Lithium battery intelligent equalization and life extension management method and system

Abstract

The present application belongs to the technical field of lithium battery management, and provides a lithium battery intelligent equalization and life extension management method and system. The method comprises: acquiring the state of charge (SOC) and the state of health (SOH) of each single battery in a lithium battery pack; determining the energy inconsistency degree of the lithium battery pack based on the SOC and SOH of each single battery; when the energy inconsistency degree meets the preset equalization trigger condition, dynamically selecting at least one first target single battery as a battery energy export and at least one second target single battery as a battery energy receiving based on the real-time difference of the SOC and SOH of each single battery; and transferring the battery energy from the first target single battery to the second target single battery through the pre-configured DC / DC bidirectional conversion circuit. The present application overcomes the limitation of only considering the difference in state of charge by capturing the dynamic changes of the SOC and SOH of the single battery, and constructs an energy inconsistency degree evaluation system.

Description

Technical Field

[0001] This invention relates to the field of lithium battery management technology, specifically to a method and system for intelligent balancing and lifespan extension management of lithium batteries. Background Technology

[0002] Portable DC power supply devices are crucial for ensuring power continuity in scenarios such as emergency power system maintenance, temporary substation backup power, and temporary power supply for outdoor equipment. Their core function relies on battery banks for energy support. These devices are typically in a "maintenance standby" state, only activated under special circumstances such as grid failures or equipment maintenance. However, as the core energy storage unit, the battery bank is prone to performance degradation or even failure during long-term static storage or intermittent use due to the following issues: First, differences in manufacturing processes, fluctuations in storage environment temperature and humidity, and the accumulation of charge-discharge cycles gradually lead to inconsistencies in capacity, internal resistance, and self-discharge rate among individual batteries within the bank, manifesting as discrepancies in state of charge (SOC) and state of health (SOH). Second, the lack of targeted periodic management means that "static losses" and "polarization accumulation" during long-term standby can lead to inaccurate battery capacity claims. Traditional devices lack integrated annual activation and online capacity verification functions, making it impossible to promptly eliminate polarization and verify actual capacity, potentially resulting in insufficient power supply during emergency activation.

[0003] Existing management solutions for conventional portable DC power supplies have significant limitations and are difficult to adapt to the requirements of the above scenarios, mainly in the following aspects: First, most of them adopt an integrated design of multiple batteries connected in series to form a DC bus. The power supply stability is completely dependent on the health status of a single battery. If a battery leaks or its internal resistance increases suddenly, it will directly cause the voltage of the entire bus to drop, or even cause the entire device to fail. This cannot meet the core requirement of "high reliability" in emergency scenarios.

[0004] Secondly, the accompanying battery management systems (BMS) generally use SOC as the sole basis for equalization triggering and control, neglecting the critical parameter SOH. For example, a battery with a high SOC but severely degraded SOH (actual capacity only 60% of the initial value) will accelerate its aging and may cause thermal runaway if it continues to be used as an energy source; while a healthy battery with a low SOC will further exacerbate the inconsistencies within the group and shorten the overall lifespan if it cannot be replenished in time.

[0005] Third, relying on conventional AC / DC rectifier modules for power supply has weak tolerance to AC input voltage fluctuations. When the AC power grid experiences short-term high voltage, low voltage, or sudden voltage drop, the rectifier module is prone to triggering protection shutdown, resulting in unstable DC bus voltage and inability to provide continuous power supply for emergency loads. This problem is particularly prominent in outdoor temporary power supply scenarios.

[0006] Fourth, there is a lack of dedicated management functions for "standby scenarios": On the one hand, there is a lack of an annual activation mechanism, making it impossible to eliminate the polarization effect caused by long-term battery idling through charge-discharge cycles, leading to gradual capacity decay; on the other hand, there is no online capacity verification function, making it impossible to verify the actual capacity of the battery in real time, making it difficult to identify deteriorated batteries in advance, and easily leading to the risk of "standby but useless" when activated in an emergency. At the same time, the traditional series architecture has extremely high requirements for battery consistency, making it impossible to mix batteries of different brands and manufacturing years, increasing maintenance costs and the difficulty of spare parts management.

[0007] Fifth, the device is large and heavy, lacks miniaturization and portability, and the primary and secondary interface access methods are cumbersome, which cannot meet the needs of "rapid transportation and rapid on-site access" in emergency maintenance scenarios. At the same time, the fixed architecture cannot be flexibly expanded or the output adjusted according to the size of the on-site load, making it difficult to match the diverse power supply needs in different scenarios.

[0008] Therefore, there is an urgent need for an intelligent management solution that is compatible with portable, split-type parallel architecture, has AC high and low voltage ride-through capability, and integrates battery activation and capacity control functions, in order to solve the problems of low reliability, lack of battery management, and poor adaptability to scenarios in existing devices, and to ensure the safety and continuity of power supply in scenarios such as power emergency maintenance and temporary backup power. Summary of the Invention

[0009] This invention provides a method and system for intelligent balancing and lifespan extension management of lithium batteries. It aims to address the limitations of existing lithium battery management systems in terms of balancing decision-making, control strategies, state estimation accuracy, and system architecture flexibility. By constructing a multi-parameter collaborative balancing decision-making mechanism that integrates state of charge (SOC) and state of health (SOH), and designing an adaptive balancing control strategy based on dynamic operating conditions, it achieves accurate and efficient energy balancing of the lithium battery pack, thereby significantly improving the consistency and overall cycle life of the lithium battery pack and reducing safety risks caused by inconsistencies in individual cells.

[0010] This invention provides a method for intelligent balancing and lifespan extension management of lithium batteries, comprising the following steps: Obtain the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack. The energy inconsistency of the lithium battery pack is determined based on the SOC and SOH of each individual cell. When the energy inconsistency meets the preset equalization trigger condition, based on the real-time difference between the SOC and SOH of each individual cell, at least one first target cell as the first target cell for energy transfer and at least one second target cell as the second target cell for energy reception are dynamically selected, and the battery energy is transferred from the first target cell to the second target cell by controlling the pre-configured DC / DC bidirectional conversion circuit. The DC / DC bidirectional converter circuit is connected between the first target cell and the second target cell and is configured to support bidirectional energy transfer. The direction, magnitude and duration of the transfer current during the energy transfer process are adjusted in real time according to the SOC difference between the first target cell and the second target cell, the SOH decay degree and the current operating temperature of the lithium battery pack.

[0011] Furthermore, the equalization trigger condition is: the duration of energy inconsistency is greater than or equal to the preset trigger delay, and any of the following sub-conditions are satisfied: (1) The energy inconsistency is greater than or equal to the preset first energy inconsistency threshold; (2) The SOC of a single cell exceeds the extreme range, which is SOC ± 8% of the average SOC of the lithium battery pack; (3) The energy inconsistency is greater than or equal to the second energy inconsistency threshold and less than the first energy inconsistency threshold, and the current working state of the lithium battery pack is either static or constant current charging and discharging; the second energy inconsistency threshold is 70% of the first energy inconsistency threshold; (4) The time interval between two consecutive equalization triggers is greater than or equal to the set minimum equalization trigger period.

[0012] Furthermore, the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack are obtained, including: A closed-loop online estimation of the State of Charge (SOC) of each individual cell is performed using a battery model-based filtering estimation algorithm. The battery model is a second-order RC equivalent circuit model, which includes a series resistor to simulate the ohmic internal resistance of the individual cell, two parallel RC networks to simulate the electrochemical polarization and concentration polarization of the individual cell, and an ideal voltage source to simulate the open-circuit voltage of the individual cell. The filtering estimation algorithm is an adaptive extended Kalman filter, which can dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix according to the real-time operating temperature and aging degree of the individual cell. A bidirectional long short-term memory network (BiLSTM) based on an attention mechanism is used to comprehensively analyze the impulse response characteristics, cycle history data, and online core capacity test data of a single cell to obtain the state of health (SOH). The impulse response characteristics include ohmic internal resistance, polarization internal resistance, and diffusion capacitance, while the cycle history data includes the number of cycles, cumulative charge-discharge capacity, and deep discharge count.

[0013] Furthermore, the pulse response characteristic parameters are obtained by applying a specific pulse excitation signal to a single cell under preset operating conditions, and by fitting the voltage response curves collected during pulse discharge and rest periods using the least squares method. The preset operating conditions are: under a constant temperature environment of 25℃±2℃, when the SOC of the single cell is 20%, 50%, and 80%, respectively, and the specific pulse excitation signal is: constant current discharge at a 1C rate for 10 seconds, followed by rest for 30 seconds.

