A battery cell adaptive equalization method in a new energy vehicle charging process

Abstract

The application discloses a battery monomer adaptive equalization method in a new energy vehicle charging process, and aims to solve the problems of existing equalization technology lag, rigidity and poor synergy with the charging process. The method comprises the following steps: collecting the total voltage, total current, and each monomer voltage and temperature of the battery pack in real time; calculating the monomer voltage dispersion index and comparing it with the threshold value to determine the starting equalization opportunity; dynamically calculating the adaptive equalization threshold based on the average state of charge, average temperature and aging degree of the battery pack; classifying each monomer according to the threshold value, and constructing an equalization target identification matrix; finally, combining the current charging stage, executing hierarchical equalization decision, and controlling the cooperative work of the charging equipment and the active equalization circuit. The application realizes the deep synergy of equalization and charging, significantly improves the charging speed, energy utilization rate and battery pack charging saturation through the dynamic threshold and hierarchical control strategy, and effectively prolongs the overall life of the battery pack.

Description

Technical Field

[0001] This invention relates to the field of battery management technology for new energy vehicles, and in particular to an adaptive equalization method for individual battery cells during the charging process of new energy vehicles. Background Technology

[0002] The power battery packs of new energy vehicles typically consist of hundreds to thousands of battery cells connected in series and parallel. Due to differences in manufacturing processes, usage environments, and aging levels, inconsistencies in capacity, internal resistance, and voltage are inevitable among battery cells. This inconsistency is particularly pronounced during charging: when the battery pack is charged with a constant current, some battery cells with smaller capacity or higher internal resistance will reach the charging cutoff voltage first, while other battery cells are not yet fully charged. To prevent overcharging damage, the battery management system must terminate the entire battery pack charging process, resulting in a decrease in the actual usable capacity of the battery pack, a shortened driving range, and energy waste.

[0003] To address these issues, battery balancing technology has been widely adopted. Existing technologies mainly include passive balancing and active balancing. Passive balancing involves discharging excess energy as heat by connecting a resistor in parallel across the higher-voltage battery cells. This method is simple in structure but has low energy efficiency, generates heat, and significantly prolongs charging time by waiting for the higher-voltage cells to discharge. It may even fail to effectively suppress voltage rise due to limited balancing current, leading to premature termination of charging. Active balancing uses components such as capacitors, inductors, or transformers to transfer energy from higher-voltage cells to lower-voltage cells or the entire battery pack, resulting in higher energy efficiency. However, most existing active balancing strategies operate after charging has completely stopped or in a static state, failing to deeply integrate with the charging process. Their balancing trigger thresholds are typically fixed, unable to adapt to the dynamic changes in battery inconsistencies under different states of charge, temperatures, and aging stages. In addition, existing balancing strategies often perform indiscriminate balancing on all cells that exceed the threshold, lacking dynamic coordination of balancing targets, balancing intensity and charging process. This can easily lead to over-balancing, increasing energy conversion losses and component stress, and may even interfere with the stability of the charging process due to frequent balancing actions.

[0004] Therefore, the existing technology has the following problems that urgently need to be solved: First, the equalization process is separated from the charging process, resulting in charging interruption or reduced efficiency; Second, the equalization triggering conditions are rigid and cannot adapt to the dynamic changes in battery state; Third, the equalization strategy is crude and lacks fine-grained coordinated control over equalization targets, timing and intensity, resulting in the overall energy efficiency and battery life failing to reach the optimal level. Summary of the Invention

[0005] To achieve the above objectives, this invention provides an adaptive balancing method for individual battery cells during the charging process of new energy vehicles. This method is cyclically executed by the battery management system during the charging process of the battery pack, and includes the following steps:

[0006] Step 1: Real-time acquisition of the total battery pack voltage, total battery pack current, terminal voltage of each individual battery cell, and temperature of each individual battery cell during the charging process;

[0007] Step 2: Calculate the dispersion index of the battery cell voltage based on the terminal voltage of all battery cells, compare the dispersion index with the preset dispersion threshold, and proceed to Step 3 when the dispersion index is greater than or equal to the dispersion threshold.

[0008] Step 3: Dynamically calculate the adaptive balancing threshold based on the average state of charge and average temperature of the battery pack. The average state of charge of the battery pack is estimated by the total voltage and total current of the battery pack, and the average temperature of the battery pack is calculated based on the temperature of all individual battery cells.

[0009] Step 4: Based on the adaptive equalization threshold, compare and classify the terminal voltage of each battery cell with the average terminal voltage of all battery cells, construct an equalization target identification matrix containing the state identifier of each battery cell, and count the number of overvoltage candidate cells.

[0010] Step 5: Based on the equalization target identification matrix, the number of overvoltage candidate cells, and the current charging stage, perform hierarchical equalization decision-making and generate control commands for the charging equipment and active equalization circuit to achieve adaptive equalization of battery cells.

[0011] Preferably, the calculation process of the dispersion index in step two includes:

[0012] Calculate the average terminal voltage of all individual battery cells;

[0013] Calculate the absolute difference between the terminal voltage of each individual battery cell and this average value;

[0014] Calculate the average of all absolute differences and use this average as the dispersion index.

