Power supply balance optimization method and power supply balance optimization system

Abstract

The present invention provides a balanced optimization method and a balanced optimization system for a power supply, which monitors and collects battery cell parameters in real time, calculates the battery cell capacity and battery cell internal resistance accordingly, and adopts an improved adaptive Kalman filter algorithm to obtain a predicted value of the state of charge, takes the comprehensive imbalance degree formed by the consistency of the state of charge, battery cell voltage, and battery cell internal resistance as the objective function, and adopts a firefly optimization algorithm to obtain a real-time balancing strategy; according to the power system fault state, battery cell temperature, battery cell voltage, and battery cell state, it is judged whether each battery cell needs to be balanced. The above method can be controlled in real time in different states of charging, discharging, and static state of the power supply, adaptively select a suitable balancing strategy, and comprehensively infer whether to turn on the balance based on a variety of data such as historical state of charge, voltage, current, temperature, and battery cell internal resistance, effectively extending the service life of the power supply and improving its application effect in electric vehicles and energy storage systems.

Description

Technical Field

[0001] The present invention relates to the field of battery management systems, and in particular to a power supply balance optimization method and a power supply balance optimization system. Background Art

[0002] In order to meet the high voltage requirements in the coal mining field, large-capacity lithium battery power supplies for mining are usually obtained by connecting multiple cells in series. However, due to the large number of cells and the large battery capacity, the difference between each cell is likely to increase, and ultimately lead to a rapid decline in battery capacity and life, seriously affecting the overall performance and life of the battery. In addition, the lithium battery power supply in the coal mining field has extremely high safety requirements and needs to comply with safety certification requirements. Among them, the "MT_T1200-2023 Safety Technical Requirements for Explosion-proof Lithium-ion Battery Power Supplies for Mining" requires that the balancing current must be less than or equal to 1 ampere, which will also limit the battery balancing speed to a certain extent. Therefore, the mining lithium battery power supply has very high requirements for the balancing function of the battery management system.

[0003] The battery balancing technology in the prior art has the following disadvantages: (1) Generally, balancing is only enabled for part of the time under specific loads such as the end of charging or in a static state, and it is impossible to achieve real-time balancing control under load switching conditions during the entire process of charging, discharging and static state; (2) It is mainly aimed at small-capacity cells with a small number of cells, and it is difficult to meet the balancing needs of hundreds of large-capacity lithium battery cells in the coal mining field; (3) The use of passive balancing with a small current and slow speed or active balancing with a large current does not meet the safety standards in the coal mining field, and cannot meet the extremely high safety requirements in the mining field; (4) Using open circuit voltage, battery state of charge (SOC) or battery state of health (SOH) as a single balancing judgment condition, it is impossible to intelligently and accurately predict the actual remaining life of the battery, and it is difficult to achieve a good balancing effect. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a power supply balance optimization method and a power supply balance optimization system, which can realize real-time regulation of different states of the lithium battery power supply, such as charging, discharging and static state, and adaptively select and start the balance strategy suitable for different load states.

[0005] In order to solve the above problems, the present invention provides a balanced optimization method for a power supply, wherein the power supply is composed of multiple battery cells connected in series, including: real-time monitoring and collecting the battery cell parameters of each battery cell; calculating the corresponding battery cell capacity and battery cell internal resistance according to the battery cell parameters of each battery cell, and using an improved adaptive Kalman filter algorithm to obtain a predicted state of charge value of each battery cell; according to the battery cell parameters of each battery cell and the predicted state of charge value, taking the comprehensive imbalance degree formed by the consistency of state of charge, the consistency of battery cell voltage, and the consistency of battery cell internal resistance as the objective function, using the firefly optimization algorithm to obtain a real-time balancing strategy, and the comprehensive imbalance degree is: ,in, BCI is the comprehensive imbalance degree, For the charge state consistency, For cell voltage consistency, The internal resistance of the battery cell is consistent, and , , , ,in, , , are respectively the voltage, state of charge and internal resistance of the ith battery cell, and d is the number of battery cells; whether each battery cell needs to be balanced is determined according to the power system fault status, battery cell temperature, battery cell voltage and battery cell status; if there are battery cells that need to be balanced, the real-time balancing strategy is executed.

[0006] In some embodiments, the battery cell parameters include battery cell voltage, battery cell temperature, charging current, charging time, discharging current, and discharging time.

[0007] In some embodiments, the power supply has more than 50 battery cells.