[0014] Furthermore, based on the real-time differences in SOC and SOH of each individual cell, at least one target cell for energy transfer and at least one target cell for energy reception are dynamically selected, including: Individual cells with a SOC value higher than a first threshold and a SOH value higher than a second threshold are identified as the first target individual cells. The first threshold is set to 105% of the average SOC value of the lithium battery pack, and the second threshold is set to 80% of the initial SOH value of the lithium battery pack. Individual cells with a SOC value below the third threshold are identified as the second target individual cells. The third threshold is set at 95% of the average SOC value of the lithium battery pack.

[0015] Furthermore, the direction, magnitude, and duration of the transfer current can be adjusted in real time based on the SOC difference between the first target cell and the second target cell, the degree of SOH decay, and the current operating temperature of the lithium battery pack, including: Based on the magnitude of the comprehensive value of energy inconsistency calculated from the SOC difference, SOH decay difference, and the current operating temperature of the lithium battery pack, at least two different power level equalization modes are triggered in stages to achieve different methods of transfer current regulation, wherein the higher power level corresponds to a larger transfer current. When the comprehensive value of energy inconsistency is in the first interval, the first power level balancing mode is triggered, and the transfer current is I1; when the comprehensive value of energy inconsistency is in the second interval, the second power level balancing mode is triggered, and the transfer current is I2; where I2 > I1, and the upper limit of the first interval is less than the lower limit of the second interval. If the duration of the comprehensive energy inconsistency value in the first interval does not reach the set first minimum holding time, or the duration of the comprehensive energy inconsistency value in the second interval does not reach the set second minimum holding time, then the first power level equalization mode will not be triggered or the second power level equalization mode will be triggered. If the current operating temperature of the lithium battery pack exceeds 80% of the temperature threshold, the transfer current corresponding to each power level will be reduced by 20%-50%; if the difference in SOH decay between the first target cell and the second target cell exceeds 50% of the SOH decay difference threshold, the adjustment rate of the transfer current will be reduced by 10%-30%.

[0016] Furthermore, the overall energy inconsistency value is calculated using the following formula: , in, Represents the overall value of energy inconsistency. Represents the SOC difference. Represents the SOC difference threshold. This represents the difference in the degree of SOH decay. The threshold representing the difference in the degree of SOH decay. This represents the current operating temperature of the lithium battery pack. Represents the temperature threshold. , , These are the weighting coefficients for the SOC difference, the SOH degradation difference, and the current operating temperature of the lithium battery pack, respectively. The SOC difference threshold is the maximum allowable difference in SOC between individual cells; the SOH decay difference threshold is the maximum allowable difference in SOH decay between individual cells; and the temperature threshold is the highest allowable temperature for normal operation of the lithium battery pack.

[0017] Furthermore, it also includes: When charging the lithium battery pack, the parameters of the charging curve of the lithium battery pack are adaptively adjusted according to the average SOH and resting time of the lithium battery pack. When the average SOH is higher than the first SOH threshold, standard charging parameters are used; when the average SOH is between the first SOH threshold and the second SOH threshold, the current value of the constant current charging stage is reduced, and the voltage value of the constant voltage charging stage is reduced by a small amount; when the average SOH is lower than the second SOH threshold, the current value of the constant current charging stage is further reduced, and the voltage value of the constant voltage charging stage is reduced by a larger amount, and the transition threshold of the constant voltage stage is triggered in advance; wherein, the parameters of the charging curve include the current value of the constant current charging stage, the voltage value of the constant voltage charging stage, and the transition threshold conditions between the constant current charging stage and the constant voltage charging stage, and the first SOH threshold is greater than the second SOH threshold.

[0018] A management system for implementing intelligent balancing and life extension management of lithium batteries, including a converter array, a main controller, a sampling module, and a protection module; The converter array consists of multiple DC / DC bidirectional conversion circuits, each corresponding to at least one individual battery cell, and is switchably connected to each individual battery cell via a matrix switching network. The DC / DC bidirectional conversion circuit adopts a bidirectional synchronous Buck-Boost topology, including power switches, energy storage components, filter capacitors, and freewheeling structures, and is configured to support switching between Buck mode, Boost mode, and synchronous rectification mode. The matrix switching network is configured to allow the positive and negative terminals of each individual battery cell to be selectively connected to the input or output terminal of any DC / DC bidirectional conversion circuit. The main controller is communicatively connected to the converter array and is configured to: acquire the SOC and SOH of each individual cell, determine the energy inconsistency and energy transfer strategy based on the SOC and SOH, control the converter array to establish an energy transfer path from the first target cell to the second target cell, and control the direction, magnitude and duration of the transfer current flowing through the path. The sampling module is configured to acquire the input / output voltage and transfer current of the DC / DC bidirectional converter circuit, with the sampling accuracy meeting the conditions that the transfer current is ≤ ±1% FS and the input / output voltage is ≤ ±0.5% FS. The protection module is configured to provide overcurrent protection, overvoltage protection, and overheat protection.

[0019] Furthermore, the main controller is also configured to dynamically generate charge and discharge power boundary parameters based on the heterogeneous characteristics of each individual battery cell, so as to achieve synergistic optimization of charging strategy, equalization strategy and thermal management strategy. The heterogeneous characteristics include differentiated rated capacity, diverse aging degree and potential performance dispersion characteristics. The charge and discharge power boundary parameters include charge and discharge rate limit, power threshold, SOC window boundary and temperature correction factor. The diverse aging degree includes cycle aging, calendar aging and storage aging. The potential performance dispersion characteristics include capacity decay rate, internal resistance growth characteristics, charge and discharge efficiency and temperature sensitivity differences.

[0020] Compared with existing technologies, this invention has the following advantages and beneficial effects: First, through a multi-parameter collaborative equalization decision-making mechanism, it achieves dynamic comprehensive evaluation of SOC and SOH, avoiding the problem of incomplete or over-equalization caused by single-parameter decision-making, and making the equalization trigger more in line with the actual state requirements of the lithium battery pack; Second, the adaptive equalization control strategy can dynamically adjust the energy transfer parameters according to the real-time operating conditions, temperature and aging degree of the lithium battery pack, significantly improving energy transfer efficiency and reducing energy loss during the equalization process; Third, the SOC estimation method based on the second-order RC equivalent circuit model and adaptive extended Kalman filter, combined with the SOH prediction model of BiLSTM network, effectively improves the accuracy and robustness of state estimation, providing reliable data support for equalization decision-making; Fourth, the combined architecture of matrix switching network and DC / DC bidirectional converter circuit enhances the system's adaptability to different numbers and specifications of single cells, improves the system's flexibility and scalability, and can meet the needs of diverse application scenarios.

[0021] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0022] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 A schematic diagram illustrating the steps of intelligent balancing and lifespan extension management for lithium batteries; Figure 2 A schematic diagram of the steps for obtaining the state of charge (SOC) and state of health (SOH) of each individual cell in a lithium battery pack; Figure 3 This is a schematic diagram of the management system structure used to implement intelligent balancing and life extension management methods for lithium batteries. Detailed Implementation

