Battery equalization control method and device, electronic equipment and storage medium

Abstract

The invention provides a battery equalization control method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a voltage value corresponding to each battery in a battery pack, and calculating a voltage difference value between the battery with the maximum voltage and the battery with the minimum voltage in the battery pack; performing filtering processing on the voltage difference value to obtain a filtered voltage difference value; taking the filtered voltage difference value as the input of a predetermined single neuron control algorithm model, performing nonlinear processing on the filtered voltage difference value by using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model, and outputting a target voltage difference state value; and calculating equalization time and a duty ratio according to the target voltage difference state value, and based on the equalization time and the duty ratio, controlling an equalization switch in the battery management system to carry out voltage equalization until the cell voltage of the battery reaches the target voltage. The battery equalization efficiency and equalization effect can be improved while the cost is reduced, so that the battery consistency is improved, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present application relates to the field of new energy vehicle technology, and in particular to a battery balancing control method, device, electronic device and storage medium. Background Art

[0002] Battery consistency refers to the uniformity of performance parameters such as voltage and capacity of each single cell in a battery pack, which has an important impact on the battery's service capacity and life. During the charging and discharging process of lithium batteries, if no defensive measures are taken for batteries with lower voltage and larger overall deviation, the performance of these batteries will gradually deviate, eventually leading to a decline in the performance of the entire battery pack. For example, batteries with lower voltages will reach the discharge cut-off voltage first, and will not be fully charged during charging, thus affecting the performance and life of the overall system. This problem of poor battery consistency will cause the range of electric vehicles operating on the market to rapidly decay, which will have a serious adverse impact on battery manufacturers and the development of the new energy industry. Therefore, the balance control strategy of the battery system is crucial.

[0003] At present, mainstream suppliers in the market generally adopt passive discharge balancing strategies. This strategy is based on the circuit topology structure, and actively discharges the higher voltage batteries through odd-even switches to release the excess electrical energy in the form of heat energy. Although active balancing can achieve energy loss without loss, its overall hardware cost is high, resulting in increased cost of the electronic control system, which does not meet the company's goal of reducing costs and increasing efficiency. Therefore, most battery manufacturers and battery management system (BMS) suppliers choose to adopt passive balancing algorithms. However, the existing passive balancing strategies still need to be optimized in terms of balancing efficiency and balancing effect.

[0004] The existing technology has the following main problems: The existing technology has large energy loss during the discharge process, resulting in a slow balancing process. The balancing target such as voltage or single cell state SOC data is affected by a variety of interference factors, resulting in unstable balancing effect and poor battery consistency. Although active balancing can improve performance, its high hardware cost is not suitable for large-scale applications. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a battery balancing control method, device, electronic device and storage medium to solve the problems of low battery balancing efficiency, poor balancing effect, high battery balancing cost and reduced battery consistency in the prior art.

[0006] According to a first aspect of an embodiment of the present application, a battery balancing control method is provided, comprising: obtaining a voltage value corresponding to each battery in a battery pack, and calculating a voltage difference between a maximum voltage battery and a minimum voltage battery in the battery pack; filtering the voltage difference to obtain a filtered voltage difference; using the filtered voltage difference as an input of a predetermined single neuron control algorithm model, performing nonlinear processing on the filtered voltage difference using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model, and outputting a target voltage difference state value; calculating a balancing time and a duty cycle according to the target voltage difference state value, and based on the balancing time and the duty cycle, controlling a balancing switch in a battery management system to perform voltage balancing until the battery cell voltage reaches the target voltage.

[0007] According to a second aspect of an embodiment of the present application, a battery balancing control device is provided, comprising: an acquisition module, configured to acquire a voltage value corresponding to each battery in a battery pack, and calculate a voltage difference between a maximum voltage battery and a minimum voltage battery in the battery pack; a filtering module, configured to filter the voltage difference to obtain a filtered voltage difference; a processing module, configured to use the filtered voltage difference as an input of a predetermined single neuron control algorithm model, and use a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model to perform nonlinear processing on the filtered voltage difference, and output a target voltage difference state value; a balancing module, configured to calculate a balancing time and a duty cycle according to the target voltage difference state value, and based on the balancing time and the duty cycle, control a balancing switch in a battery management system to perform voltage balancing until the battery cell voltage reaches the target voltage.

[0008] According to a third aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the above method are implemented when the processor executes the computer program.

[0009] According to a fourth aspect of an embodiment of the present application, a readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0010] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0011] By obtaining the voltage value corresponding to each battery in the battery pack, and calculating the voltage difference between the maximum voltage battery and the minimum voltage battery in the battery pack; filtering the voltage difference to obtain the filtered voltage difference; using the filtered voltage difference as the input of the predetermined single neuron control algorithm model, using the neuron self-learning algorithm and vector converter in the single neuron control algorithm model, performing nonlinear processing on the filtered voltage difference, and outputting the target voltage difference state value; calculating the balancing time and duty cycle according to the target voltage difference state value, and based on the balancing time and duty cycle, controlling the balancing switch in the battery management system to perform voltage balancing until the battery cell voltage reaches the target voltage. This application can improve the battery balancing efficiency and balancing effect while reducing costs, thereby improving battery consistency and service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 is a flowchart of a battery balancing control method provided in an embodiment of the present application;

[0014] Figure 2 It is a structural diagram of the extended Kalman filter algorithm model provided in the embodiment of the present application;

[0015] Figure 3 It is a structural schematic diagram of a single neuron control algorithm model provided in an embodiment of the present application;

[0016] Figure 4 It is a schematic diagram of the system structure combining the extended Kalman filter algorithm and the single neuron control algorithm provided in an embodiment of the present application;

[0017] Figure 5 is a structural schematic diagram of a battery balancing control device provided in an embodiment of the present application;

[0018] Figure 6 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0020] Battery consistency refers to the consistency of a group of batteries in terms of performance, including parameters such as voltage and consistency. This has a critical impact on the battery's capacity and life. During the charging and discharging process of lithium batteries, if no defensive measures are taken for batteries with lower voltage and larger overall deviation, the performance of these batteries will gradually become discrete, eventually leading to a decline in the performance of the entire battery group. Specifically, the batteries with lower voltages in the battery pack will reach the discharge cut-off voltage first, and these batteries cannot be fully charged to the cut-off voltage during charging.