[0015] Preferably, the process of dynamically calculating the adaptive equilibrium threshold in step three specifically includes:

[0016] The average state of charge of the battery pack is estimated by using the ampere-hour integration method, combined with the total current of the battery pack and the nominal capacity of the battery pack.

[0017] Calculate the average temperature of all individual battery cells to obtain the average temperature of the battery pack;

[0018] Based on the average state of charge and average temperature of the battery pack, a baseline threshold is obtained by querying a pre-established threshold mapping table. The threshold mapping table is obtained by testing the constant current charging characteristics of individual battery cells at different temperatures and states of charge and interpolating the values.

[0019] The baseline threshold is multiplied by an aging compensation coefficient to obtain the final adaptive equalization threshold. The aging compensation coefficient is determined based on the number of charge cycles of the battery pack and gradually increases from 1 as the number of charge cycles increases.

[0020] Preferably, the process of constructing the balanced target recognition matrix in step four specifically includes:

[0021] Calculate the voltage difference between the terminal voltage of each individual battery cell in the current cycle and the average terminal voltage of all individual battery cells;

[0022] When the voltage difference of a certain battery cell is greater than zero and greater than the adaptive equalization threshold, the status of that battery cell is set as "overvoltage candidate".

[0023] When the voltage difference of a certain battery cell is less than zero and the absolute value of the voltage difference is greater than the adaptive equalization threshold, the status of that battery cell is set to "undervoltage candidate".

[0024] When the absolute value of the voltage difference of a certain battery cell is less than or equal to the adaptive equalization threshold, the status flag of that battery cell is set to "hold".

[0025] The number of battery cells with the status marked as "overvoltage candidate" is recorded as the overvoltage candidate cell number.

[0026] Preferably, step five, which involves performing a hierarchical equilibrium decision, specifically includes the following sub-steps:

[0027] The current charging stage is determined to be either constant current charging stage or constant voltage charging stage, and the determination is based on whether the average state of charge of the battery pack is lower than the preset state of charge switching point.

[0028] If the current phase is constant current charging and the number of overvoltage candidate cells is zero, but there are battery cells with the status label "undervoltage candidate" in the equalization target identification matrix, then the active equalization circuit will transfer energy from the battery pack bus to the battery cells with the status label "undervoltage candidate" while maintaining the charging equipment at the maximum allowable current.

[0029] If the current stage is constant current charging and the number of overvoltage candidate cells is greater than zero, the charging equipment is controlled to reduce the output current to the preset safe current value, and the active balancing circuit is controlled to transfer energy from the battery cells with the largest voltage difference marked as "overvoltage candidate" to the battery cells or battery pack bus marked as "undervoltage candidate".

[0030] If the current state is a constant voltage charging stage, the charging equipment is controlled to maintain a constant output voltage, and the active balancing circuit is controlled to transfer energy from all battery cells marked as "overvoltage candidate" to the battery pack bus with a preset balancing current.

[0031] Preferably, during the constant current charging phase and the equalization control process where the number of overvoltage candidate cells is greater than zero, a real-time monitoring step is also included:

[0032] Real-time monitoring of the terminal voltage of all battery cells marked as "overvoltage candidate";

[0033] When the voltage difference of a battery cell whose status is marked as "overvoltage candidate" drops to less than or equal to the adaptive balancing threshold, the status of the battery cell is updated to "hold", and the balancing energy transfer operation for that battery cell is stopped.

[0034] Preferably, the generation of control commands for the charging device in step five specifically includes generating a current command for adjusting the output current of the charging device or a voltage command for adjusting the output voltage of the charging device. The current command and voltage command are sent through the communication link between the battery management system and the charging device.

[0035] Preferably, the active balancing circuit is an inductor-based bidirectional flyback balancing circuit or a switched capacitor-based balancing circuit. The active balancing circuit is controlled by the battery management system and is capable of bidirectional energy transfer between any two battery cells or between a battery cell and the battery pack bus.

[0036] Preferably, the state of charge switching point is determined based on the typical state of charge value corresponding to the end of the constant current charging phase of the battery pack when the voltage begins to rise rapidly. This typical state of charge value is obtained by performing standard charging tests on similar battery packs.

[0037] Preferably, the method continues to collect data from step one during the equalization control in step five, and terminates the current round of equalization control and returns to step one for the next cycle when any of the following conditions are met: the status identifier of all battery cells in the equalization target identification matrix is ​​"hold"; the charging process is terminated by an external command; the battery management system detects that the total voltage of the battery pack, the total current of the battery pack, or the temperature of any battery cell exceeds a safety threshold.

[0038] The beneficial effects of this invention are:

[0039] 1. This invention deeply embeds and coordinates the equalization process throughout the entire charging process, intelligently switching equalization strategies according to different charging stages, effectively avoiding the problem of pausing or prematurely terminating charging while waiting for equalization. This method allows the charging system to maintain high power input for a longer period, thereby shortening the overall charging time. Simultaneously, the active energy transfer mechanism employed, compared to traditional passive equalization which relies primarily on heat dissipation, significantly reduces energy waste and improves the energy utilization efficiency of the entire charging process.