[0024] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0025] This invention provides a method for intelligent balancing and lifespan extension management of lithium batteries, such as... Figure 1 As shown, it includes the following steps: Obtain the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack. The energy inconsistency of the lithium battery pack is determined based on the SOC and SOH of each individual cell. When the energy inconsistency meets the preset equalization trigger condition, based on the real-time difference between the SOC and SOH of each individual cell, at least one first target cell as the first target cell for energy transfer and at least one second target cell as the second target cell for energy reception are dynamically selected, and the battery energy is transferred from the first target cell to the second target cell by controlling the pre-configured DC / DC bidirectional conversion circuit. The DC / DC bidirectional converter circuit is connected between the first target cell and the second target cell and is configured to support bidirectional energy transfer. The direction, magnitude and duration of the transfer current during the energy transfer process are adjusted in real time according to the SOC difference between the first target cell and the second target cell, the SOH decay degree and the current operating temperature of the lithium battery pack. In a specific application, such as a large energy storage power station using a 48-cell series-parallel lithium battery cluster, during grid-connected charging and discharging, the monitoring system acquires the SOC and SOH data of each individual cell every 5 minutes. During a 3-hour charging process, the system detected that the SOC of cells 18 (SOC=89%, SOH=85%) and 22 (SOC=90%, SOH=84%) in the third series of lithium batteries was significantly higher than that of other cells in the same group (average SOC=82%). The energy inconsistency of this series of lithium batteries was calculated to be 9%, reaching the preset equalization trigger condition (≥7%). Subsequently, analysis revealed that the SOH of these two individual cells was relatively low (below the cluster average SOH). The battery with a SOC of 88% was selected as the first target cell. Simultaneously, cells with lower SOCs (SOC=78%, SOH=90%) and 45 (SOC=77%, SOH=89%) were selected as the second target cells. Since the current operating temperature of the lithium battery pack is 25℃ (ideal temperature) and the SOC difference is relatively large (approximately 12%), the initial transfer current of the DC / DC bidirectional converter was set to 2.5A. As the SOC difference gradually decreased during energy transfer, the current was automatically reduced to 0.8A. After 40 minutes of dynamic balancing, the SOC of the first target cell dropped to 83%, while the SOC of the second target cell rose to 82%. The energy inconsistency of the lithium battery pack decreased to 3%, ensuring the consistency of charging and discharging of the entire battery cluster. This improved the charging and discharging efficiency of the energy storage station by approximately 5% and reduced capacity loss due to overcharging of high-SOC cells, extending the expected lifespan of the lithium battery pack by approximately 1.5 years.

[0026] The working principle of the above technical solution is as follows: First, the system collects the state of charge (SOC, i.e., the percentage of the battery's current charge relative to its rated capacity) and state of health (SOH, i.e., the ratio of the battery's current actual capacity to its nominal capacity, reflecting the degree of battery aging and degradation) of all individual cells in the lithium battery pack in real time. Next, based on the acquired SOC and SOH of each individual cell, a specific algorithm is used to comprehensively calculate and quantify the energy inconsistency of the entire lithium battery pack. This indicator comprehensively reflects the energy imbalance of the individual cells within the pack. When the calculated energy inconsistency reaches or exceeds a preset balancing trigger condition (e.g., the inconsistency exceeds a certain set threshold), the system initiates the balancing management process. At this time, based on the real-time differences in SOC and SOH of each individual cell, at least one cell with a relatively high state of charge that can release energy is dynamically selected as the first target cell (energy exporter), and at least one cell with a relatively low state of charge that needs energy replenishment is selected as the second target cell (energy receiver). Subsequently, the system controls the pre-built DC / DC bidirectional converter between the first and second target cells. The circuit begins operation. A DC / DC bidirectional converter is a power conversion device that enables bidirectional flow of electrical energy between two ports. It can boost or buck the electrical energy of the first target cell and transfer it to the second target cell, and can also reverse the energy transfer when necessary, thus supporting flexible bidirectional energy transfer. During the energy transfer process, the system dynamically adjusts the direction (ensuring flow from the first target to the second target), magnitude (precisely controlling the energy transfer rate), and duration (ensuring just the right amount of energy transfer, achieving a balanced effect without over-charging or discharging) of the transfer current based on the SOC difference between the first and second target cells (a larger difference usually requires a larger initial transfer force), the degree of SOH decay (for batteries with low SOH and severe aging, it may be necessary to limit the magnitude and duration of the current involved in energy transfer to avoid further damage from over-charging and discharging), and the current operating temperature of the lithium battery pack (excessive or insufficient temperature will affect battery performance and safety, requiring adjustment of parameters such as current accordingly). Ultimately, this achieves energy balancing of the individual cells within the lithium battery pack, thereby effectively extending the lifespan of the entire lithium battery pack.

[0027] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, the energy inconsistency of the lithium battery pack can be accurately quantified by real-time monitoring of the SOC and SOH parameters of individual cells, ensuring that equalization management is activated in a timely manner when necessary; the target individual cells for energy transfer and reception are dynamically selected, and the transfer current parameters of the DC / DC bidirectional converter circuit are adjusted in real time in combination with the SOC difference, SOH decay degree and operating temperature, thereby achieving refined and intelligent energy equalization, improving the consistency of charging and discharging of the lithium battery pack and the overall charging and discharging efficiency, avoiding the problems of overcharging of high SOC individual cells and overcharging and discharging of low SOH individual cells, thus significantly extending the cycle life of the lithium battery pack while ensuring the safety of battery use.

[0028] In one embodiment, the equalization triggering condition is: the duration of energy inconsistency is greater than or equal to a preset trigger delay, and any of the following sub-conditions are satisfied: (1) The energy inconsistency is greater than or equal to the preset first energy inconsistency threshold; (2) The SOC of a single cell exceeds the extreme range, which is SOC ± 8% of the average SOC of the lithium battery pack; (3) The energy inconsistency is greater than or equal to the second energy inconsistency threshold and less than the first energy inconsistency threshold, and the current working state of the lithium battery pack is either static or constant current charging and discharging; the second energy inconsistency threshold is 70% of the first energy inconsistency threshold; (4) The time interval between two consecutive equalization triggers is greater than or equal to the set minimum equalization trigger period; Assuming a lithium battery pack has a first energy inconsistency threshold of 15% and a second energy inconsistency threshold of 10.5% (15% × 70%), a preset trigger delay of 30 seconds, a minimum equalization trigger cycle of 2 hours, and an average SOC of 50%, then the extreme range is 42%-58% (50% ± 8%). The following is an example of whether the specific equalization trigger conditions can be implemented: The energy inconsistency is 16% and lasts for 35 seconds. Since 16% ≥ 15% (satisfying sub-condition 1) and the duration is 35 seconds ≥ 30 seconds, equilibrium is triggered. The energy inconsistency is 12%, the duration is 40 seconds, a certain single cell has a SOC of 60%, the average SOC is 50%, the extreme range is 42%-58%, 60% is out of range (satisfying sub-condition 2), the duration is 40 seconds ≥ 30 seconds, triggering equalization; The energy inconsistency is 11%, lasting for 30 seconds. The lithium battery pack is in a static state. 11% ≥ 10.5% and < 15% (meeting the threshold requirement in sub-condition 3). The working state is static, lasting for 30 seconds ≥ 30 seconds, triggering equalization. The last equalization trigger time was 10:00. The energy inconsistency this time is 14% and the duration is 30 seconds. The current time is 12:10. The condition 14% ≥ 15% is not met. The SOC of the individual cells is all in the range of 42%-58%. The condition 14% ≥ 10.5% is not met, but the cells are currently in a dynamic charge and discharge state (sub-condition 3 is not met). However, the interval between adjacent triggers is 130 minutes ≥ 120 minutes (sub-condition 4 is met), so equalization is triggered. The energy inconsistency is 10%, lasting for 25 seconds. If 10% < 10.5% (sub-conditions 1, 2, and 3 are not met), and the duration is 25 seconds < 30 seconds, the equilibrium will not be triggered. The energy inconsistency is 12%, lasting for 30 seconds. The last equalization trigger time was 11:30, and the current time is 13:20. 12% ≥ 10.5% and < 15%, but it is currently in a dynamic charging and discharging state (sub-condition 3 is not met). The adjacent interval is 110 minutes < 120 minutes (sub-condition 4 is not met), so equalization is not triggered.