[0021] This situation will lead to an overall decline in the performance of the battery system, especially in applications such as new energy vehicles, which will cause the range to decay rapidly, and have an adverse impact on battery manufacturers and the development of the new energy industry. Therefore, the battery system requires an effective balancing control strategy to improve battery consistency, extend battery life and enhance overall performance.

[0022] Currently, the first-class suppliers in the market mainly adopt the passive discharge type balancing strategy. This strategy is based on the circuit topology structure and uses odd-even switches to actively discharge the battery with higher voltage, releasing the excess electrical energy in the form of heat energy. Although active balancing can achieve energy loss, the overall hardware cost is high, resulting in an increase in the cost of the electronic control system, which is not in line with the company's goal of reducing costs and increasing efficiency.

[0023] Due to cost reasons, battery manufacturers and battery management system (BMS) suppliers generally adopt passive balancing algorithm design. However, the existing passive balancing algorithms need to be optimized in terms of balancing efficiency and effect. Specific problems include: balancing targets such as voltage or single cell state SOC (State of Charge) data are affected by a variety of interference factors. The target data needs to be filtered to reduce interference and ensure a more ideal balancing effect and better battery consistency.

[0024] Therefore, under the existing passive balancing strategy, the battery balancing efficiency and effect are not ideal, and the existing battery balancing strategy still has the following problems:

[0025] 1) Low balancing efficiency: The existing technology has large energy loss during the discharge process, resulting in a slow balancing process and low efficiency.

[0026] 2) Poor balancing effect: Since the balancing target data (such as voltage and SOC) is disturbed, the balancing effect is unstable and the battery consistency is poor.

[0027] 3) High balancing cost: Although active balancing can improve performance, its hardware cost is high and is not suitable for large-scale applications.

[0028] Therefore, an improved battery balancing strategy is needed that can reduce costs while improving balancing efficiency and effect to ensure the consistency and life of the battery pack.

[0029] In view of the deficiencies in the prior art, this application proposes a balancing control strategy based on a neuron adaptive intelligent filtering algorithm to solve the problems of low balancing efficiency, poor balancing effect and high cost in the prior art. The technical solution of this application mainly adopts a single neuron PID strategy (single neuron control algorithm), combined with an extended Kalman filter algorithm (EKF) to denoise the voltage target value, and observe, predict and correct according to parameter inputs such as temperature and current, so as to obtain an accurate balancing target voltage value in real time, thereby facilitating the calculation of the balancing time and the closing control of the balancing switch, and ultimately making the entire system more consistent.

[0030] The contents of the technical solution of the present application are described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] Figure 1 It is a flowchart of a battery balancing control method provided in an embodiment of the present application. Figure 1 The battery balancing control method can be executed by a battery management system. Figure 1 As shown, the battery balancing control method may specifically include:

[0032] S101, obtaining the voltage value corresponding to each battery in the battery pack, and calculating the voltage difference between the battery with the maximum voltage and the battery with the minimum voltage in the battery pack;

[0033] S102, filtering the voltage difference to obtain a filtered voltage difference;

[0034] S103, using the filtered voltage difference as an input of a predetermined single neuron control algorithm model, performing nonlinear processing on the filtered voltage difference using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model, and outputting a target voltage difference state value;

[0035] S104, calculating the balancing time and the duty cycle according to the target voltage difference state value, and based on the balancing time and the duty cycle, controlling the balancing switch in the battery management system to perform voltage balancing until the battery cell voltage reaches the target voltage.

[0036] First, the embodiment of the present application uses a battery management system to monitor the voltage value of each battery in the vehicle battery pack, and calculates the voltage difference between the maximum voltage battery and the minimum voltage battery in the battery pack according to the voltage value of each battery obtained in real time. In practical applications, the voltage difference can be calculated using the following formula: V(k) = V max -V min , where V(k) represents the voltage difference, V max Indicates the maximum voltage value in the battery pack, V min Indicates the minimum voltage value in the battery pack.

[0037] In some embodiments, before filtering the voltage difference, the method further includes:

[0038] A functional relationship between the voltage difference in each stage and the preset parameters is established, and a filtering algorithm model is constructed using the functional relationship. Independent system noise and measurement noise are introduced into the filtering algorithm model, wherein the system noise and measurement noise are both Gaussian white noise and have corresponding covariance matrices.

[0039] Specifically, the calculated pressure difference value V(k) is input into the extended Kalman filter algorithm structure. The extended Kalman filter algorithm can handle nonlinear systems and is particularly suitable for situations where the battery pressure difference is associated with multiple parameters such as temperature, charge and discharge power, and battery aging status.

[0040] First, a functional relationship between the voltage difference and preset parameters in each stage (including charging, discharging, and resting) is established. The preset parameters include temperature, charge and discharge current, and battery aging status. These parameters will affect the performance and voltage difference of the battery, and the functional relationship can reflect the law of the voltage difference changing with these parameters.