[0040] 2. This invention overcomes the limitations of traditional fixed-threshold equalization methods by introducing an adaptive equalization threshold dynamically calculated based on real-time battery status and aging levels. This allows the trigger sensitivity of the equalization system to automatically optimize according to the actual battery operating conditions. Combined with the constructed "equalization target identification matrix" and "hierarchical equalization decision" logic, it achieves refined selection and control of the equalization target, timing, direction, and intensity. This avoids unnecessary circuit actions and losses caused by coarse equalization, making the equalization action more precise, necessary, and efficient.

[0041] 3. This invention, through real-time and precise equalization intervention, continuously suppresses the widening trend of voltage and state of charge inconsistencies among battery cells during charging, ensuring a high degree of coordination in the electrochemical states of each cell. This helps mitigate the accelerated aging of some cells caused by prolonged overcharging or undercharging tendencies, promoting the synchronization of aging rates among all cells within the battery pack. In the long term, this method can significantly improve the overall cycle life and reliability of the battery pack, ensuring its usable capacity throughout its entire lifespan. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of the steps of the method of the present invention;

[0044] Figure 2 This is a flowchart of the step three of the method of the present invention, which describes the dynamic calculation of the adaptive equilibrium threshold.

[0045] Figure 3 This is a flowchart illustrating the steps of constructing the balanced target recognition matrix in step four of the method of the present invention. Detailed Implementation

[0046] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0047] Please see Figures 1-3 This invention provides an adaptive equalization method for battery cells during the charging process of new energy vehicles. This method is periodically executed by the central processing unit of the battery management system throughout the entire charging process of the battery pack. First, the analog front-end acquisition circuit of the battery management system synchronously acquires the total voltage between the positive and negative terminals of the battery pack, the total current in the series circuit, the voltage across each battery cell, and the surface temperature of each battery cell in each control cycle. These data are converted into digital signals and transmitted to the central processing unit.

[0048] Subsequently, the central processing unit invokes a calculation program to calculate a statistic reflecting the degree of dispersion, namely the dispersion index, based on the acquired voltage data of all individual battery cells. This dispersion index is compared with a preset benchmark threshold. If the dispersion index is less than the threshold, it indicates that the battery pack consistency is good, and the next data acquisition cycle begins directly. If the dispersion index reaches or exceeds the threshold, it is determined that equalization control needs to be activated.

[0049] Next, the central processing unit dynamically calculates an adaptive equalization voltage difference threshold based on the estimated average state of charge of the battery pack, the calculated average temperature of the battery pack, and the battery aging status.

[0050] Then, using this adaptive threshold, the real-time voltage of each battery cell is compared and classified with the average voltage of all cells. Each cell is labeled with a status identifier of "overvoltage candidate", "undervoltage candidate" or "holding", and the total number of "overvoltage candidate" cells is counted to form the balance target identification matrix for this control cycle.

[0051] Finally, based on the identification matrix, the total number of overvoltage candidate cells, and the current charging stage (constant current or constant voltage), the central processing unit executes hierarchical decision-making logic, generates specific control commands, synchronously adjusts the output power of the charging equipment, and controls the active balancing circuit to perform directional energy transfer between battery cells or between a cell and the battery pack bus, until the balancing condition is released.

[0052] In one possible implementation, after receiving voltage samples from all battery cells within a control cycle, the central processing unit of the battery management system first sums these voltage values ​​and divides them by the total number of cells to obtain the average voltage of the battery cells. Then, the program sequentially calculates the difference between the voltage sample value of each battery cell and the aforementioned average value, and takes the absolute value of this difference to obtain a set of absolute deviation values ​​reflecting the deviation of each cell from the average level. Finally, this set of absolute deviation values ​​is summed again and divided by the total number of cells; the resulting value is defined as the dispersion index.

[0053] The entire calculation process is implemented by embedded software, requiring no additional dedicated hardware. By calculating this dispersion index, the system can use an objective and quantitative value to characterize the degree of inconsistency in the voltage of all individual cells within the battery pack, avoiding the bias of judging the overall state based solely on the voltage of individual cells. This index acts as a master switch for triggering subsequent complex balancing logic, ensuring that the balancing system is activated only when necessary, reducing unnecessary control actions and energy loss, and improving the overall energy efficiency and stability of the system.

[0054] In one possible implementation, the process of dynamically calculating the adaptive equalization threshold first estimates the average state of charge of the battery pack using the ampere-hour integration method. This involves integrating the collected total current of the battery pack over time and combining this with the nominal capacity of the batteries. The average temperature of the battery pack is then obtained by taking the arithmetic mean of the temperature samples from all individual cells.

[0055] The system memory pre-stores a two-dimensional threshold mapping table calibrated experimentally. This table takes the average state of charge and average temperature of the battery pack as input and outputs a corresponding reference voltage difference threshold. The mapping table was established by conducting constant current charging tests on the same type of battery at different states of charge and temperatures in the laboratory, recording the typical difference between the individual cell voltage and the average voltage at the inflection point where the voltage begins to rise nonlinearly, and then filling the entire working range using an interpolation algorithm.

[0056] Furthermore, the system reads the number of complete charging cycles the battery pack has completed, stored in non-volatile memory, and determines an aging compensation coefficient greater than or equal to 1 based on the curve obtained through pre-existing aging experiments. The final adaptive balancing threshold is obtained by multiplying the aging compensation coefficient by the queried baseline threshold. This implementation method allows the balancing judgment threshold to be dynamically adjusted according to the battery's real-time operating state (state of charge, temperature) and long-term health state (degree of aging). It avoids oversensitivity when the battery is in good condition and improves the timeliness of intervention when the battery is in a harsh condition or aging, significantly enhancing the adaptability and intelligence of the balancing control.