[0029] The working principle of the above technical solution is as follows: by comprehensively considering factors such as the degree of energy inconsistency of the lithium battery pack, the deviation of the state of charge (SOC) of individual cells, the current working state of the lithium battery pack, and the time interval between two equalization operations, the equalization function is activated at the appropriate time to achieve effective protection and performance optimization of the lithium battery pack. First, the duration of energy inconsistency must be greater than or equal to the preset trigger delay. This delay setting is to avoid false triggering caused by instantaneous fluctuations or interference signals, and to ensure that the detected energy inconsistency state is real and continuous, thereby improving the accuracy and reliability of equalization triggering. Secondly, based on satisfying the above duration condition, further determine whether any one of the following four sub-conditions is met: The first sub-condition is that the energy inconsistency is greater than or equal to the preset first energy inconsistency threshold. The first energy inconsistency threshold is usually set to a relatively high value, which means that the energy difference between individual cells in the lithium battery pack is already quite significant, reaching the level where equalization intervention is necessary to prevent the difference from further expanding and seriously affecting the overall performance and lifespan of the lithium battery pack. The second sub-condition focuses on the SOC state of individual cells. When the SOC of any individual cell exceeds the extreme range defined by ±8% of the average SOC of the lithium battery pack, equalization is triggered. SOC (State of Charge) refers to the state of charge of the battery, that is, the percentage of the current usable capacity of the battery relative to its rated capacity. If the SOC of an individual cell deviates too much from the average SOC of the lithium battery pack, it means that the individual cell may be on the verge of overcharging or over-discharging, or its compatibility with other individual cells has deteriorated. At this time, equalization can bring the SOC of each individual cell back to a reasonable range, ensuring the safe operation and consistency of the lithium battery pack. The third sub-condition requires that the energy inconsistency is greater than or equal to the second energy inconsistency threshold (which is 70% of the first energy inconsistency threshold), but less than the first energy inconsistency threshold, and that the current operating state of the lithium battery pack is either quiescent or constant current charging / discharging. This means that when the energy inconsistency is at a moderate level (not yet reaching the severity of the first threshold), equalization is not triggered immediately, but needs to be judged in conjunction with the operating state of the lithium battery pack. Equalization is chosen to be performed in the quiescent or constant current charging / discharging state because the operating conditions of the lithium battery pack are relatively stable in these states. Equalization at this time can achieve a more precise adjustment effect and has less impact on the current main charging / discharging process of the lithium battery pack. It can more effectively utilize specific operating stages to improve battery consistency without wasting energy or affecting the main task under complex and variable operating conditions. The fourth sub-condition is that the time interval between two adjacent equalization triggers is greater than or equal to the set minimum equalization trigger period. This is to prevent equalization operations from being performed too frequently. The equalization process itself also consumes a certain amount of energy, and too frequent operations may put unnecessary burden on the battery management system and the battery itself. Setting a minimum equalization trigger period can ensure that after an equalization operation is completed, the lithium battery pack has enough time to stabilize its state or to observe the changing trend of energy inconsistency, avoiding repeated equalization triggers in a short period of time, and improving system efficiency and stability. The beneficial effects of the above technical solution are as follows: By using the solution provided in this embodiment, the duration of energy inconsistency is taken as a premise, and combined with the satisfaction of any one of the four sub-conditions, namely, different threshold levels of energy inconsistency, deviation range of single cell SOC, real-time working status of lithium battery pack, and time interval limit of equalization operation, it is possible to comprehensively determine whether to start equalization, thereby achieving accurate, reliable and efficient control of the timing of lithium battery pack equalization.

[0030] In one embodiment, such as Figure 2 As shown, the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack are obtained, including: A closed-loop online estimation of the State of Charge (SOC) of each individual cell is performed using a battery model-based filtering estimation algorithm. The battery model is a second-order RC equivalent circuit model, which includes a series resistor to simulate the ohmic internal resistance of the individual cell, two parallel RC networks to simulate the electrochemical polarization and concentration polarization of the individual cell, and an ideal voltage source to simulate the open-circuit voltage of the individual cell. The filtering estimation algorithm is an adaptive extended Kalman filter, which can dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix according to the real-time operating temperature and aging degree of the individual cell. A bidirectional long short-term memory network (BiLSTM) based on an attention mechanism is used to comprehensively analyze the impulse response characteristics, cycle history data, and online core capacity test data of a single cell to obtain the state of health (SOH). The impulse response characteristics include ohmic internal resistance, polarization internal resistance, and diffusion capacitance, while the cycle history data includes the number of cycles, cumulative charge-discharge capacity, and deep discharge count.

[0031] The working principle of the above technical solution is as follows: by integrating model-based filtering estimation with deep learning-based network analysis, the SOC and SOH of individual cells in the lithium battery pack can be accurately obtained. For the estimation of the state of charge (SOC), an adaptive extended Kalman filter algorithm based on a second-order RC equivalent circuit model is used for closed-loop online estimation. This second-order RC equivalent circuit model uses a series resistor to simulate the ohmic internal resistance of a single cell, which reflects the instantaneous resistance encountered when current flows through the internal materials of the cell. Simultaneously, the model includes two parallel RC networks: one to simulate the electrochemical polarization phenomenon of a single cell, i.e., the voltage hysteresis caused by charge accumulation and consumption during the electrochemical reaction; and the other to simulate concentration polarization, which is the voltage change caused by the uneven distribution of reactant concentrations within the cell. Furthermore, the model also includes... An ideal voltage source is used to simulate the open-circuit voltage of a single battery cell, i.e., the terminal voltage of the battery under no-load conditions. The adaptive extended Kalman filter algorithm can dynamically adjust the process noise covariance matrix (which characterizes the noise level caused by the uncertainty of the system model) and the measurement noise covariance matrix (which reflects the noise characteristics introduced during the sensor measurement process) based on the real-time operating temperature of the single battery cell during operation (temperature changes will significantly affect the electrochemical characteristics and internal resistance of the battery) and the degree of aging (the performance degradation of the battery after long-term use). This ensures the accuracy and robustness of SOC estimation under different operating conditions. For obtaining State of Health (SOH), a Bidirectional Long Short-Term Memory (BiLSTM) network based on an attention mechanism is used to comprehensively analyze the impulse response characteristic parameters, cycle history data, and online core capacity test data of individual cells. Specifically, the impulse response characteristic parameters include ohmic internal resistance (consistent with the series resistance in the SOC estimation model; its magnitude directly reflects the patency of the internal conductive pathways of the battery, and it increases with aging), polarization internal resistance (internal resistance related to electrochemical polarization and concentration polarization, which also changes with battery aging), and diffusion capacitance (capacitive characteristics characterizing the ion diffusion process inside the battery, reflecting changes in the battery's ability to store and release charge); cycle history data includes the number of battery cycles (the number of times the battery completes a full charge-discharge cycle, an important indicator of battery aging), and cumulative charge-discharge capacity (…). The total charge and discharge capacity of the battery from the start of use to the present reflects the battery's total energy throughput, as does the number of deep discharge cycles (the number of times the battery discharges to a lower state of charge, which has a significant impact on battery life). Online capacity test data is the actual capacity data of the battery obtained in real time during normal battery operation through specific charge and discharge strategies (such as low-current constant-current charge and discharge), which can directly reflect the battery's current effective capacity state. The BiLSTM network can effectively capture long-term dependencies in time series data, and its bidirectional propagation characteristics allow it to use both past and future information for analysis. The introduction of the attention mechanism enables the network to automatically focus on the more critical features and time segments for SOH assessment when processing this data, thereby improving the accuracy of SOH estimation and achieving accurate judgment of battery health status.

[0032] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, real-time and accurate monitoring of the current state of charge of the battery is achieved through a refined battery model and adaptive filtering algorithm. At the same time, by using an advanced deep learning network combined with multi-dimensional battery characteristic parameters and historical data, a comprehensive and accurate assessment of the battery health status is achieved. The two complement each other and together provide key technical support for the efficient management and safe and reliable operation of lithium battery packs.

[0033] In one embodiment, the pulse response characteristic parameters are obtained by applying a specific pulse excitation signal to a single cell under preset operating conditions, and by fitting the voltage response curves collected during pulse discharge and rest periods using the least squares method. The preset operating conditions are: under a constant temperature environment of 25℃±2℃, when the SOC of the single cell is 20%, 50%, and 80%, respectively, and the specific pulse excitation signal is: constant current discharge at a 1C rate for 10 seconds, followed by rest for 30 seconds.

[0034] The working principle of the above technical solution is as follows: In a constant temperature environment of 25℃±2℃, for a single battery cell, its state of charge (SOC) is first precisely adjusted to three representative levels: 20%, 50%, and 80%. These different SOC states constitute the preset test conditions, aiming to comprehensively examine the battery characteristics under different charge levels. Subsequently, at each SOC state, a specific pulse excitation signal is applied to the battery, specifically a 1C discharge rate for 10 seconds of constant current discharge, followed by a 30-second rest period. During this process, voltage response data of the battery during the pulse discharge phase and the subsequent rest period are continuously collected, thus obtaining a complete voltage-time curve. Finally, the least squares method is used to fit the collected voltage response curve with the theoretical model. By minimizing the sum of squares error between the actual data and the model prediction, the pulse response characteristic parameters that characterize the battery's electrochemical reaction kinetics and internal structural features are accurately solved. These parameters reflect the voltage change law of the battery under specific excitation, thus providing key basis for battery performance evaluation and state estimation.

[0035] The beneficial effects of the above technical solution are as follows: by adopting the solution provided in this embodiment, the pulse response data of the battery can be obtained in a standardized constant temperature environment and under multiple SOC states. Characteristic parameters are extracted through scientific fitting methods, effectively eliminating the interference of temperature fluctuations and single charge states on the test results, making the obtained parameters more representative and accurate.

