluation vectors in chronological order to obtain a real-time state evaluation vector sequence;

[0155] The first fusion sequence is obtained by extracting a subsequence of the first fusion value changing over time from the real-time state evaluation vector sequence;

[0156] The first fused value sequence is subjected to continuous difference processing to obtain the change in its instantaneous rate of decrease;

[0157] Identify periods in which the instantaneous rate of descent continuously increases as the acceleration phase of descent;

[0158] Based on the duration and rate of increase of the descent acceleration phase, the rate of change of the constant current charging current with time is determined as the individualized charging current decay slope.

[0159] In this invention, when the information extraction module performs the reverse calculation of the discharge cutoff voltage rise threshold, it is specifically used for:

[0160] The second fusion sequence is obtained by extracting a subsequence of the second fusion value changing over time from the real-time state evaluation vector sequence;

[0161] By differentiating the second fusion sequence, the variation curve of the second fusion sequence is obtained;

[0162] Locate the inflection point on the curve where the voltage transitions from a stable phase to an increasing phase, and determine the inflection point as the threshold for raising the discharge cutoff voltage.

[0163] Specifically, real-time state evaluation vectors are periodically collected at fixed time intervals, and all collected real-time state evaluation vectors are arranged in chronological order of collection time to form an ordered sequence of real-time state evaluation vectors.

[0164] Furthermore, from the arranged real-time state evaluation vector sequence, the first fusion value contained in each real-time state evaluation vector is extracted, and these first fusion values ​​are sequentially associated according to their corresponding acquisition time order to form a subsequence that only reflects the change law of the first fusion value over time. This subsequence is the first fusion sequence.

[0165] Furthermore, the first fusion sequence is subjected to continuous difference processing. The first fusion value corresponding to two adjacent time points in the first fusion sequence is selected. The first fusion value of the previous time point is subtracted from the first fusion value of the next time point to obtain the change of the first fusion value within the adjacent time interval.

[0166] The change here is the instantaneous descent rate at the corresponding time point. By recording all instantaneous descent rates in chronological order, we can obtain the change of the instantaneous descent rate over time.

[0167] Furthermore, by comparing the instantaneous descent rates at adjacent time points, the trend of the instantaneous descent rate is determined. When the instantaneous descent rate at the later time point is greater than that at the earlier time point, and this increasing trend continues, this sustained time interval is defined as the descent acceleration phase.

[0168] Furthermore, the start and end times of the descent acceleration phase are statistically analyzed, and the difference between the two is calculated to obtain the duration of the descent acceleration phase. At the same time, the difference between the instantaneous descent rate at the start and end of the descent acceleration phase is calculated to obtain the magnitude of the rate increase. Combining the duration of the descent acceleration phase and the magnitude of the rate increase, the rate of change of the constant current charging current with time is calculated. This rate of change is the individualized charging current decay slope.

[0169] Furthermore, from the real-time state evaluation vector sequence, the second fusion value contained in each real-time state evaluation vector is extracted, and these second fusion values ​​are sequentially associated according to their corresponding acquisition time order to form a subsequence that only reflects the change law of the second fusion value over time. This subsequence is the second fusion sequence.

[0170] Furthermore, the second fusion sequence is differentiated, and the second fusion values ​​corresponding to two adjacent time points in the second fusion sequence are selected. The difference between the second fusion value at the next time point and the second fusion value at the previous time point is calculated. Then, the difference is divided by the time interval between the two time points to obtain the rate of change of the second fusion value at the corresponding time point. The rates of change of all time points are arranged in chronological order to form a curve reflecting the change of the rate of change of the second fusion value with time.

[0171] Furthermore, observe the change curve of the second fusion sequence and identify the change stage of the curve. When the curve is in a stable stage, its corresponding change rate remains relatively constant and fluctuates very little. When the curve starts to enter the rising stage from the stable stage, the change rate will show a continuous increasing trend. The time point when the curve changes from the stable stage to the rising stage is determined as the inflection point. This inflection point is the threshold for the rise of the discharge cutoff voltage.