[0041] Furthermore, the establishment of the functional relationship may include the following steps:

[0042] Data collection and analysis: Collect voltage data from multiple battery samples at different temperatures, charge and discharge currents, and different aging conditions. By analyzing this data, determine the correlation between the voltage difference and each preset parameter.

[0043] Functional relationship modeling: Based on the collected data, a functional relationship model between the voltage difference and temperature, charge and discharge current, and battery aging status is established using statistical analysis methods or machine learning algorithms. This model is used to predict the voltage difference under specific parameter conditions.

[0044] Model verification and optimization: Verify the accuracy of the model through experiments, and adjust the model parameters if necessary to improve the prediction accuracy. The final functional relationship model will be used as the input of the extended Kalman filter algorithm.

[0045] Furthermore, after constructing the functional relationship model, the extended Kalman filter algorithm (EKF) is constructed using the model. The extended Kalman filter algorithm can handle nonlinear systems and is particularly suitable for situations where the battery pressure difference is associated with multiple parameters such as temperature, charge and discharge power, and battery aging status.

[0046] Furthermore, independent system noise and measurement noise are introduced when constructing the EKF model. Both system noise and measurement noise are Gaussian white noise, with covariances of Q and R respectively. These noises are used to simulate the uncertainty and measurement errors that may exist in the actual system, thereby improving the robustness and accuracy of the filtering algorithm.

[0047] Furthermore, the calculated voltage difference is input into the extended Kalman filter algorithm structure. The EKF algorithm is processed by the following steps:

[0048] Prediction step: Based on the current state estimate and voltage difference, predict the state at the next moment.

[0049] Update step: Combine actual observation data to update state estimates and correct prediction errors.

[0050] Iterative process: The prediction and update steps are repeated until the state estimate converges or the preset number of iterations is reached.

[0051] It should be noted that in the technical solution of this application, "state estimation" refers to the prediction and calculation of the internal state of the battery system (such as battery voltage, SOC (battery state of charge), battery temperature and other parameters) through an algorithm. Since some state parameters are difficult to measure directly or obtain in real time in actual operation, the extended Kalman filter algorithm recursively estimates these states by combining existing observation data and system models to obtain a more accurate internal state of the system.

[0052] Specifically, the process of state estimation can include two key steps: prediction and update.

[0053] Among them, prediction refers to the calculation of the system state at the next moment through the model based on the existing state information (such as previous voltage, temperature, etc.), which is the prediction of the future state. Update refers to the correction of the predicted value in combination with the actual measured observation data to make the estimated value closer to the actual state of the system.

[0054] In battery management systems, state estimation is usually used to predict key parameters of the battery, such as voltage changes, capacity differences, and temperature. These estimation results play an important role in subsequent battery balancing control, energy management, and other operations.

[0055] The structure and construction process of the extended Kalman filter algorithm model provided in the embodiment of the present application are described in detail below in conjunction with the accompanying drawings and specific embodiments. Figure 2 Schematic diagram of the structure of the extended Kalman filter algorithm model provided in the embodiment of the present application. Figure 2 As shown, the structure and construction process of the extended Kalman filter algorithm model in the embodiment of the present application may include the following contents:

[0056] The extended Kalman filter algorithm can handle nonlinear systems, especially the voltage difference of the battery PACK monomer is related to multiple parameters such as temperature, charge and discharge power, battery aging status, etc., so the extended Kalman filter algorithm can improve the accuracy of target value processing. The extended Kalman filter algorithm is a recursive operation of a series of mathematical formulas. First, the functional relationship of the voltage difference in each stage and each temperature range is established, and then the established functional relationship is used as the input of the extended Kalman filter algorithm structure. Independent system noise and measurement noise will be introduced when establishing the model. These two noises are Gaussian white noise, and the corresponding covariances are Q and R respectively. From the perspective of the overall model algorithm structure, this filtering algorithm is suitable for working conditions where the overall use environment changes rapidly. The specific establishment steps and equation calculation process are as follows:

[0057] First, establish the functional relationship between the voltage difference and temperature, the current during vehicle charging and discharging, and the battery aging state. During the entire life cycle of the battery pack, the voltage performance will be complex and changeable. For example, overcharge and overdischarge, low or high temperature, and inconsistent polarization state of later aging caused by battery production materials. The complex and changeable performance of the single cell voltage will also cause fluctuations in the voltage difference, so the first step formula is as follows:

[0058]

[0059] Among them, i represents the current dependent variable, t represents the temperature dependent variable, and soh represents the battery health dependent variable.

[0060] Next, V(k) is used as the input of the system and introduced into the extended filtering algorithm structure, and finally output t k Represents the final corrected target pressure difference state value, where ω k and μ k That is, they are independent Gaussian white noise. Figure 2 The structure of the extended Kalman filter algorithm model shown can be expressed using the following formula:

[0061]

[0062] Among them, X k is the process state variable, V k is the voltage difference, y kis the final pressure difference value after the extended Kalman filter algorithm, A k , B k , C k , D k are coefficients, w k and μ k is the system white noise introduced when establishing the equation.

[0063] Finally, the system state equation and observation equation are established according to the extended Kalman filter algorithm as shown below:

[0064]

[0065] Among them, q k , Q k 、r k , R k All are noise estimators; according to the above formula, q k , Q k 、r k , R k The system noise estimator, i.e., the state equation, is then iteratively updated according to the covariance matrix until the state estimation equation closest to the actual value is obtained.