[0057] In one possible implementation, the central processing unit first calculates the difference between the voltage of each individual battery cell in the current cycle and the calculated average voltage of all cells. Then, it compares and logically determines the voltage difference of each cell with the adaptive balancing threshold calculated in step three. If the voltage difference of a battery cell is greater than 0 and its value is greater than the adaptive balancing threshold, the cell is marked as an "overvoltage candidate." If the voltage difference of a battery cell is less than 0 and its absolute value is greater than the adaptive balancing threshold, the cell is marked as an "undervoltage candidate."

[0058] Battery cells that do not meet either of the above two conditions—that is, cells whose absolute voltage difference is less than or equal to the adaptive balancing threshold—are marked as "maintained." After the program iterates through all battery cells and completes the above marking, it stores the status identifier of each cell in an array or matrix structure in memory. This structure is the balancing target identification matrix. Simultaneously, the program uses a counting function to count the number of all identifiers marked as "overvoltage candidate" in this matrix, obtaining the number of overvoltage candidate cells. This implementation process accurately classifies and locates inconsistent cells within the battery pack, transforming the global inconsistency problem into operational targets for specific "overvoltage candidate" and "undervoltage candidate" cells. This provides a clear basis for subsequent differentiated and precise balancing control, avoiding blind allocation of balancing energy.

[0059] In one possible implementation, the decision-making layer first determines the current charging stage: the central processing unit compares the estimated average state of charge (SOC) of the battery pack with a preset SOC switching point. If the SOC is lower than this point, it is determined to be in the constant current charging stage; otherwise, it is determined to be in the constant voltage charging stage. Specifically, the decision is executed as follows: if it is in the constant current charging stage and the number of overvoltage candidate cells is 0, but there are undervoltage candidate cells in the identification matrix, the control command causes the charging equipment to maintain the maximum allowable current, and simultaneously activates the active balancing circuit to replenish energy to these undervoltage candidate cells from the battery pack bus.

[0060] If the charging phase is constant current and the number of overvoltage candidate cells is greater than 0, the control command first requests the charging equipment to reduce the output current to a lower safe current value. Then, it immediately controls the active balancing circuit to transfer energy from the top few overvoltage candidate cells to undervoltage candidate cells or feed it back to the battery pack bus. If the charging phase is constant voltage, the control command maintains the output voltage of the charging equipment constant and uses a smaller constant current to control the active balancing circuit to transfer energy from the overvoltage candidate cells to the battery pack bus. This layered implementation achieves tight coupling between the balancing strategy and the charging process, adopting the optimal strategy at different stages. It can preventively bridge differences in the early stage of charging, quickly eliminate charging bottlenecks in the middle stage, and finely complete balancing at the end stage, thereby maximizing charging speed and charging saturation while ensuring safety.

[0061] In one possible implementation, during the constant current charging phase and while the system is performing energy transfer operations on "overvoltage candidate" cells, the monitoring program continuously reads the real-time voltage values ​​of these operated cells in parallel. The program periodically recalculates the difference between the real-time voltage of these cells and the continuously updated average voltage of all cells. Once a cell being equalized is detected whose voltage difference drops to less than or equal to the current adaptive equalization threshold, the monitoring logic immediately triggers a state update. The cell's state flag in the equalization target identification matrix is ​​changed from "overvoltage candidate" to "hold".

[0062] As the status flag is updated, the control logic immediately sends a command to the active balancing circuit to stop transferring energy from that specific cell. Subsequently, the control system selects the cell with the largest current voltage difference from the remaining "overvoltage candidate" cells as the new balancing source and continues energy transfer. This process achieves dynamic management of the balancing target, ensuring that the balancing operation is just right. Once the inconsistency of a cell is corrected to an acceptable range, the operation on that cell is immediately stopped, avoiding the energy and time waste caused by over-balancing and reducing the thermal stress on the balancing circuit components.

[0063] In one possible implementation, the central processing unit (CPU) of the battery management system generates corresponding digital control instructions based on the results of the hierarchical equalization decision. When it is necessary to adjust the charging current, a digital current instruction containing the target current value is generated; when it is necessary to adjust or maintain the charging voltage, a digital voltage instruction containing the target voltage value is generated. These instructions are encapsulated into data frames according to a predefined communication protocol by a dedicated communication controller integrated within the battery management system.

[0064] The encapsulated data frame is sent to the external charging device controller via an isolated communication interface, such as a controller area network bus or pulse width modulation signal. The charging device controller parses the received instructions and adjusts its power output module, thereby changing the actual output charging current or charging voltage. This implementation clarifies the collaborative working mechanism between the battery management system and the external charging device. Through a standardized communication link, it achieves precise intervention in the charging process, ensuring that actions such as "adjusting the charging current" or "maintaining the charging voltage" can be reliably and accurately executed. This is a key technical aspect enabling coordinated charging and equalization.