[0008] In some embodiments, the step of using the improved adaptive Kalman filter algorithm to obtain the state of charge prediction value of each battery cell further includes: establishing a state space model, the state space model includes state variables, a state transfer matrix, and an observation matrix, wherein the state of charge is used as the state variable, and the charging current, the discharging current, the battery cell voltage, and the battery cell internal resistance are used as input variables / output variables; initializing the improved adaptive Kalman filter, setting an initial state of charge estimate, as well as an initial estimation error covariance matrix and an initial system noise covariance matrix; predicting the state of charge estimate and updating the state of charge estimate, wherein the state of charge estimate of the next time step is predicted by the state transfer matrix, and the state of charge estimate is updated by the observation matrix according to the real-time measurement value and the state of charge estimate, and the real-time measurement value includes the battery cell voltage, the charging current, and the discharging current; cyclically executing the steps of predicting the state of charge estimate and updating the state of charge estimate until a preset number of iterations is met; and using the state of charge estimate of each battery cell as the state of charge prediction value of each battery cell.

[0009] In some embodiments, in the step of predicting the state of charge estimation value and updating the state of charge estimation value, the estimation error covariance matrix and the system noise covariance matrix are updated.

[0010] In some embodiments, the step of using the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each battery cell also obtains the maximum voltage difference and the maximum temperature difference, wherein the maximum voltage difference is used to calculate the balancing accuracy, and the maximum temperature difference is used to determine the power supply failure.

[0011] In some embodiments, in the step of using the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each battery cell, the relationship between the open circuit voltage and the charge state of the battery cell is fitted into a linear relationship.

[0012] In some embodiments, the step of using the firefly optimization algorithm to obtain a real-time equilibrium strategy further includes: establishing an optimization model, taking the comprehensive imbalance as the objective function; initializing the fireflies, randomly setting the initial positions of the fireflies in the high-dimensional solution space, calculating the value of the objective function of the fireflies at the initial positions as the current minimum value, setting the initial absolute brightness of each firefly, defining a method for calculating the relative attraction between fireflies, and the relative brightness between fireflies; updating the positions of the fireflies, defining a position update formula, so that each firefly moves one step toward a firefly with a brightness higher than itself to a new position, wherein the step length is related to a step length coefficient; adjusting the step length coefficient; evaluating the objective function, calculating the value of the objective function of the firefly at the new position; recording the optimal solution, and when the new value of the objective function is less than the current minimum value, updating the current minimum value, and recording the corresponding firefly position as the current optimal solution; looping through iterations until the number of iterations or accuracy requirements are met, each iteration including the steps of updating the positions of the fireflies, adjusting the step length coefficient, evaluating the objective function, and recording the current optimal solution; using the current optimal solution as the real-time equilibrium strategy.

[0013] In some embodiments, the position of the fireflies , the initial absolute brightness of each firefly is: ,in, , , are the cell voltage, state of charge, and internal resistance of the cell in the i-th cell, respectively. 1 , k 2 , k 3 are the weight coefficients of voltage, state of charge and internal resistance respectively; the relative attraction between the fireflies is: ,in is the initial attraction of fireflies, is the light absorption coefficient, is the distance between the i-th firefly and the j-th firefly, and the calculation formula is: , the relative brightness between the fireflies is: , the position update formula is: ,in and are the positions of the i-th and j-th fireflies at time t, is the step size coefficient, i.e. the attraction between fireflies.

[0014] In some embodiments, in the step of determining whether each battery cell needs to be balanced based on the power system fault status, battery cell temperature, battery cell voltage and battery cell status, the battery cells that need to be balanced meet the following conditions at the same time: the power control system where the battery cell is located currently has no faults in self-inspection; the battery cell is in the non-mid-charging, non-late-discharging, and non-early-stationary state; the battery cell temperature is less than or equal to 50 degrees Celsius; and the battery cell voltage is greater than or equal to 2.8 volts.

[0015] In some embodiments, if the battery cell is in a charging state and the cumulative charging time of the battery cell is greater than 0.5 hours and less than 1.5 hours, the battery cell is in the middle charging state; if the battery cell is in a discharging state and the cumulative discharging time of the battery cell is greater than 1.5 hours, the battery cell is in the late discharging state; if the battery cell is in a static state and the cumulative static time of the battery cell is less than 1.5 hours, the battery cell is in the early static state.

[0016] In some embodiments, when the real-time balancing strategy is executed, the balancing switches of the battery cells that need balancing are turned on, and the balancing switches of the battery cells that do not need balancing are turned off.

[0017] In some embodiments, if there are no battery cells that need to be balanced, the battery cell parameters of each battery cell continue to be monitored and collected in real time.