[0066] Through the above method of this embodiment, this embodiment can fully consider factors such as temperature, charge and discharge current and battery aging status when processing battery voltage difference, thereby improving the accuracy and effect of battery balancing control. This method can not only effectively filter out noise, but also dynamically adjust the battery balancing strategy to ensure the consistency and service life of the battery pack.

[0067] In some embodiments, filtering the voltage difference includes:

[0068] Initialize the state estimation and error covariance matrix of the filtering algorithm model, and set the process noise covariance matrix and the measurement noise covariance matrix;

[0069] According to the current state estimation and voltage difference, the state estimation and error covariance matrix of the next moment are predicted by using the state equation;

[0070] According to the actual observation value and observation equation, the state estimation and error covariance matrix of the next moment are updated;

[0071] Iterative prediction and updating are performed on the state estimation and error covariance matrix until the state estimation converges or the preset number of iterations is reached to obtain the final state estimation equation.

[0072] Specifically, in the initialization phase, before the filtering process begins, the state estimation and error covariance matrix of the extended Kalman filter algorithm model need to be initialized. The initialization process may include the following:

[0073] State estimation initialization: Initialize the state estimation based on the initial state of the system. The initial state can be determined based on the historical data or experience data of the system.

[0074] Error covariance matrix initialization: Set the initial error covariance matrix to describe the uncertainty of the initial state estimate.

[0075] Process noise covariance matrix setting: Set the process noise covariance matrix to describe the uncertainty of the system in the process.

[0076] Measurement noise covariance matrix setting: Set the measurement noise covariance matrix to describe the errors that may occur during the measurement process.

[0077] Furthermore, in the prediction phase, after initialization is completed, the state estimation and the error covariance matrix at the next moment are predicted using the state equation according to the state estimation and the voltage difference at the current moment. The prediction phase process may include the following:

[0078] State estimation prediction: Based on the current state estimation and the known voltage difference, the state estimation at the next moment is calculated using the state equation. This step predicts the state variables based on the system model.

[0079] Error covariance matrix prediction: Based on the prediction results of state estimation, the error covariance matrix is ​​updated to reflect the uncertainty of the predicted state. This step is also based on the system model and the influence of process noise.

[0080] Further, in the update phase, after the prediction is completed, the predicted state and covariance matrix are updated using the observation equation. The update phase process may include the following:

[0081] Calculate Kalman gain: Calculate Kalman gain based on the error covariance matrix and the measurement noise covariance matrix. Kalman gain is used to weigh the difference between the predicted value and the actual measured value.

[0082] State estimate update: The predicted state estimate is corrected by the observation equation using the actual observations. The state estimate is updated by comparing the actual measurements with the predicted values ​​and taking a weighted average based on the Kalman gain.

[0083] Error covariance matrix update: According to the updated results of the state estimate, the error covariance matrix is ​​adjusted to reflect the uncertainty of the updated state estimate.

[0084] Furthermore, in the iteration phase, the above prediction and update steps are iterated in a loop until convergence or a predetermined number of iterations is reached. The iteration phase process may include the following:

[0085] Iterative prediction and update: Repeat the steps of the prediction and update phases to gradually approach the actual state variables. In each iteration, the state estimate and the error covariance matrix are updated.

[0086] Convergence determination: After each iteration, check whether the state estimate has converged to a stable value. If the state estimate changes very little in several consecutive iterations, or reaches the preset number of iterations, it is considered to have converged.

[0087] It should be noted that in the technical solution of this application, the current state estimation is obtained by the extended Kalman filter (EKF) algorithm. EKF combines the historical state of the system with the current observation data and estimates the current system state through a series of prediction and update steps.

[0088] Specifically, the state estimation process is divided into the following three steps:

[0089] 1. Prediction steps:

[0090] First, the current state is predicted through the system model (i.e., the state equation) based on the state estimate of the previous moment and the input voltage difference, temperature, battery aging state and other influencing parameters. This prediction step does not rely on current observation data, but is derived using historical data and system models.

[0091] In addition to the predicted state itself, an error covariance matrix is ​​calculated based on the process noise of the system to describe the uncertainty of the prediction.

[0092] 2. Update steps:

[0093] After obtaining the predicted state at the current moment, EKF corrects the predicted value through the actual observation data at the current moment (such as battery voltage, temperature, etc.).

[0094] The observation data is linked to the system state through the observation equation. Based on the prediction error and measurement noise, the Kalman gain is calculated. The Kalman gain determines the combination ratio of the predicted value to the actual observed value to optimize the accuracy of the state estimation.

[0095] Finally, the predicted state estimate is combined with the actual observed value using the Kalman gain to obtain the state estimate at the current moment. Through this process, the prediction error can be reduced and the estimated value can be closer to the actual state.

[0096] 3. Loop iteration:

[0097] The estimated state value after each calculation will be used as the input for the next moment, and the above prediction and update steps will be repeated to continuously adjust and correct the state.

[0098] Through the method of the present embodiment described above, the extended Kalman filter algorithm can effectively filter the voltage difference, gradually approach the actual state variable, and improve the accuracy and effect of battery balancing control. This process fully considers the nonlinear characteristics of the system and the influence of measurement noise, thereby achieving accurate management of the battery system.

[0099] In some embodiments, a neuron self-learning algorithm and a vector converter are used in a single neuron control algorithm model to perform nonlinear processing on the filtered voltage difference, including:

[0100] Set the target pressure difference value and the current pressure difference value, and calculate the error between the target pressure difference value and the current pressure difference value;

[0101] Input the error into a vector converter, and use the vector converter to convert the error into a plurality of neuron input components;

[0102] The neuron self-learning algorithm is used to adaptively adjust the weight of each neuron input component according to the current state, and the weight is normalized;

[0103] Calculate the control amount at the current moment, calculate the new control amount based on the control amount at the previous moment and the increment of the control amount at the current moment, and output the new control amount as a control signal.