[0172] Finally, the determined individualized charging current decay slope and the discharge cutoff voltage rise threshold are integrated, and the charging and discharging parameters of the backup battery are adjusted based on the two. This makes the charging current decrease with time according to the decay slope, and the discharge process adjusts the cutoff voltage at the moment corresponding to the rise threshold, thus forming an individualized charging and discharging trajectory adapted to the backup battery.

[0173] In summary, this invention utilizes periodic data acquisition and continuous analysis to achieve real-time tracking of state responses. The module periodically acquires real-time state evaluation vectors and dynamically captures the balance and evolution trends of the core electrochemical processes within the battery by tracking the relative magnitudes and directions of change of charge transfer-related weights and diffusion impedance-related weights in multiple vectors. This avoids the lag in responding to real-time battery state changes inherent in traditional static adjustment strategies, ensuring that adjustments to charge and discharge parameters are synchronized with changes in the battery's internal state, significantly improving the timeliness of strategy adaptation.

[0174] In summary, this embodiment accurately reverse-engineers key charging and discharging parameters to achieve targeted optimization. For the charging scenario, by analyzing the changes in the weights related to the first fusion value, the charging current decay slope is dynamically determined, ensuring a precise match between the charging current decay rate and the efficiency changes in the charge transfer process. This avoids both underutilization of capacity due to excessively rapid current decay and overcharging risks caused by excessively slow decay. For the discharging scenario, by analyzing the fluctuations in the weights related to the second fusion value, the discharge cutoff voltage rise threshold is located. The cutoff voltage is adjusted in a timely manner at the inflection point where the material diffusion process becomes limited, preventing damage to the battery's internal structure from over-discharge and achieving refined control over the entire charging and discharging process.

[0175] In summary, this embodiment generates individualized charge and discharge trajectories to adapt to individual battery differences. The module formulates strategies based on the real-time state assessment vector analysis results specific to each battery, fully taking into account the differences in electrochemical characteristics formed by different batteries in the manufacturing process and usage history. This allows each battery to obtain a charge and discharge scheme adapted to its own state, significantly improving the consistency between the charge and discharge process and the actual state of the battery.

[0176] In summary, this embodiment achieves the core objective of the closed-loop protection system. This module transforms the abstract state vector output by the feature fusion module into directly executable charging and discharging parameters, constructing a complete closed loop of "state perception - feature fusion - decision execution." Its individualized trajectory optimizes charging and discharging efficiency and capacity output, and effectively reduces internal polarization losses and structural damage to the battery by avoiding mismatched charging and discharging behaviors, providing crucial support for extending battery cycle life and ensuring long-term energy storage stability.

[0177] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0178] The embodiments described above can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A backup battery charging and discharging system, characterized in that, The system includes a parameter acquisition module, a dynamic response signal generation module, a signal analysis module, a feature fusion module, and a control decision module, wherein: The parameter acquisition module is used to acquire the reference electrochemical parameter set of the target backup battery; The dynamic response signal generation module is used to apply a composite micro-perturbation signal to the backup battery during the constant current charging and discharging phase, and simultaneously acquire the loop current to obtain the dynamic response signal. The signal analysis module is used to perform relaxation time distribution analysis on dynamic response signals, separating the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process. The feature fusion module is used to normalize and weight the first feature subset based on charge transfer rate, and at the same time normalize and weight the second feature subset based on diffusion impedance. The two weighted feature sets are then fused into a single vector to obtain the real-time state evaluation vector. The control decision module is used to periodically acquire real-time state evaluation vectors. By continuously analyzing the relative magnitude and direction of change of two sets of weights in multiple vectors, it dynamically inversely calculates the attenuation slope of the charging current or the rise threshold of the discharge cutoff voltage, forming an individualized charging and discharging trajectory.

2. The backup battery charging and discharging system as described in claim 1, characterized in that, When the parameter acquisition module acquires the reference electrochemical parameter set of the target backup battery, it is specifically used for: After the backup battery has been left to stand and reach electrochemical equilibrium, its open-circuit voltage and temperature are measured as a steady-state reference. Apply DC current excitation to the backup battery and record the time-domain curve of its voltage recovery process after the excitation is removed; Based on the decay characteristics of the voltage recovery time-domain curve, the fast relaxation time constant and the slow relaxation time constant are extracted as dynamic reference parameters. The reference electrochemical parameter set is composed of steady-state reference parameters and dynamic reference parameters.