[0018] In order to solve the above problems, the present invention also provides a balanced optimization system for a power supply, comprising: a cell data acquisition module: the cell data acquisition module is used to monitor and collect cell parameters of each cell in real time; a data processing and analysis module: connected to the cell data acquisition module, the data processing and analysis module is used to calculate the corresponding cell capacity and cell internal resistance according to the cell parameters of each cell, and adopt the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each cell; a balanced decision generation module: connected to the cell data acquisition module and the data processing and analysis module, the balanced decision generation module is used to obtain a real-time balanced strategy based on the cell parameters of each cell and the charge state prediction value, with the charge state consistency, cell voltage consistency, and cell internal resistance consistency combined to form a comprehensive imbalance as the objective function, and the firefly optimization algorithm is adopted. The comprehensive imbalance is: ,in, BCI is the comprehensive imbalance degree, For the charge state consistency, For cell voltage consistency, The internal resistance of the battery cell is consistent, and , , , ,in, , , are respectively the voltage, state of charge and internal resistance of the ith battery cell, and d is the number of batteries; a balancing condition judgment module: connected to the battery cell data acquisition module, the balancing condition judgment module is used to judge whether each battery cell needs to be balanced according to the power system fault state, battery cell temperature, battery cell voltage and battery cell state. If there are batteries that need to be balanced, the real-time balancing strategy is executed. If all batteries do not need to be balanced, the battery cell parameters of each battery cell continue to be monitored and collected in real time, and the battery cell state includes charging state, discharging state, and static state; a balancing condition execution module: connected to the balancing decision generation module and the balancing condition judgment module, the balancing condition execution module is used to execute the real-time balancing strategy, control the circuit board of the battery cell management unit in real time through the instructions of the battery cluster management unit, turn on the balancing switch of the battery cell that needs to be balanced and turn off the balancing switch of the battery cell that does not need to be balanced.

[0019] The above technical solution monitors and collects the cell parameters of each cell in real time; calculates the corresponding cell capacity and cell internal resistance according to the cell parameters of each cell, and uses the improved adaptive Kalman filter algorithm to obtain the predicted value of the state of charge of each cell, takes the comprehensive imbalance formed by the consistency of the state of charge, the consistency of the cell voltage, and the consistency of the cell internal resistance as the objective function, and uses the firefly optimization algorithm to obtain the real-time balancing strategy; judges whether each cell needs to be balanced according to the power system fault state, cell temperature, cell voltage and cell state, and executes the real-time balancing strategy when there are cells that need to be balanced. The above balancing optimization method can be controlled in real time in different states of charging, discharging and static state of the lithium battery power supply, adaptively selects the balancing strategy suitable for different load states, and comprehensively calculates and evaluates whether it is necessary to start balancing based on various data such as historical state of charge, voltage, current, temperature and cell internal resistance, effectively extending the service life of the lithium battery and improving its application effect in electric vehicles and energy storage systems.

[0020] It should be understood that the above general description and the detailed description below are exemplary and explanatory only and cannot limit the present invention. The techniques, methods and devices known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the techniques, methods and devices should be considered as part of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments of the present invention are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of a power supply balancing optimization method provided by an embodiment of the present invention;

[0023] Figure 2 It is a flow chart of obtaining a predicted value of the state of charge of a battery cell using an improved adaptive Kalman filter algorithm provided by an embodiment of the present invention;

[0024] Figure 3 is a flow chart of a real-time balancing strategy obtained by using a firefly optimization algorithm provided by an embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of a power supply balancing optimization system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Please refer to Figure 1~Figure 3 , Figure 1 is a flow chart of a power supply balancing optimization method provided by an embodiment of the present invention, Figure 2 is a flow chart of obtaining a predicted value of the state of charge of a battery cell using an improved adaptive Kalman filter algorithm provided by an embodiment of the present invention. Figure 3 It is a flow chart of obtaining a real-time balancing strategy using a firefly optimization algorithm provided by an embodiment of the present invention.

[0028] like Figure 1 As shown, the power supply balancing optimization method includes: step S11, real-time monitoring and collecting the cell parameters of each cell; step S12, calculating the corresponding cell capacity and cell internal resistance according to the cell parameters of each cell, and using the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each cell; step S13, according to the cell parameters of each cell and the charge state prediction value, taking the charge state consistency, cell voltage consistency, and cell internal resistance consistency as the objective function, and using the firefly optimization algorithm to obtain a real-time balancing strategy; step S14, judging whether each cell needs to be balanced according to the power system fault state, cell temperature, cell voltage and cell state, if there are cells that need to be balanced, executing the real-time balancing strategy.

[0029] The number of battery cells of the power supply is greater than 50. For example, the number of battery cells is 51, 60, 70, 100, 120, etc. In this embodiment, the number of battery cells is 60.

[0030] Referring to step S11, the cell parameters of each cell are monitored and collected in real time. In this step, the cell parameters include cell voltage, cell temperature, charging current, charging time, discharge current, and discharge time. In this embodiment, the cell parameters are monitored and collected in real time by the monitoring circuit board and sensor device of the battery module unit (BMU) in the large-capacity power supply system for mining.