[0104] Specifically, first, the state variable processed by the extended Kalman filter algorithm is input into a single neuron proportional-integral-differential (PID) controller. This state variable is based on the accurate voltage difference obtained after the extended Kalman filter algorithm is processed.

[0105] Furthermore, a target pressure difference value is set and the current pressure difference value is obtained. By comparing the target pressure difference value and the current pressure difference value, the error between the two is calculated. The error reflects the degree of deviation between the current voltage difference value and the ideal state, and is the basis for subsequent adjustments.

[0106] Furthermore, the calculated error is input into a vector converter. The vector converter converts the error into multiple neuron input components, usually three input components. These components may represent different parts of the error, such as proportional, integral and differential parts, so as to fully reflect the error characteristics.

[0107] Furthermore, the neuron self-learning algorithm is used to adaptively adjust the weight of each neuron input component according to the current state. The weight adjustment is to optimize the output of the controller and make the control effect more accurate. The adjusted weight will be normalized to ensure that the sum of the weights of each component is a constant, thereby maintaining the stability of the system.

[0108] Furthermore, the control quantity increment at the current moment is calculated based on the control quantity at the current moment and the control quantity at the previous moment. Specifically, the single neuron PID controller will calculate a new control quantity based on the current error and the adjusted weight. This control quantity reflects the adjustment force that the current system needs to apply to reduce the error and approach the target pressure difference value.

[0109] Furthermore, the calculated new control amount is output as a control signal. The control signal is used to adjust the equalization process of the battery pack, specifically including adjusting the working state of the equalization resistor, so as to achieve the consistency of the voltage of each single cell of the battery pack.

[0110] For example, within a predetermined period, three pressure difference values ​​are selected as a group of training samples, and a weighted coefficient equation is established using these samples. The weighted coefficient equation is as follows:

[0111]

[0112] Among them, w1(k), w2(k), w3(k) are full value coefficients, K uI , K uP , K uD is the adjustment coefficient of neuron PID, and the input state value is obtained as a relatively stable weighting coefficient, so as to obtain a high-performance and stable pressure difference u(k) target value according to the neuron PID structure. The single neuron PID controller continuously adjusts the weight coefficient according to these pressure difference values ​​through the neuron self-learning algorithm, so that the system outputs a stable control signal and realizes efficient control of the battery pack voltage difference.

[0113] In the above process, the specific steps of weight adjustment and control amount calculation are as follows:

[0114] Weight adjustment: According to the current state, the weight is updated through the self-learning algorithm. The weight of each input component will be corrected according to the current system error and the control amount at the previous moment, and the weight sum will be ensured to be a constant through normalization.

[0115] Control quantity calculation: Combine the current error and the adjusted weight to calculate the control quantity at the current moment. The new control quantity will be used to adjust the balancing resistor to achieve balanced control of the battery pack.

[0116] The structure and working principle of the single neuron control algorithm model provided in the embodiment of the present application are described in detail below in conjunction with the accompanying drawings and specific embodiments. Figure 3 Schematic diagram of the structure of the single neuron control algorithm model provided in the embodiment of the present application. Figure 3 As shown, the structure and working principle of the single neuron control algorithm model of the embodiment of the present application may include the following contents:

[0117] The single neuron PID controller is an algorithm based on the PID feedback architecture and the fusion of the neuron algorithm with self-learning and adaptive capabilities. It can not only realize nonlinear processing and online parameter adjustment, but also has the advantages of simple structure, stability and robustness. The specific neuron PID control structure is as follows Figure 3 shown.

[0118] Neurons can simulate the human brain to perform nonlinear processing on complex real-time data inputs, and the PID structure can realize feedback tracking of changes in specific values. Therefore, the fusion of the two algorithms can help solve engineering science problems and has far-reaching significance. The balancing strategy proposed in this application will process voltage data based on neuron PID and the extended Kalman filter algorithm to make the overall goal more stable, improve balancing efficiency, shorten balancing time, and achieve a higher level of consistency for the entire battery PACK.

[0119] The embodiment of the present application combines the extended Kalman filter algorithm with the neuron PID algorithm. At the system input end of the neuron PID, the voltage difference is combined with the influencing factors for observation and correction, which further improves the accuracy and stability of the target balancing value, and prevents the target battery cell voltage from failing to achieve the balancing effect during the balancing process, or the balancing time is too long, resulting in a low cell voltage, which causes a reduction in available capacity. Figure 4 As shown, Figure 4 This is a schematic diagram of the system structure combining the extended Kalman filter algorithm and the single neuron control algorithm provided in the embodiment of the present application. According to the above progressive relationship, the specific output formula of the neuron is as follows:

[0120]

[0121] Among them, u(k) represents the target pressure difference state value output last, u(k-1) represents the pressure difference value output at the previous moment, and K u represents the gain factor, x i (k) represents the state variable, i.e., the input training sample of the neuron. The three pressure difference values ​​can be combined into a group of samples for training. i (k) represents the weighting coefficient, Represents the average weighting coefficient.

[0122] Through the above-mentioned method of this embodiment, this embodiment can effectively use the single neuron control algorithm to perform nonlinear processing on the voltage difference of the battery pack, improve the accuracy and efficiency of battery balancing control, and ensure the consistency and service life of the battery pack.