[0036] In one embodiment, based on the real-time differences in the SOC and SOH of each individual cell, at least one first target cell for energy transfer and at least one second target cell for energy reception are dynamically selected, including: Individual cells with a SOC value higher than a first threshold and a SOH value higher than a second threshold are identified as the first target individual cells. The first threshold is set to 105% of the average SOC value of the lithium battery pack, and the second threshold is set to 80% of the initial SOH value of the lithium battery pack. Individual cells with a SOC value below the third threshold are identified as the second target individual cells. The third threshold is set at 95% of the average SOC value of the lithium battery pack.

[0037] The working principle of the above technical solution is as follows: Based on the real-time differences in the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack, the energy distribution of the lithium battery pack is dynamically optimized. For example, if a single cell currently has a charge of 45Ah and a rated capacity of 50Ah, its SOC is 90%. If a single cell initially has a capacity of 50Ah but its capacity decays to 40Ah after use, its SOH is 80%. First, the system monitors and calculates the average SOC value of the entire lithium battery pack and the initial SOH value of each individual cell in real time (as a benchmark). Then, a first threshold is set at 105% of the average SOC value of the lithium battery pack, and a second threshold is set at 80% of the initial SOH value of the lithium battery pack. Individual cells that simultaneously meet the conditions of having an SOC value higher than the first threshold (i.e., relatively sufficient charge) and an SOH value higher than the second threshold (i.e., good health and able to withstand energy output) are selected as the first target individual cells. These cells will serve as energy output sources. At the same time, a third threshold is set at 95% of the average SOC value of the lithium battery pack. Individual cells with an SOC value lower than the third threshold (i.e., relatively insufficient charge) are selected as the first target individual cells. The first target battery cell is identified as the second target battery cell. These cells will act as energy receivers. Through this dynamic selection mechanism, energy can be transferred from the first target battery cell, which is fully charged and in good condition, to the second target battery cell, which is undercharged. This balances the SOC levels of each battery cell and reduces the performance degradation or lifespan reduction of the lithium battery pack caused by differences in individual cells. For example, assuming a lithium battery pack consists of 10 battery cells with a calculated average SOC of 80%, the first threshold is 80% × 105% = 84%, and the third threshold is 80% × 95% = 84%. 76%; If the initial SOH of the lithium battery pack is 100%, then the second threshold is 100% × 80% = 80%. At this time, if a certain single cell has an SOC of 86% (higher than 84%) and an SOH of 85% (higher than 80%), it is selected as the first target single cell; if another single cell has an SOC of 72% (lower than 76%), it is selected as the second target single cell. Then, the energy transfer process will be started to transfer part of the energy of the 86% SOC single cell to the 72% SOC single cell to achieve the balance of the SOC of the single cells in the lithium battery pack.

[0038] The beneficial effects of the above technical solution are as follows: by adopting the solution provided in this embodiment, it is possible to accurately identify individual cells in the lithium battery pack that have energy transfer capability and meet the health standards, and at the same time quickly locate individual cells with insufficient power that need to be replenished. Through this targeted dynamic screening mechanism, the energy loss and low efficiency problems caused by the indiscriminate treatment of all individual cells in the traditional equalization method are avoided.

[0039] In one embodiment, the direction, magnitude, and duration of the transfer current can be adjusted in real time based on the SOC difference between the first target cell and the second target cell, the degree of SOH decay, and the current operating temperature of the lithium battery pack, including: Based on the magnitude of the comprehensive value of energy inconsistency calculated from the SOC difference, SOH decay difference, and the current operating temperature of the lithium battery pack, at least two different power level equalization modes are triggered in stages to achieve different methods of transfer current regulation, wherein the higher power level corresponds to a larger transfer current. When the comprehensive value of energy inconsistency is in the first interval, the first power level balancing mode is triggered, and the transfer current is I1; when the comprehensive value of energy inconsistency is in the second interval, the second power level balancing mode is triggered, and the transfer current is I2; where I2 > I1, and the upper limit of the first interval is less than the lower limit of the second interval. If the duration of the comprehensive energy inconsistency value in the first interval does not reach the set first minimum holding time, or the duration of the comprehensive energy inconsistency value in the second interval does not reach the set second minimum holding time, then the first power level equalization mode will not be triggered or the second power level equalization mode will be triggered. If the current operating temperature of the lithium battery pack exceeds 80% of the temperature threshold, the transfer current corresponding to each power level will be reduced by 20%-50%; if the difference in SOH decay between the first target cell and the second target cell exceeds 50% of the SOH decay difference threshold, the adjustment rate of the transfer current will be reduced by 10%-30%.

[0040] The working principle of the above technical solution is as follows: by comprehensively considering the state differences of individual cells in the lithium battery pack and environmental factors, intelligent and precise balance control is achieved, specifically: First, the system collects the SOC difference between the first and second target single cells in real time, which is the difference in the current percentage of charge of the two cells; the SOH decay difference, which is the difference in the performance degradation of the two cells after long-term use; and the current operating temperature of the lithium battery pack. Based on these three key parameters, the system calculates a comprehensive energy inconsistency value, which fully reflects the overall imbalance between the target single cells in terms of energy state, health status, and current operating environment. Next, based on the magnitude of the comprehensive energy inconsistency value, the system will trigger at least two different power level balancing modes in stages to adjust the transfer current in different ways. The power level is directly related to the magnitude of the transfer current. A higher power level corresponds to a larger transfer current, which means that energy is transferred from one cell to another faster, in order to reduce the energy gap between them more quickly. When the calculated comprehensive energy inconsistency value is in the preset first interval, the system will trigger the first power level balancing mode, and the transfer current is set to I1. When the comprehensive energy inconsistency value is in the higher second interval, the second power level balancing mode is triggered, and the corresponding transfer current is I2. The value of I2 is greater than I1, and the upper limit of the first interval is set to be less than the lower limit of the second interval, ensuring the clarity of the interval division and the rigor of the triggering logic, and avoiding frequent jitter during mode switching. To prevent the balancing mode from being falsely triggered due to instantaneous fluctuations, a duration judgment mechanism is set up. If the duration of the comprehensive value of energy inconsistency in the first interval has not reached the preset first minimum holding time, or the duration of the value in the second interval has not reached the preset second minimum holding time, the corresponding first power level balancing mode or second power level balancing mode will not be triggered temporarily. This means that the balancing operation will only be started after the imbalance state has been maintained stably for a sufficiently long time and it is confirmed that it is not a short-term disturbance, thus ensuring the accuracy and necessity of the balancing action. In addition, a dynamic correction mechanism based on temperature and SOH decay differences is introduced to ensure the safety of lithium battery packs and extend their service life. If the current operating temperature of the lithium battery pack exceeds 80% of the set temperature threshold, it indicates that the lithium battery pack may be in a relatively high-temperature operating environment. Excessive temperature will accelerate battery aging and may bring safety risks. At this time, the transfer current corresponding to each power level is reduced by 20%-50%. By reducing the transfer current, the heat generated during the equalization process can be reduced, and the temperature can be prevented from rising further, thereby protecting the lithium battery pack. In addition, if the difference in SOH decay between the first target cell and the second target cell exceeds 50% of the SOH decay difference threshold, it indicates that there is a significant difference in the health status of the two cells. In order to avoid excessive current impact and additional damage to the cells with poor health, the adjustment rate of the transfer current is reduced by 10%-30%. The adjustment rate refers to the speed at which the magnitude of the transfer current changes. Reducing the adjustment rate makes the current change more gradual, reduces the stress on the internal structure of the battery, helps protect the battery with more severe decay, and slows down its performance degradation.

[0041] The beneficial effects of the above technical solution are as follows: By using the solution provided in this embodiment, the energy inconsistency comprehensive value is calculated through multi-parameter fusion for graded balancing, and the duration is used to prevent false triggering. At the same time, the current magnitude and adjustment rate are dynamically adjusted according to the temperature and battery health differences, so as to maximize the safety and service life of the lithium battery pack while ensuring the balancing effect.