[0065] In one possible implementation, the active balancing circuit can be a bidirectional flyback topology based on inductor energy storage. This circuit includes a high-speed switching network controlled by a battery management system, a multi-winding transformer or a series of independent inductors, and necessary rectifier and filter components. The switching network periodically switches on and off according to control commands, connecting "overvoltage candidates" that need to release energy to the primary winding of the transformer, thus enabling inductor energy storage.

[0066] Subsequently, by switching the switching network, the stored energy is transferred to the secondary winding connected to the "undervoltage candidate" cell or the battery pack bus, completing the energy transfer. Another feasible implementation is a switching capacitor-based equalization circuit, which achieves voltage equalization by controlling the connection order of capacitors between different cells and utilizing capacitor charging and discharging. Regardless of the topology, this circuit has the ability to perform bidirectional energy transfer between any two specified battery cells, or between any battery cell and the battery pack common bus.

[0067] This implementation provides a physical basis for the efficient redistribution of energy within the battery pack, and its controllability and flexibility are the fundamental guarantee for achieving the technical feature of directional and quantitative energy transfer between different cells.

[0068] In one possible implementation, the switching point is a key parameter used to determine the transition of the battery pack from the constant current charging stage to the constant voltage charging stage. Its determination is based on standard characteristic tests of the battery pack model. Under laboratory conditions, a complete constant current-constant voltage charging test is performed on the battery pack model at the rated current, and the curves showing the change of the battery pack terminal voltage over time or with the state of charge are recorded. Analyzing these curves, the inflection point corresponding to the transition of the battery pack terminal voltage from an approximately linear increase to reaching and stabilizing at the charging cutoff voltage plateau is identified.

[0069] Read the average state of charge (SOC) value of the battery pack at the inflection point; this value is the theoretical transition point. Considering the tolerance and safety margin in practical applications, this theoretical value can be appropriately modified to ultimately determine the preset SOC switching point used as the basis for the judgment logic in the program.

[0070] This implementation method ensures the accuracy of stage judgment, enabling hierarchical balancing decisions to be executed based on correct charging stage information. This is a prerequisite for ensuring the correct application of the strategy differences between the "constant current stage" and the "constant voltage stage".

[0071] In one possible implementation, while the equalization control command is executed in step five, the real-time data acquisition process in step one continues to run at a fixed cycle. The system checks preset termination conditions in each control cycle. The first condition is to check the equalization target identification matrix. If the status flags of all battery cells in the matrix have changed to "hold," it indicates that the current inconsistency problem has been resolved, and this round of equalization control is complete.

[0072] The second condition is to monitor the charging process status. If the charging device has stopped outputting or a stop charging command has been received from the outside, the equalization process will be terminated immediately.

[0073] The third condition is safety monitoring. If any data collected in real time, including the total voltage of the battery pack, the total current, or the temperature of any single cell, exceeds its respective preset safety alarm threshold, the system will immediately terminate all operations to enter a safety protection state.

[0074] When any of the above conditions are met, the current centralized balancing control process is terminated, and the system returns to its initial monitoring state, beginning a completely new data acquisition and judgment cycle. This implementation method ensures the integrity and safety of the control process, enabling the system to automatically exit the balancing state and continue monitoring, thus guaranteeing the long-term reliable operation of the power battery.

[0075] Example

[0076] This embodiment uses a new energy vehicle equipped with a ternary lithium-ion battery pack charging at a DC fast charging station as an application scenario. The battery pack has a nominal voltage of 400V and a nominal capacity of 80Ah, consisting of 100 battery cells connected in series. The implementation of this invention is accomplished by the battery management system within the battery pack, which possesses the capabilities for voltage acquisition, current acquisition, temperature acquisition, communication, and control of the active balancing circuit.

[0077] This embodiment describes in detail the complete process and details of the battery management system executing the adaptive balancing method during DC fast charging of the new energy vehicle:

[0078] Step 1: Collect global data of the battery pack and individual battery data in real time during the charging process.

[0079] The battery management system's data acquisition module synchronously collects the following parameters in 100-millisecond control cycles:

[0080] Total battery pack voltage: obtained through a high-voltage sampling circuit connected to the positive and negative terminals of the battery pack.

[0081] Total battery pack current: obtained through a Hall current sensor connected in series in the battery pack circuit; charging current is defined as positive.

[0082] Terminal voltage of each battery cell: The voltage difference between the two ends of each cell is collected sequentially through a multiplexer and an analog-to-digital converter.

[0083] The temperature of each battery cell is obtained by a negative temperature coefficient thermistor sensor attached to the surface of each battery cell.

[0084] In a specific control cycle, the collected data is as follows: the total battery pack voltage is 365.2V, and the total battery pack current (i.e., charging current) is 120A (0.5C rate). The individual battery cell voltages range from 3.640V to 3.665V, and the individual battery cell temperatures range from 25℃ to 28℃.

[0085] Step 2: Calculate the dispersion index of the battery cell voltage and determine whether to activate the equalization control for this cycle.

[0086] The central processing unit (CPU) of the battery management system calculates a voltage dispersion index based on the terminal voltages of 100 individual battery cells collected in the current cycle. First, it calculates the arithmetic mean of these 100 voltage values, assuming it to be 3.652V. Next, it calculates the absolute difference between each cell's voltage and this average value; for example, for a cell with a voltage of 3.665V, the absolute difference is 0.013V. Finally, it calculates the average of these 100 absolute differences to obtain the dispersion index, assuming it to be 0.008V.