3. A backup battery charging and discharging system as described in claim 2, characterized in that, The dynamic response signal generation module applies a composite micro-perturbation signal to the backup battery during the constant current charge / discharge phase, specifically for: Based on the fast relaxation time constant in the reference electrochemical parameter set, the short-period fluctuation component is determined; Based on the lumped slow relaxation time constant of the reference electrochemical parameters, the long-period fluctuation component is determined. By superimposing the short-period fluctuation component with the long-period fluctuation component, a composite micro-perturbation signal is obtained.

4. A backup battery charging and discharging system as described in claim 3, characterized in that, When performing relaxation time distribution analysis on the dynamic response signal, the signal analysis module is specifically used for: Based on the correspondence between the dynamic response signal and the composite micro-perturbation signal, the frequency domain impedance response of the backup battery in the current state is established; The frequency domain impedance response is converted into a continuous distribution spectrum with the relaxation time constant as the abscissa and the characteristic intensity as the ordinate, thus obtaining the relaxation time distribution spectrum.

5. A backup battery charging and discharging system as described in claim 4, characterized in that, When the signal analysis module separates the first feature subset characterizing the charge transfer process and the second feature subset characterizing the matter diffusion process, it is specifically used for: On the relaxation time distribution spectrum, determine the relative distances between the time constants of each characteristic distribution and the fast relaxation time constant and the slow relaxation time constant, respectively; Among all feature distributions, those that are relatively closer to the fast relaxation time constant are assigned to the first feature subset. Among all the feature distributions, those that are relatively closer to the slow relaxation time constant are assigned to the second feature subset.

6. A backup battery charging and discharging system as described in claim 1, characterized in that, When the feature fusion module performs normalization and weighting of the first feature subset based on charge transfer rate, it is specifically used for: Obtain the change in the area of ​​each feature distribution in the first feature subset over adjacent analysis periods; The weight value is determined based on the proportion of the change in each characteristic distribution to the total change in the subset.

7. A backup battery charging and discharging system as described in claim 6, characterized in that, When the feature fusion module performs normalization weighting of the second feature subset based primarily on diffusion impedance, it is specifically used for: The trend of the peak intensity of each feature distribution in the second feature subset changes with the battery's discharged or charged capacity, thus obtaining the feature distribution trend; The weight values ​​are determined based on the monotonicity and steepness of the characteristic distribution trend.

8. A backup battery charging and discharging system as described in claim 7, characterized in that, When the feature fusion module fuses the two weighted sets of features into a single vector to obtain the real-time state evaluation vector, it is specifically used for: Using the weight values ​​determined for the first feature subset, the corresponding feature distribution areas are weighted and summed to obtain the first fusion value; Using the weight values ​​determined for the second feature subset, the corresponding feature distribution areas are weighted and summed to obtain the second fusion value; The real-time state evaluation vector is composed of the first fusion value and the second fusion value.

9. A backup battery charging and discharging system as described in claim 8, characterized in that, When the control decision module performs dynamic reverse calculation of the charging current attenuation slope, it is specifically used for: Arrange multiple real-time state evaluation vectors in chronological order to obtain a real-time state evaluation vector sequence; The first fusion sequence is obtained by extracting a subsequence of the first fusion value changing over time from the real-time state evaluation vector sequence; The first fused value sequence is subjected to continuous difference processing to obtain the change in its instantaneous rate of decrease; Identify periods in which the instantaneous rate of descent continuously increases as the acceleration phase of descent; Based on the duration and rate of increase of the descent acceleration phase, the rate of change of the constant current charging current with time is determined as the individualized charging current decay slope.

10. A backup battery charging and discharging system as described in claim 9, characterized in that, When performing the reverse discharge cutoff voltage rise threshold calculation, the information extraction module is specifically used for: The second fusion sequence is obtained by extracting a subsequence of the second fusion value changing over time from the real-time state evaluation vector sequence; By differentiating the second fusion sequence, the variation curve of the second fusion sequence is obtained; Locate the inflection point on the curve where the voltage transitions from a stable phase to an increasing phase, and determine the inflection point as the threshold for raising the discharge cutoff voltage.

Images