[0042] In one embodiment, the overall energy inconsistency value is calculated using the following formula: , in, Represents the overall value of energy inconsistency. Represents the SOC difference. Represents the SOC difference threshold. This represents the difference in the degree of SOH decay. The threshold representing the difference in the degree of SOH decay. This represents the current operating temperature of the lithium battery pack. Represents the temperature threshold. , , These are the weighting coefficients for the SOC difference, the SOH degradation difference, and the current operating temperature of the lithium battery pack, respectively. The SOC difference threshold is the maximum allowable difference in SOC between individual cells; the SOH decay difference threshold is the maximum allowable difference in SOH decay between individual cells; and the temperature threshold is the highest allowable temperature for normal operation of the lithium battery pack.

[0043] The working principle of the above technical solution is as follows: By constructing a calculation model for the comprehensive value R of energy inconsistency, a quantitative assessment of the energy state inconsistency of lithium battery packs is achieved. Its core is: First, to clarify the three key dimensions affecting the energy consistency of lithium battery packs, namely the SOC (State of Charge) difference between individual cells, the difference in SOH (State of Health) degradation, and the current operating temperature of the lithium battery pack. For the SOC dimension, the ratio of the SOC difference to the SOC difference threshold is used to characterize the degree of actual SOC difference relative to the maximum allowable difference; the larger this ratio, the more severe the SOC inconsistency. For the SOH dimension, similarly, the ratio of the SOH degradation difference to the SOH degradation difference threshold is used to measure the proportion of SOH degradation differences between individual cells exceeding the allowable range; the higher this ratio, the more prominent the inconsistency at the SOH level. For the temperature dimension, the ratio of the current operating temperature of the lithium battery pack to the temperature threshold is directly used. This ratio reflects the degree to which the current operating temperature of the lithium battery pack approaches or exceeds the maximum allowable operating temperature. Excessive temperature itself will exacerbate the inconsistency of battery performance and affect the overall energy state. Next, weighting coefficients are assigned to the three evaluation indicators, and the sum of the three is 1, to reflect the relative importance of each factor in the comprehensive inconsistency assessment. Finally, the three weighted ratios are added together to obtain the comprehensive energy inconsistency value R. The magnitude of the R value comprehensively reflects the overall energy inconsistency of the lithium battery pack in the three key aspects of SOC, SOH, and temperature. The larger the R value, the worse the energy state consistency of the lithium battery pack, which may require equalization processing or other maintenance measures to ensure its safe and stable operation and effective utilization. Through this formula, multiple inconsistency indicators with different dimensions and physical meanings can be integrated into a unified quantitative indicator, providing a scientific and intuitive basis for the battery management system to judge the health status of the lithium battery pack and formulate equalization strategies.

[0044] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, the energy inconsistency of the lithium battery pack can be comprehensively and quantitatively evaluated from multiple key dimensions, overcoming the limitations of traditional single-parameter evaluation. When the R value obtained reaches or exceeds the preset threshold, the equalization control strategy can be triggered in time, thereby effectively avoiding problems such as overcharging, over-discharging or excessive degradation of some batteries due to uneven energy distribution. This helps to extend the cycle life of the entire lithium battery pack, improve the safety and stability of the battery system operation, and reduce maintenance costs.

[0045] In one embodiment, it also includes: When charging the lithium battery pack, the parameters of the charging curve of the lithium battery pack are adaptively adjusted according to the average SOH and resting time of the lithium battery pack. When the average SOH is higher than the first SOH threshold, standard charging parameters are used; when the average SOH is between the first SOH threshold and the second SOH threshold, the current value of the constant current charging stage is reduced, and the voltage value of the constant voltage charging stage is reduced by a small amount; when the average SOH is lower than the second SOH threshold, the current value of the constant current charging stage is further reduced, and the voltage value of the constant voltage charging stage is reduced by a larger amount, and the transition threshold of the constant voltage stage is triggered in advance; wherein, the parameters of the charging curve include the current value of the constant current charging stage, the voltage value of the constant voltage charging stage, and the transition threshold conditions between the constant current charging stage and the constant voltage charging stage, and the first SOH threshold is greater than the second SOH threshold.

[0046] The working principle of the above technical solution is as follows: by adaptively adjusting the charging curve parameters according to the state of health (SOH) of the lithium battery pack, fine-grained charging management of batteries with different health levels can be achieved. Specifically: First, the average SOH value of the lithium battery pack is obtained as the basis for adjusting the charging strategy. The average SOH reflects the overall health level of the entire lithium battery pack and is a key indicator for judging the degree of battery aging. When the average SOH of the lithium battery pack is higher than the first SOH threshold, it indicates that the lithium battery pack is in good health and the degree of aging is relatively mild. At this time, standard charging parameters are used for charging. Standard charging parameters are usually determined during the battery design stage and can ensure charging efficiency while minimizing damage to healthy batteries. When the average SOH is between the first SOH threshold and the second SOH threshold, it means that the lithium battery pack has shown a certain degree of aging. At this time, in order to reduce the stress on the aging battery and extend its service life, the system will reduce the current value of the constant current charging stage. The lower charging current can reduce the chemical reaction rate and heat generation inside the battery. At the same time, the voltage value of the constant voltage charging stage is reduced by a small margin. Appropriately reducing the constant voltage can prevent the battery from aggravating side reactions due to excessive voltage when fully charged, thereby slowing down the capacity decay of the battery. When the average SOH is below the second SOH threshold, the lithium battery pack is severely aged and its performance deteriorates significantly. At this point, a more conservative charging strategy is needed: continue to reduce the current value in the constant current charging stage, further reducing the current compared to the previous stage to minimize the internal stress of the battery during charging; and significantly reduce the voltage value in the constant voltage charging stage. A larger voltage reduction can more effectively protect the severely aged battery and prevent overcharging and excessive chemical loss; in addition, the transition threshold of the constant voltage stage will be triggered earlier. This means that the battery will switch from the constant current stage to the constant voltage stage when it reaches a relatively low charge or voltage, reducing the duration of high current charging and further reducing the battery loss under high current, thereby extending the cycle life and safe use time of the severely aged battery as much as possible. The first SOH threshold is greater than the second SOH threshold, forming a health status judgment gradient from high to low. This corresponds to different levels of charging parameter adjustment strategies. The entire adaptive adjustment process dynamically changes the constant current, constant voltage, and the constant current to constant voltage conversion point in the charging curve, thus realizing personalized and protective charging of lithium battery packs under different SOH states.

[0047] The beneficial effects of the above technical solution are as follows: Using the solution provided in this embodiment, charging parameters can be dynamically adjusted according to the overall health status of the lithium battery pack, achieving precise matching between the charging strategy and the degree of battery aging. For lithium battery packs in good health, standard charging parameters can ensure charging efficiency. For moderately aged lithium battery packs, by reducing the constant current and slightly reducing the constant voltage, battery loss is reduced while maintaining charging speed. For severely aged lithium battery packs, further reducing the current, significantly reducing the voltage, and switching to the constant voltage stage earlier can minimize internal stress and chemical loss during charging, effectively slowing down the rate of battery performance degradation, significantly extending the cycle life of the lithium battery pack, and reducing safety risks caused by improper charging of aged batteries, thus improving the safety and stability of the lithium battery pack during use.

[0048] Management systems for implementing intelligent balancing and lifespan extension management methods for lithium batteries, such as Figure 3 As shown, it includes a converter array, a main controller, a sampling module, and a protection module; The converter array consists of multiple DC / DC bidirectional conversion circuits, each corresponding to at least one individual battery cell, and is switchably connected to each individual battery cell via a matrix switching network. The DC / DC bidirectional conversion circuit adopts a bidirectional synchronous Buck-Boost topology, including power switches, energy storage components, filter capacitors, and freewheeling structures, and is configured to support switching between Buck mode, Boost mode, and synchronous rectification mode. The matrix switching network is configured to allow the positive and negative terminals of each individual battery cell to be selectively connected to the input or output terminal of any DC / DC bidirectional conversion circuit. The main controller is communicatively connected to the converter array and is configured to: acquire the SOC and SOH of each individual cell, determine the energy inconsistency and energy transfer strategy based on the SOC and SOH, control the converter array to establish an energy transfer path from the first target cell to the second target cell, and control the direction, magnitude and duration of the transfer current flowing through the path. The sampling module is configured to acquire the input / output voltage and transfer current of the DC / DC bidirectional converter circuit, with the sampling accuracy meeting the conditions that the transfer current is ≤ ±1% FS and the input / output voltage is ≤ ±0.5% FS. The protection module is configured to provide overcurrent protection, overvoltage protection, and overheat protection.