[0123] In some embodiments, calculating the balancing time and the duty cycle according to the target pressure difference state value includes:

[0124] According to the preset voltage mapping relationship under different discharge rates or charge rates, the interpolation method is used to find the capacity difference corresponding to the target voltage difference state value; based on the found capacity difference, the balancing time is calculated using the matching balancing resistor, and the duty cycle is determined based on the capacity difference, balancing current, balancing efficiency and balancing time.

[0125] Specifically, the battery management system (BMS) uses interpolation to find the capacity difference corresponding to the target voltage difference state value at the application level. Battery manufacturers have pre-made voltage and current mapping diagrams (maps) at different discharge rates or charge rates, as well as voltage and state of charge (SOC) mapping diagrams.

[0126] Furthermore, in the application layer of the BMS, the interpolation method is used to find the corresponding capacity difference in the preset mapping relationship according to the current target pressure difference state value. The interpolation method can effectively derive the required intermediate value from the discrete data set to ensure the accuracy of the calculation.

[0127] Furthermore, according to the found capacity difference, the balancing time is calculated using the matching balancing resistor. First, a balancing resistor matching the battery pack is selected to ensure that the resistor can effectively consume excess energy during the balancing process.

[0128] Furthermore, the required balancing time is calculated based on the capacity difference and the balancing resistance. The calculation of the balancing time takes into account the energy transfer required by the battery pack during the balancing process to ensure that the battery voltage consistency is achieved within a reasonable time range.

[0129] Furthermore, after calculating the balancing time, the duty cycle is further determined. First, the energy conversion efficiency of the battery system during the balancing process is determined based on the capacity difference, balancing current and balancing efficiency. The balancing efficiency reflects the effective utilization of energy in the battery during the discharge or charging process.

[0130] Furthermore, the duty cycle during the balancing process is calculated using the known capacity difference, balancing current, balancing efficiency and balancing time. The duty cycle reflects the proportion of time that the balancing circuit is in the on state in a cycle, thereby ensuring that the balancing resistor works within a reasonable time to achieve battery consistency.

[0131] For example, in one example, a normal battery manufacturer will make a map of voltage and current at different discharge rates or charge rates, as well as a map of voltage and SOC. In actual applications, the BMS application layer uses interpolation to find the capacity difference corresponding to the target voltage difference state value. Then, based on the matching balancing resistor, the required balancing time is calculated. Finally, based on the capacity difference, balancing current, balancing efficiency and balancing time, the duty cycle is calculated. For example, it can be calculated according to the following formula: capacity difference = balancing current * balancing efficiency * duty cycle * balancing time.

[0132] Through the above method of this embodiment, this embodiment can accurately calculate the balancing time and duty cycle according to the target pressure difference state value, ensuring that the battery management system can effectively perform battery balancing control and improve the consistency and service life of the battery pack.

[0133] In some embodiments, based on the balancing time and the duty cycle, controlling the balancing switch in the battery management system to perform voltage balancing includes:

[0134] According to the balancing time and duty cycle, the balancing parameters in the battery management system are set, and the balancing switch is controlled to connect the balancing resistor to start the voltage balancing process;

[0135] Use equalizing resistors to passively discharge batteries that exceed the voltage threshold, and use resistors to dissipate excess battery energy in the form of heat to gradually reduce the voltage;

[0136] The cell voltage of the battery is continuously monitored, and the working state of the balancing switch is adjusted in real time so that the voltage balancing process is carried out according to the preset balancing time and duty cycle.

[0137] Specifically, first, the balancing parameters in the battery management system (BMS) are set according to the calculated balancing time and duty cycle. For example, the pre-calculated balancing time parameters are input into the BMS to ensure that the balancing process can be completed within the set time. The calculated duty cycle parameters are input into the BMS to ensure that the balancing switch is turned on and off according to the set time ratio in each cycle.

[0138] Furthermore, according to the set balancing time and duty cycle, the balancing switch is controlled to connect the balancing resistor to start the voltage balancing process. For example, the balancing MOS switch of the corresponding battery cell is controlled by the BMS to connect the balancing resistor to the battery cell that needs to be balanced. The balancing resistor starts to work and passively discharges the battery that exceeds the voltage threshold, dissipating the excess energy in the form of heat.

[0139] Furthermore, the battery with a higher voltage is passively discharged using an equalizing resistor. For example, the equalizing resistor dissipates the excess energy in the battery in the form of heat by dissipating heat through resistance. The passive discharge process gradually reduces the voltage of the battery until the voltage reaches the target voltage value.

[0140] In this embodiment, the target voltage is the minimum voltage of the battery pack. Balancing mainly uses resistor discharge to dissipate the battery pack with a higher voltage in the form of heat energy, so the target voltage is the minimum voltage of the battery pack.

[0141] Furthermore, during the voltage balancing process, the battery cell voltage is continuously monitored, and the working state of the balancing switch is adjusted in real time. For example, the BMS monitors the voltage of the battery cell in real time to ensure that the voltage change is within the preset range. According to the real-time monitoring data, the on and off states of the balancing switch are dynamically adjusted to ensure that the balancing process is carried out according to the preset balancing time and duty cycle. When the voltage of the battery cell drops to the target voltage value, the BMS turns off the balancing switch and ends the balancing process.

[0142] For example, in the specific implementation process, the BMS system controls the balancing MOS switch to turn on the balancing resistor and start voltage balancing according to the set balancing time and duty cycle. The balancing resistor passively discharges the battery cells with higher voltages, releases the excess energy in the battery through resistance heat dissipation, and gradually reduces the battery voltage. The BMS continuously monitors the voltage of the battery cells and adjusts the working state of the balancing switch in real time according to the monitoring data to ensure the stability and effectiveness of the balancing process. When the battery cell voltage reaches the target voltage value, the BMS turns off the balancing switch to complete the voltage balancing.