[0087] The preset dispersion threshold is 0.010V, which was determined based on extensive experimental statistics and represents the basic level of acceptable inconsistency in the battery pack during the charging process. The calculated dispersion index of 0.008V is compared with the dispersion threshold of 0.010V. Since 0.008V is less than 0.010V, the battery management system determines that the current battery pack inconsistency is within an acceptable range, and without initiating equalization, it directly returns to step one to begin data acquisition for the next 100-millisecond cycle.

[0088] As charging progresses, the inconsistency between individual battery cells may worsen. In a later cycle, the calculated dispersion index becomes 0.012V, which is greater than the dispersion threshold of 0.010V. At this point, the battery management system determines that equalization intervention needs to be initiated, and the process proceeds to step three.

[0089] Step 3: Dynamically calculate the adaptive balancing threshold under the current charging state.

[0090] This step is one of the core elements of the invention's creativity, aiming to adapt the equalization triggering conditions to the dynamic state of the battery.

[0091] Estimating the average state of charge (SOC) of the battery pack: The battery management system estimates the SOC using the ampere-hour integration method based on the total battery pack current continuously collected in step one. The initial SOC is obtained from the stored value read at power-on or by looking up the open-circuit voltage in a table. In this cycle, the estimated average SOC of the battery pack is 65%.

[0092] Calculate the average temperature of the battery pack: Calculate the arithmetic mean of the temperatures of all 100 individual battery cells collected in step one, and the average temperature of the battery pack is 26.5℃.

[0093] Threshold Mapping Table: A two-dimensional threshold mapping table is pre-stored in the battery management system's memory. The table is built as follows: In a laboratory environment, constant current charging tests are performed on the same type of battery cells at three temperature points (10℃, 25℃, and 40℃) and three states of charge (CBC) points (20%, 50%, and 80%). The typical difference between the cell voltage and the average voltage is recorded for each (temperature, CBC) combination when a cell voltage begins to deviate significantly from the average voltage, indicating that the cutoff voltage is about to be reached. For example, at 25℃ and 50% CBC, the typical difference is 0.015V; at 40℃ and 80% CBC, the typical difference is 0.010V. A continuous threshold lookup table covering temperatures from 0-50℃ and CBCs from 0-100% is constructed using a binary linear interpolation algorithm. Based on the current average temperature of 26.5℃ and average CBC of 65%, this mapping table is consulted, and a baseline threshold, assumed to be 0.013V, is calculated through interpolation.

[0094] Aging compensation coefficient is applied: The battery management system records the historical number of full charge cycles of the battery pack, assuming the current number is 300. Based on accelerated aging test data for this battery model, a relationship curve between the aging compensation coefficient and the number of cycles has been established. At 300 cycles, the corresponding aging compensation coefficient is 1.15. This coefficient is greater than 1, meaning that inconsistency will worsen as the battery ages, requiring a relaxation of the equalization judgment threshold to avoid overly frequent equalization actions.

[0095] Calculate the final adaptive equalization threshold: Multiply the baseline threshold of 0.013V by the aging compensation coefficient of 1.15 to obtain the adaptive equalization threshold of 0.015V used in this cycle.

[0096] Step 4: Construct the balanced target identification matrix for the current period.

[0097] The battery management system classifies each cell based on the adaptive equalization threshold of 0.015V obtained in step three.

[0098] Calculate the difference between the terminal voltage of each battery cell in the current cycle and the average voltage of all cells (3.652V) calculated in step two.

[0099] Iterate through each cell: Assume cell number 15 has a voltage of 3.670V, with a voltage difference of +0.018V. This value is greater than 0 and greater than the adaptive equalization threshold of 0.015V, therefore cell number 15 is marked as "overvoltage candidate". Assume cell number 88 has a voltage of 3.630V, with a voltage difference of -0.022V. Its absolute value of 0.022V is greater than 0.015V, therefore cell number 88 is marked as "undervoltage candidate". Assuming that the absolute value of the voltage difference for most cells is within 0.015V, their status is marked as "hold".

[0100] After identifying all 100 monomers, 3 monomers were identified as "overvoltage candidate," meaning there were 3 overvoltage candidate monomers. Simultaneously, 1 monomer was identified as "undervoltage candidate."

[0101] Step 5: Implement hierarchical equilibrium decision-making and dynamic equilibrium control.

[0102] This step is another core aspect of the invention, enabling precise coordination with the charging process.

[0103] Determine the charging stage: Based on the estimated average state of charge of the battery pack of 65% in step three, and the preset state of charge switching point (usually set to about 80%-85%, depending on the battery characteristics, set to 82% in this example), determine that the current charging stage is constant current charging stage.

[0104] Layered decision-making and execution: Since the current stage is constant current charging and the number of overvoltage candidate cells (3) is greater than 0, the system enters the logic of "decision layer two".

[0105] Adjusting the charging current: The battery management system immediately sends a command to the DC charging station via a communication link (such as a CAN bus) to reduce the charging current from 120A to a preset safe current value, such as 30A. This is to prevent the voltage of overvoltaged cells from continuing to rise rapidly and triggering the danger threshold before equalization takes effect.