[0049] The working principle of the above technical solution is as follows: After the system is powered on and started, the sampling module first collects key parameters such as voltage and current of all individual cells in the lithium battery pack in real time. The sampling accuracy of the transfer current reaches ≤±1%FS, and the sampling accuracy of the input / output voltage reaches ≤±0.5%FS, ensuring the accuracy and reliability of the collected data and providing accurate basis for the decision of the main controller. The main controller maintains a communication connection with the sampling module and the converter array. It receives the voltage and current data of individual cells transmitted by the sampling module and calculates the SOC and SOH of each individual cell based on this data. Then, the main controller determines the energy inconsistency between individual cells in the lithium battery pack based on the acquired SOC and SOH data, that is, it determines which individual cells have too high or too low charge, and which individual cells are in good or poor health. In this way, it comprehensively formulates a reasonable energy transfer strategy and identifies the first target individual cell and the second target individual cell that need to be transferred. The converter array consists of multiple DC / DC bidirectional conversion circuits, each corresponding to at least one individual battery cell. A matrix switching network enables switchable connections between the individual battery cells and the matrix switching network. The core function of the matrix switching network is to precisely control the positive and negative terminals of each individual battery cell to selectively connect to the input or output of any DC / DC bidirectional conversion circuit, according to the instructions of the main controller. This allows for flexible construction of energy transfer paths. Once the main controller determines the energy transfer strategy, it sends control signals to the matrix switching network and the corresponding DC / DC bidirectional conversion circuits, controlling the matrix switching network to switch between the first target individual battery cell and the input of a DC / DC bidirectional conversion circuit. Simultaneously, it connects the second target individual battery cell to the output of that circuit (or adjusts the input and output directions according to the energy flow direction), thus establishing an energy transfer path from the first target individual battery cell to the second target individual battery cell. The DC / DC bidirectional converter circuit adopts a bidirectional synchronous Buck-Boost topology. This topology includes power switches, energy storage components (usually inductors), filter capacitors, and a freewheeling structure (implemented by synchronous rectifier switches in synchronous topologies). This topology allows the DC / DC bidirectional converter circuit to flexibly support Buck mode (step-down mode, where the higher input voltage is converted to a lower voltage for the second target cell when the input voltage is higher than the output voltage), Boost mode (step-up mode, where the lower input voltage is converted to a higher voltage for the second target cell when the input voltage is lower than the output voltage), and synchronous rectification mode (when energy flows in reverse or under specific conditions). In the switching state, the switching is achieved by reducing conduction losses and improving conversion efficiency through synchronous switching transistors; the operating mode switching logic of the DC / DC bidirectional converter circuit is as follows: Energy output mode: the topology operates in Buck mode, and the input voltage is stepped down by an inductor before output; Energy receiving mode: the topology operates in Boost mode, and the input voltage is stepped up by an inductor before output; Synchronous rectification mode: triggered when the difference between the input voltage and the output voltage is ≤0.2V, reducing conversion losses; Mode switching conditions: the energy output terminal is defined as the input terminal, and the energy inflow terminal is defined as the output terminal. When the input voltage is ≥ the output voltage +0.2V, it switches to Buck mode; when the input voltage is ≤ the output voltage -0.2V, it switches to Boost mode. The main controller precisely controls the direction (determines whether energy flows from the first target cell to the second target cell or vice versa, although balancing is usually from high to low), magnitude (controls the rate of energy transfer), and duration (controls the total amount of energy transfer) of the transfer current flowing through the established energy transfer path by controlling the turn-on and turn-off timing and duty cycle of the power switching transistors in the DC / DC bidirectional converter circuit. This ensures that energy is efficiently and stably transferred from the first target cell to the second target cell, achieving energy balance among the individual cells in the lithium battery pack. Throughout the energy balance management process, the protection module continuously monitors the system's operating status, including the input and output voltages of each DC / DC bidirectional converter circuit, the current flowing through them, and the temperature of the converter array and lithium battery pack. Once an abnormality such as overcurrent (current exceeding the safety threshold), overvoltage (voltage exceeding the safety threshold), or overheating (temperature exceeding the safety threshold) is detected in the DC / DC bidirectional converter circuit, the protection module will immediately trigger the protection mechanism, such as cutting off the corresponding power path, shutting down the DC / DC converter circuit, or issuing an alarm signal, to prevent the fault from escalating and to protect the safe and reliable operation of the lithium battery pack and the management system itself, avoiding battery damage or safety accidents caused by abnormal conditions.

[0050] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, through the collaborative work of each module, the management system can realize intelligent equalization management of the lithium battery pack, effectively improve the inconsistency of SOC and SOH between individual cells, thereby extending the service life of the entire lithium battery pack.

[0051] In one embodiment, the main controller is further configured to dynamically generate charge / discharge power boundary parameters based on the heterogeneous characteristics of each individual battery cell, so as to achieve synergistic optimization of charging strategy, balancing strategy and thermal management strategy; the heterogeneous characteristics include differentiated rated capacity, diverse aging degree and potential performance dispersion characteristics; the charge / discharge power boundary parameters include charge / discharge rate limit, power threshold, SOC window boundary and temperature correction factor; the diverse aging degree includes cycle aging, calendar aging and storage aging; and the potential performance dispersion characteristics include capacity decay rate, internal resistance growth characteristics, charge / discharge efficiency and temperature sensitivity differences.

[0052] The working principle of the above technical solution is as follows: The main controller acquires the heterogeneous characteristic parameters of each individual cell in the lithium battery pack in real time or periodically to achieve fine management of the charging and discharging process. These heterogeneous characteristics specifically cover three main aspects: First, differentiated rated capacity, that is, the inherent capacity difference of different individual cells at the time of manufacture; second, diverse aging degree, which includes cycle aging caused by the accumulation of charge and discharge cycles, calendar aging that will occur over time even if not used, and storage aging caused under specific storage conditions. These aging factors work together to cause different degrees of performance degradation of individual cells; third, potential performance dispersion characteristics, specifically manifested as inconsistent capacity decay rates, differences in internal resistance growth characteristics, different charging and discharging efficiencies, and differences in temperature sensitivity. Based on the accurate identification and quantitative analysis of the heterogeneous characteristics of these individual cells, the main controller can dynamically generate charge and discharge power boundary parameters for each individual cell. These key boundary parameters mainly include: the upper limit of charge and discharge rate, which determines the ratio of the maximum current that the individual cell can withstand during charging and discharging to the battery capacity; the power threshold, which clarifies the maximum power limit of the individual cell during charging and discharging; the SOC window boundary, which defines the upper and lower limits of the SOC of the individual cell during normal operation; and the temperature correction factor, which is used to dynamically adjust the above parameters according to the real-time temperature to adapt to the changes in battery performance under different temperature conditions. By generating such personalized charge and discharge power boundary parameters, the main controller can effectively optimize the charging strategy, balancing strategy, and thermal management strategy in a coordinated manner. Regarding the charging strategy, based on the boundary parameters of each individual battery cell, overcharging of severely aged or low-capacity cells can be avoided, while fully utilizing the charging capacity of high-performance cells to improve overall charging efficiency and safety. At the balancing strategy level, based on the characteristic differences and dynamic boundaries between individual batteries, the timing and extent of balancing can be determined more accurately. Through active or passive balancing, the SOC and voltage of each individual battery cell are kept within a reasonable range, reducing overall performance degradation caused by imbalance. The thermal management strategy, combined with temperature correction factors and the temperature sensitivity differences of each individual battery cell, provides targeted heat dissipation or insulation control for abnormally hot or temperature-sensitive cells, preventing localized overheating or overcooling from negatively impacting battery performance and lifespan. The synergistic effect of these three strategies ultimately maximizes the overall performance of the lithium battery pack while ensuring its safety and extending its lifespan.

[0053] The beneficial effects of the above technical solution are as follows: by adopting the solution provided in this embodiment, the individual differences of individual cells in the lithium battery pack can be fully considered, and the deep coupling and synergistic optimization of charging, balancing and thermal management strategies can be achieved by dynamically generating charge and discharge power boundary parameters.

[0054] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for intelligent balancing and lifespan extension management of lithium batteries, characterized in that, include: Obtain the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack. The energy inconsistency of the lithium battery pack is determined based on the SOC and SOH of each individual cell. When the energy inconsistency meets the preset equalization trigger condition, based on the real-time difference between the SOC and SOH of each individual cell, at least one first target cell as the first target cell for energy transfer and at least one second target cell as the second target cell for energy reception are dynamically selected, and the battery energy is transferred from the first target cell to the second target cell by controlling the pre-configured DC / DC bidirectional conversion circuit. The DC / DC bidirectional converter circuit is connected between the first target cell and the second target cell and is configured to support bidirectional energy transfer. The direction, magnitude and duration of the transfer current during the energy transfer process are adjusted in real time according to the SOC difference between the first target cell and the second target cell, the SOH decay degree and the current operating temperature of the lithium battery pack.

2. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 1, characterized in that, The equalization trigger condition is: the duration of energy inconsistency is greater than or equal to the preset trigger delay, and any of the following sub-conditions are met: (1) The energy inconsistency is greater than or equal to the preset first energy inconsistency threshold; (2) The SOC of a single cell exceeds the extreme range, which is SOC ± 8% of the average SOC of the lithium battery pack; (3) The energy inconsistency is greater than or equal to the second energy inconsistency threshold and less than the first energy inconsistency threshold, and the current working state of the lithium battery pack is either static or constant current charging and discharging. The second energy inconsistency threshold is 70% of the first energy inconsistency threshold; (4) The time interval between two consecutive equalization triggers is greater than or equal to the set minimum equalization trigger period.

3. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 1, characterized in that, Obtain the state of charge (SOC) and state of health (SOH) of each individual cell in the lithium battery pack, including: A closed-loop online estimation of the State of Charge (SOC) of each individual cell is performed using a battery model-based filtering estimation algorithm. The battery model is a second-order RC equivalent circuit model, which includes a series resistor to simulate the ohmic internal resistance of the individual cell, two parallel RC networks to simulate the electrochemical polarization and concentration polarization of the individual cell, and an ideal voltage source to simulate the open-circuit voltage of the individual cell. The filtering estimation algorithm is an adaptive extended Kalman filter, which can dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix according to the real-time operating temperature and aging degree of the individual cell. A bidirectional long short-term memory network (BiLSTM) based on an attention mechanism is used to comprehensively analyze the impulse response characteristics, cycle history data, and online core capacity test data of a single cell to obtain the state of health (SOH). The impulse response characteristics include ohmic internal resistance, polarization internal resistance, and diffusion capacitance, while the cycle history data includes the number of cycles, cumulative charge-discharge capacity, and deep discharge count.

4. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 3, characterized in that, The pulse response characteristic parameters are obtained by applying a specific pulse excitation signal to a single cell under preset operating conditions, and by fitting the voltage response curves collected during pulse discharge and rest periods using the least squares method. The preset operating conditions are: under a constant temperature environment of 25℃±2℃, when the SOC of the single cell is 20%, 50%, and 80%, respectively. The specific pulse excitation signal is: constant current discharge at a 1C rate for 10 seconds, followed by rest for 30 seconds.

5. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 1, characterized in that, Based on the real-time differences in SOC and SOH of each individual cell, at least one target cell for energy transfer and at least one target cell for energy reception are dynamically selected, including: Individual cells with a SOC value higher than a first threshold and a SOH value higher than a second threshold are identified as the first target individual cells. The first threshold is set to 105% of the average SOC value of the lithium battery pack, and the second threshold is set to 80% of the initial SOH value of the lithium battery pack. Individual cells with a SOC value below the third threshold are identified as the second target individual cells. The third threshold is set at 95% of the average SOC value of the lithium battery pack.

6. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 1, characterized in that, The direction, magnitude, and duration of the transfer current can be adjusted in real time based on the SOC difference between the first and second target cells, the degree of SOH decay, and the current operating temperature of the lithium battery pack, including: Based on the magnitude of the comprehensive value of energy inconsistency calculated from the SOC difference, SOH decay difference, and the current operating temperature of the lithium battery pack, at least two different power level equalization modes are triggered in stages to achieve different methods of transfer current regulation, wherein the higher power level corresponds to a larger transfer current. When the comprehensive value of energy inconsistency is in the first interval, the first power level balancing mode is triggered, and the transfer current is I1; when the comprehensive value of energy inconsistency is in the second interval, the second power level balancing mode is triggered, and the transfer current is I2; where I2 > I1, and the upper limit of the first interval is less than the lower limit of the second interval. If the duration of the comprehensive energy inconsistency value in the first interval does not reach the set first minimum holding time, or the duration of the comprehensive energy inconsistency value in the second interval does not reach the set second minimum holding time, then the first power level equalization mode will not be triggered or the second power level equalization mode will be triggered. If the current operating temperature of the lithium battery pack exceeds 80% of the temperature threshold, the transfer current corresponding to each power level will be reduced by 20%-50%; if the difference in SOH decay between the first target cell and the second target cell exceeds 50% of the SOH decay difference threshold, the adjustment rate of the transfer current will be reduced by 10%-30%.

7. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 6, characterized in that, The overall value of energy inconsistency is calculated using the following formula: ， in, Represents the overall value of energy inconsistency. Represents the SOC difference. Represents the SOC difference threshold. This represents the difference in the degree of SOH decay. The threshold representing the difference in SOH decay level This represents the current operating temperature of the lithium battery pack. Represents the temperature threshold. , , These are the weighting coefficients for the SOC difference, the SOH degradation difference, and the current operating temperature of the lithium battery pack, respectively. The SOC difference threshold is the maximum allowable difference in SOC between individual cells; the SOH decay difference threshold is the maximum allowable difference in SOH decay between individual cells; and the temperature threshold is the highest allowable temperature for normal operation of the lithium battery pack.

8. The intelligent balancing and lifespan extension management method for lithium batteries according to claim 1, characterized in that, Also includes: When charging the lithium battery pack, the parameters of the charging curve of the lithium battery pack are adaptively adjusted according to the average SOH and resting time of the lithium battery pack. When the average SOH is higher than the first SOH threshold, standard charging parameters are used; when the average SOH is between the first SOH threshold and the second SOH threshold, the current value of the constant current charging stage is reduced, and the voltage value of the constant voltage charging stage is reduced by a small amount; when the average SOH is lower than the second SOH threshold, the current value of the constant current charging stage is further reduced, and the voltage value of the constant voltage charging stage is reduced by a larger amount, and the transition threshold of the constant voltage stage is triggered in advance; wherein, the parameters of the charging curve include the current value of the constant current charging stage, the voltage value of the constant voltage charging stage, and the transition threshold conditions between the constant current charging stage and the constant voltage charging stage, and the first SOH threshold is greater than the second SOH threshold.

9. A management system for implementing the intelligent balancing and lifespan extension management method for lithium batteries as described in any one of claims 1 to 8, characterized in that, Includes converter array, main controller, sampling module and protection module; The converter array consists of multiple DC / DC bidirectional conversion circuits, each corresponding to at least one individual battery cell, and is switchably connected to each individual battery cell via a matrix switching network. The DC / DC bidirectional conversion circuit adopts a bidirectional synchronous Buck-Boost topology, including power switches, energy storage components, filter capacitors, and freewheeling structures, and is configured to support switching between Buck mode, Boost mode, and synchronous rectification mode. The matrix switching network is configured to allow the positive and negative terminals of each individual battery cell to be selectively connected to the input or output terminal of any DC / DC bidirectional conversion circuit. The main controller is communicatively connected to the converter array and is configured to: acquire the SOC and SOH of each individual cell, determine the energy inconsistency and energy transfer strategy based on the SOC and SOH, control the converter array to establish an energy transfer path from the first target cell to the second target cell, and control the direction, magnitude and duration of the transfer current flowing through the path. The sampling module is configured to acquire the input / output voltage and transfer current of the DC / DC bidirectional converter circuit, with the sampling accuracy meeting the conditions of transfer current ≤ ±1%FS and input / output voltage ≤ ±0.5%FS; The protection module is configured to provide overcurrent protection, overvoltage protection, and overheat protection.

10. The management system according to claim 9, characterized in that, The main controller is also configured to dynamically generate charge and discharge power boundary parameters based on the heterogeneous characteristics of each individual battery cell, so as to achieve synergistic optimization of charging strategy, equalization strategy and thermal management strategy. Heterogeneous characteristics include differentiated rated capacity, diverse aging degree and potential performance dispersion characteristics. Charge and discharge power boundary parameters include charge and discharge rate limit, power threshold, SOC window boundary and temperature correction factor. Diverse aging degree includes cycle aging, calendar aging and storage aging. Potential performance dispersion characteristics include capacity decay rate, internal resistance growth characteristics, charge and discharge efficiency and temperature sensitivity differences.

Images