[0143] Through the above method of this embodiment, this embodiment can effectively control the balancing switch in the battery management system to perform voltage balancing based on the balancing time and duty cycle, thereby ensuring the consistency and service life of the battery pack.

[0144] According to the technical solution provided in the embodiments of the present application, the technical solution of the present application has at least the following advantages:

[0145] By introducing the extended Kalman filter algorithm model, the voltage difference is accurately filtered, and multiple factors such as temperature, charge and discharge current, and battery aging status are taken into account. System noise and measurement noise are filtered out, and the calculated target balanced voltage difference is more in line with the actual operating status of the battery system, improving the accuracy and reliability of the voltage difference.

[0146] By using the precise target balanced voltage difference, the single neuron control algorithm performs nonlinear processing on the filtered state variables, calculates the balanced time and duty cycle, and controls the balanced switch in the battery management system to perform voltage balanced, ensuring that the voltages of each battery cell in the battery pack are consistent, significantly improving the consistency of the battery system, thereby ensuring the customer's available capacity and extending the battery life.

[0147] This application uses the extended Kalman filter algorithm and the single neuron control algorithm to systematically optimize the battery balancing control process and achieve a high degree of algorithm standardization. Through this standardized algorithm software, the battery management system in the industry is easier to upgrade functions and iterate algorithms, which improves the flexibility and maintainability of the system and helps promote technological progress and development of the entire industry.

[0148] In summary, the technical solution of the present application has significant technical effects in battery balancing control, which not only improves calculation accuracy and system consistency, but also promotes industry standardization and technology upgrades.

[0149] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0150] Figure 5 Schematic diagram of the structure of the battery balancing control device provided in the embodiment of the present application. Figure 5 As shown, the battery balancing control device includes:

[0151] The acquisition module 501 is configured to acquire the voltage value corresponding to each battery in the battery pack and calculate the voltage difference between the battery with the maximum voltage and the battery with the minimum voltage in the battery pack;

[0152] The filtering module 502 is configured to filter the voltage difference to obtain a filtered voltage difference;

[0153] The processing module 503 is configured to use the filtered voltage difference as an input of a predetermined single neuron control algorithm model, perform nonlinear processing on the filtered voltage difference by using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model, and output a target voltage difference state value;

[0154] The balancing module 504 is configured to calculate the balancing time and duty cycle according to the target voltage difference state value, and control the balancing switch in the battery management system to perform voltage balancing based on the balancing time and duty cycle until the battery cell voltage reaches the target voltage.

[0155] In some embodiments, Figure 5 Before filtering the voltage difference, the construction module 505 establishes a functional relationship between the voltage difference in each stage and the preset parameters, uses the functional relationship to build a filtering algorithm model, and introduces independent system noise and measurement noise into the filtering algorithm model, wherein the system noise and the measurement noise are both Gaussian white noise and have corresponding covariance matrices.

[0156] In some embodiments, Figure 5The filtering module 502 initializes the state estimation and error covariance matrix of the filtering algorithm model, and sets the process noise covariance matrix and the measurement noise covariance matrix; predicts the state estimation and error covariance matrix at the next moment using the state equation according to the state estimation and voltage difference at the current moment; updates the state estimation and error covariance matrix at the next moment according to the actual observation value and the observation equation; performs iterative prediction and update on the state estimation and error covariance matrix until the state estimation converges or reaches a preset number of iterations, thereby obtaining the final state estimation equation.

[0157] In some embodiments, Figure 5 The processing module 503 sets the target pressure difference value and the current pressure difference value, and calculates the error between the target pressure difference value and the current pressure difference value; inputs the error into a vector converter, and uses the vector converter to convert the error into multiple neuron input components; uses the neuron self-learning algorithm to adaptively adjust the weight of each neuron input component according to the current state, and normalizes the weight; calculates the control amount at the current moment, calculates the new control amount according to the control amount at the previous moment and the increment of the control amount at the current moment, and outputs the new control amount as a control signal.

[0158] In some embodiments, the single neuron control algorithm model is expressed by the following formula:

[0159]

[0160] Among them, u(k) represents the target pressure difference state value output last, u(k-1) represents the pressure difference value output at the previous moment, and K u represents the gain factor, x i (k) represents the state variable, w i (k) represents the weighting coefficient, Represents the average weighting coefficient.

[0161] In some embodiments, Figure 5 The balancing module 504 uses the interpolation method to find the capacity difference corresponding to the target voltage difference state value according to the preset voltage mapping relationship under different discharge rates or charging rates; calculates the balancing time according to the found capacity difference using the matching balancing resistor, and determines the duty cycle according to the capacity difference, balancing current, balancing efficiency and balancing time.

[0162] In some embodiments, Figure 5The balancing module 504 sets the balancing parameters in the battery management system according to the balancing time and the duty cycle, and controls the balancing switch to connect the balancing resistor to start the voltage balancing process; uses the balancing resistor to passively discharge the battery that exceeds the voltage threshold, and uses the resistor to dissipate the excess energy of the battery in the form of heat to gradually reduce the voltage; continuously monitors the battery cell voltage, and adjusts the working state of the balancing switch in real time, so that the voltage balancing process is carried out according to the preset balancing time and duty cycle.

[0163] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0164] Figure 6 Schematic diagram of the structure of the electronic device 6 provided in the embodiment of the present application. Figure 6 As shown, the electronic device 6 of this embodiment includes: a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, the steps in the above-mentioned various method embodiments are implemented. Alternatively, when the processor 601 executes the computer program 603, the functions of each module / unit in the above-mentioned various device embodiments are implemented.