[0106] Initiation and Control of Equalization: Simultaneously, the battery management system controls the operation of an inductor-based bidirectional flyback active equalization circuit. This circuit can transfer energy between any selected cell and the battery pack bus. The system first identifies the "overvoltage candidate" cell (e.g., cell number 15, difference +0.018V) with the largest voltage difference based on the equalization target identification matrix. Then, the equalization circuit is controlled to transfer energy from cell number 15 to the battery pack bus. During the equalization process, the voltage of all "overvoltage candidate" cells is continuously monitored.

[0107] Dynamic Adjustment of Equalization Target: After a period of equalization, the voltage of cell number 15 drops to 3.658V, and its difference from the average voltage decreases to +0.014V, which is less than the adaptive equalization threshold of 0.015V. The battery management system immediately updates the status of cell number 15 to "Hold" and stops transferring energy from that cell. Then, the system automatically selects the next "overvoltage candidate" cell with the largest voltage difference (e.g., cell number 7) as the new equalization source and continues energy transfer. This process continues.

[0108] Collaborative energy transfer objective: During the equalization process, the system can also select to directly transfer energy from the "overvoltage candidate" cell to the "undervoltage candidate" cell (cell number 88) based on the circuit topology and capability, to achieve more efficient intra-packet direct equalization.

[0109] Cycle and Termination: While performing the above equalization control, the system continuously performs data acquisition in parallel with step one. As equalization progresses, the number of overvoltage candidate cells may decrease. The current round of centralized equalization control terminates when one of the following conditions is met: all cell status flags change to "hold"; the charging pile stops charging (current drops to 0); or any temperature exceeding the safety threshold of 45°C is detected. After termination, the system returns to step one and begins a new monitoring cycle.

[0110] In another charging scenario: when the battery pack's average state of charge reaches 85% and enters the constant-voltage charging stage, the system executes "Decision Layer Three" logic. The battery management system controls the charging pile to maintain a constant output voltage (e.g., 400V), and the charging current naturally decreases. At this time, even if there are a few "overvoltage candidate" cells, the system controls the balancing circuit to smoothly transfer the energy of these cells to the battery pack bus with a small, fixed balancing current (e.g., 1A) until charging is complete, ensuring that all cells are fully balanced and achieve a higher degree of charge saturation.

[0111] To demonstrate the effectiveness of this invention, the following two comparative examples were set up for comparative testing with the embodiments of this invention. Test conditions: A simulated battery pack (100 ternary lithium batteries in series) was charged at 25°C, starting from 20% state of charge, with a 0.5C current until the charging station automatically shut off. Initially, the capacity of 5 individual cells was artificially set to be low (95%), and the internal resistance of 5 individual cells was artificially set to be high (120%).

[0112] Comparison Projects Comparative Example 1: Traditional Fixed Threshold Active Equalization Comparative Example 2: Passive Equalization (Discharge Resistor) Embodiment of the present invention: Adaptive Equilibrium Method Equilibrium Strategy Description Using a fixed voltage difference threshold (e.g., 0.020V), balancing is initiated as soon as a single cell exceeds the tolerance during the entire charging process, and the balancing logic is simple. Using a fixed voltage difference threshold (e.g., 0.015V), the overvoltage cells are resistively discharged throughout the charging process. Adaptive threshold and hierarchical decision-making, as described in the embodiments, are employed in deep collaboration with the charging phase. Total charging time 98 minutes 105 minutes 92 minutes Average state of charge at the end of charging 95.2% 93.8% 96.5% Total energy consumption of the equilibrium process The active balancing circuit reduces the total charging energy loss by approximately 0.8%. Resistive heating results in approximately 3.5% of the total charging energy being lost. The active balancing circuit reduces the total charging energy loss by approximately 0.5%. The difference between the highest and lowest single-cell voltages at the end of charging 0.025V 0.030V 0.015V Voltage drop difference after 1 hour of standing after full charge The larger value indicates that some monomers are not actually fully filled. The large value indicates that some monomers are not actually fully filled and there is a large inconsistency. Small values ​​indicate that the charging saturation of each cell is consistent and high. evaluate While balanced and effective, it lacks coordination, takes a long time, and fails to optimally improve charging saturation. It results in significant energy waste, prolonged charging time, limited balancing capabilities, and poor performance. It features fast charging speed, high energy utilization, high balancing accuracy, and the highest final charging saturation.

[0113] As can be seen from the comparison table above, the embodiments of the present invention are significantly superior to traditional methods in terms of charging efficiency, energy utilization, and balancing effect.