[0165] Exemplarily, the computer program 603 may be divided into one or more modules / units, which are stored in the memory 602 and executed by the processor 601 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 603 in the electronic device 6.

[0166] The electronic device 6 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 6 may include, but is not limited to, a processor 601 and a memory 602. Those skilled in the art will appreciate that Figure 6 It is only an example of the electronic device 6 and does not constitute a limitation of the electronic device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0167] The processor 601 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0168] The memory 602 may be an internal storage unit of the electronic device 6, for example, a hard disk or memory of the electronic device 6. The memory 602 may also be an external storage device of the electronic device 6, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 6. Further, the memory 602 may also include both an internal storage unit of the electronic device 6 and an external storage device. The memory 602 is used to store computer programs and other programs and data required by the electronic device. The memory 602 may also be used to temporarily store data that has been output or is to be output.

[0169] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0170] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0171] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0172] In the embodiments provided in the present application, it should be understood that the disclosed devices / computer equipment and methods can be implemented in other ways. For example, the device / computer equipment embodiments described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0173] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0175] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, which may be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0176] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A battery balancing control method, characterized in that: include: Obtaining the voltage value corresponding to each battery in the battery pack, and calculating the voltage difference between the maximum voltage battery and the minimum voltage battery in the battery pack; Performing filtering on the voltage difference to obtain a filtered voltage difference; The filtered voltage difference is used as an input of a predetermined single neuron control algorithm model, a neuron self-learning algorithm and a vector converter are used in the single neuron control algorithm model to perform nonlinear processing on the filtered voltage difference and output a target voltage difference state value; The equalization time and the duty cycle are calculated according to the target voltage difference state value, and based on the equalization time and the duty cycle, the equalization switch in the battery management system is controlled to perform voltage equalization until the cell voltage of the battery reaches the target voltage.

2. The method according to claim 1, characterized in that Before filtering the voltage difference, the method further includes: A functional relationship between the voltage difference at each stage and the preset parameters is established, a filtering algorithm model is constructed using the functional relationship, and independent system noise and measurement noise are introduced into the filtering algorithm model, wherein the system noise and measurement noise are both Gaussian white noise and have corresponding covariance matrices.

3. The method according to claim 1, characterized in that The filtering process of the voltage difference comprises: Initializing the state estimation and error covariance matrix of the filtering algorithm model, and setting the process noise covariance matrix and the measurement noise covariance matrix; According to the current state estimation and voltage difference, the state estimation and error covariance matrix of the next moment are predicted by using the state equation; According to the actual observation value and the observation equation, the state estimation and the error covariance matrix of the next moment are updated; Iterative prediction and updating are performed on the state estimation and the error covariance matrix until the state estimation converges or reaches a preset number of iterations, thereby obtaining a final state estimation equation.

4. The method according to claim 1, characterized in that: The method of using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model to perform nonlinear processing on the filtered voltage difference includes: Setting a target pressure difference value and a current pressure difference value, and calculating an error between the target pressure difference value and the current pressure difference value; Inputting the error into a vector converter, and converting the error into a plurality of neuron input components using the vector converter; The neuron self-learning algorithm is used to adaptively adjust the weight of each neuron input component according to the current state, and the weight is normalized; The control amount at the current moment is calculated, and a new control amount is calculated based on the control amount at the previous moment and the increment of the control amount at the current moment, and the new control amount is output as a control signal.

5. The method according to claim 4, characterized in that The single neuron control algorithm model is expressed by the following formula: Among them, u(k) represents the target pressure difference state value output last, u(k-1) represents the pressure difference value output at the previous moment, and K u represents the gain factor, x i (k) represents the state variable, w i (k) represents the weighting coefficient, Represents the average weighting coefficient.

6. The method according to claim 1, characterized in that The calculating of the balancing time and the duty cycle according to the target pressure difference state value comprises: According to the preset voltage mapping relationship under different discharge rates or charge rates, the capacity difference corresponding to the target voltage difference state value is found by using the interpolation method; according to the found capacity difference, the balancing time is calculated using a matching balancing resistor, and the duty cycle is determined according to the capacity difference, balancing current, balancing efficiency and balancing time.

7. The method according to claim 1, characterized in that The controlling the balancing switch in the battery management system to perform voltage balancing based on the balancing time and the duty cycle includes: According to the balancing time and duty cycle, the balancing parameters in the battery management system are set, and the balancing switch is controlled to connect the balancing resistor to start the voltage balancing process; The equalizing resistor is used to passively discharge the battery that exceeds the voltage threshold, and the resistor is used to dissipate the excess energy of the battery in the form of heat to gradually reduce the voltage; The cell voltage of the battery is continuously monitored, and the working state of the equalization switch is adjusted in real time, so that the voltage equalization process is performed according to the preset equalization time and duty cycle.

8. A battery balancing control device, characterized in that: include: An acquisition module is configured to acquire a voltage value corresponding to each battery in the battery pack and calculate a voltage difference between a battery with a maximum voltage and a battery with a minimum voltage in the battery pack; A filtering module is configured to filter the voltage difference to obtain a filtered voltage difference; a processing module configured to use the filtered voltage difference as an input of a predetermined single neuron control algorithm model, perform nonlinear processing on the filtered voltage difference by using a neuron self-learning algorithm and a vector converter in the single neuron control algorithm model, and output a target voltage difference state value; The balancing module is configured to calculate the balancing time and duty cycle according to the target voltage difference state value, and based on the balancing time and duty cycle, control the balancing switch in the battery management system to perform voltage balancing until the battery cell voltage reaches the target voltage.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images