[0114] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0115] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for adaptive balancing of individual battery cells during the charging process of new energy vehicles, characterized in that, The method is executed cyclically by the battery management system during the charging process of the battery pack, and the method includes the following steps: Step 1: Real-time acquisition of the total battery pack voltage, total battery pack current, terminal voltage of each individual battery cell, and temperature of each individual battery cell during the charging process; Step 2: Calculate the dispersion index of the battery cell voltage based on the terminal voltage of all battery cells, compare the dispersion index with the preset dispersion threshold, and proceed to Step 3 when the dispersion index is greater than or equal to the dispersion threshold. Step 3: Dynamically calculate the adaptive balancing threshold based on the average state of charge and average temperature of the battery pack. The average state of charge of the battery pack is estimated by the total voltage and total current of the battery pack, and the average temperature of the battery pack is calculated based on the temperature of all individual battery cells. Step 4: Based on the adaptive equalization threshold, compare and classify the terminal voltage of each battery cell with the average terminal voltage of all battery cells, construct an equalization target identification matrix containing the state identifier of each battery cell, and count the number of overvoltage candidate cells. Step 5: Based on the equalization target identification matrix, the number of overvoltage candidate cells, and the current charging stage, perform hierarchical equalization decision-making and generate control commands for the charging equipment and active equalization circuit to achieve adaptive equalization of battery cells. The execution of the hierarchical equilibrium decision specifically includes the following sub-steps: The current charging stage is determined to be either constant current charging stage or constant voltage charging stage, and the determination is based on whether the average state of charge of the battery pack is lower than the preset state of charge switching point. If the current phase is constant current charging and the number of overvoltage candidate cells is zero, but there are battery cells with the status label "undervoltage candidate" in the equalization target identification matrix, then the active equalization circuit will transfer energy from the battery pack bus to the battery cells with the status label "undervoltage candidate" while maintaining the charging equipment at the maximum allowable current. If the current stage is constant current charging and the number of overvoltage candidate cells is greater than zero, the charging equipment is controlled to reduce the output current to the preset safe current value, and the active balancing circuit is controlled to transfer energy from the battery cells with the largest voltage difference marked as "overvoltage candidate" to the battery cells or battery pack bus marked as "undervoltage candidate". If the current state is a constant voltage charging stage, the charging equipment is controlled to maintain a constant output voltage, and the active balancing circuit is controlled to transfer energy from all battery cells marked as "overvoltage candidate" to the battery pack bus with a preset balancing current. The equalization control process during the constant current charging stage, where the number of overvoltage candidate cells is greater than zero, also includes a real-time monitoring step: Real-time monitoring of the terminal voltage of all battery cells whose status is marked as "overvoltage candidate"; When the voltage difference of a battery cell whose status is marked as "overvoltage candidate" drops to less than or equal to the adaptive balancing threshold, the status of the battery cell is updated to "hold", and the balancing energy transfer operation for that battery cell is stopped.

2. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The calculation process of the dispersion index in step two includes: Calculate the average terminal voltage of all individual battery cells; Calculate the absolute difference between the terminal voltage of each individual battery cell and this average value; Calculate the average of all absolute differences and use this average as the dispersion index.

3. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The process of dynamically calculating the adaptive equilibrium threshold in step three specifically includes: The average state of charge of the battery pack is estimated by using the ampere-hour integration method, combined with the total current of the battery pack and the nominal capacity of the battery pack. Calculate the average temperature of all individual battery cells to obtain the average temperature of the battery pack; Based on the average state of charge and average temperature of the battery pack, a baseline threshold is obtained by querying a pre-established threshold mapping table. The threshold mapping table is obtained by testing the constant current charging characteristics of individual battery cells at different temperatures and states of charge and interpolating the values. The baseline threshold is multiplied by an aging compensation coefficient to obtain the final adaptive equalization threshold. The aging compensation coefficient is determined based on the number of charge cycles of the battery pack and gradually increases from 1 as the number of charge cycles increases.

4. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The process of constructing the balanced target recognition matrix in step four specifically includes: Calculate the voltage difference between the terminal voltage of each individual battery cell in the current cycle and the average terminal voltage of all individual battery cells; When the voltage difference of a certain battery cell is greater than zero and greater than the adaptive equalization threshold, the status of that battery cell is set to "overvoltage candidate". When the voltage difference of a certain battery cell is less than zero and the absolute value of the voltage difference is greater than the adaptive equalization threshold, the status of that battery cell is set to "undervoltage candidate". When the absolute value of the voltage difference of a certain battery cell is less than or equal to the adaptive equalization threshold, the status flag of that battery cell is set to "hold". The number of battery cells with the status marked as "overvoltage candidate" is recorded as the overvoltage candidate cell number.

5. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The step five, generating control commands for the charging device, specifically includes generating current commands for adjusting the output current of the charging device or voltage commands for adjusting the output voltage of the charging device. The current commands and voltage commands are sent through the communication link between the battery management system and the charging device.

6. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The active balancing circuit is either an inductor-based bidirectional flyback balancing circuit or a switched capacitor-based balancing circuit. The active balancing circuit is controlled by the battery management system and is capable of bidirectional energy transfer between any two battery cells or between a battery cell and the battery pack bus.

7. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The state of charge switching point is determined based on the typical state of charge value corresponding to the end of the constant current charging phase of the battery pack when the voltage begins to rise rapidly. This typical state of charge value is obtained by performing standard charging tests on similar battery packs.

8. The adaptive equalization method for battery cells during the charging process of a new energy vehicle according to claim 1, characterized in that, The method continues to collect data from step one during the equalization control in step five, and terminates the current round of equalization control and returns to step one for the next cycle when any of the following conditions are met: the status flags of all battery cells in the equalization target identification matrix are "hold"; the charging process is terminated by an external command; or the battery management system detects that the total voltage of the battery pack, the total current of the battery pack, or the temperature of any battery cell exceeds a safety threshold.

Images