[0031] The power supply in this embodiment is a large-capacity power supply system for mining, and the number of battery cells is 60. The voltage V of each battery cell is collected. 1 -V 60 , Charging current Ic 1 -Ic 60 , discharge current Id 1 -Id 60 , temperature T 1 -T 60 and charging time Tc 1 -Tc 60 , charging time Td 1 -Td 60 wait.

[0032] Referring to step S12, the corresponding cell capacity and cell internal resistance are calculated according to the cell parameters of each cell, and the improved adaptive Kalman filter algorithm is used to obtain the charge state prediction value of each cell.

[0033] In this step, the step of using the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each battery cell further includes: step S121, establishing a state space model, the state space model includes state variables, state transfer matrix, observation matrix, wherein the charge state is used as the state variable, and the charging current, discharge current, battery cell voltage, and battery cell internal resistance are used as input variables and output variables; step S122, initializing the improved adaptive Kalman filter, setting the initial charge state estimation value, as well as the initial estimation error covariance matrix, and the initial system noise covariance matrix; step S123, obtaining the charge state prediction value. The state estimation value is estimated and the state of charge estimation value is updated, the state of charge estimation value of the next time step is predicted through the state transfer matrix, and the state of charge estimation value is updated through the observation matrix according to the real-time measurement value and the state of charge estimation value, and the real-time measurement value includes the battery cell voltage, charging current, and discharging current; step S124, judging whether the number of iterations of step S123 meets the preset number of first iterations, if so, executing step S125, if not, continuing to execute step S123; step S125, obtaining the state of charge estimation value of each battery cell as the state of charge prediction value of each battery cell.

[0034] Referring to step S121, a state space model is established, wherein the state space model includes state variables, a state transfer matrix, and an observation matrix, wherein the state of charge is used as the state variable, and the charging current, the discharging current, the cell voltage, and the cell internal resistance are used as input variables / output variables. In this embodiment, the state space model is a nonlinear state space model. In addition, the relationship between the open circuit voltage and the state of charge of the cell is fitted as a linear relationship, and varies with temperature. Experimental verification shows that these two assumptions are consistent with the cell experimental results.

[0035] Referring to step S122, the improved adaptive Kalman filter is initialized to set the initial state of charge estimation value, the initial estimation error covariance matrix, and the initial system noise covariance matrix, wherein the system noise covariance matrix is ​​used to describe the system error.

[0036] Referring to step S123, the state of charge estimate is predicted and updated, the state of charge estimate of the next time step is predicted through the state transfer matrix, and the state of charge estimate is updated through the observation matrix according to the real-time measurement value and the state of charge estimate, and the real-time measurement value includes the cell voltage, the charging current, and the discharging current. Specifically, in the process of predicting the state of charge estimate of the next time step through the state transfer matrix, the influence of process noise is considered; in the process of updating the state of charge estimate through the observation matrix, the influence of observation noise is considered. In this step, the calculation of the Kalman gain is also included, and the estimation error covariance matrix and the system noise covariance matrix are updated.

[0037] Referring to step S124, determine whether the number of iterations of step S123 meets the first preset number of iterations. If yes, execute step S125. If no, continue to execute step S123. The first preset number of iterations is 20 to 50. For example, the preset number of iterations is set to 20, 25, 30, 40, and 50 times. In this embodiment, the first preset number of iterations is 30. Each time step S123 is executed, the number of iterations is increased by one. When the number of iterations is less than 30, continue to execute step S123. When the number of iterations is equal to 30, the iteration ends and step S125 is executed.

[0038] Referring to step S125, the estimated state of charge value of each battery cell is used as the predicted state of charge value of each battery cell. That is, when the iteration stops, the estimated state of charge value of each battery cell obtained in the last iteration is used as the predicted state of charge value of each battery cell.

[0039] In this embodiment, the step of using the improved adaptive Kalman filter algorithm to obtain the predicted value of the state of charge of each battery cell also obtains the maximum voltage difference and the maximum temperature difference, wherein the maximum voltage difference is used to calculate the balancing accuracy, and the maximum temperature difference is used to determine the power failure. In some embodiments, the step of using the improved adaptive Kalman filter algorithm to obtain the predicted value of the state of charge of each battery cell also calculates the average voltage and average temperature of the battery cell to evaluate the effect of voltage balancing.

[0040] Continue reading Figure 1, refer to step S13, according to the cell parameters of each cell and the predicted state of charge, the comprehensive imbalance formed by the consistency of state of charge, cell voltage and cell internal resistance is used as the objective function, and the firefly optimization algorithm is used to obtain a real-time balancing strategy. The firefly algorithm (FA) is an optimization algorithm based on swarm intelligence, which simulates the flashing behavior of fireflies to find the optimal solution to the problem. In this step, one firefly represents one cell, and the brightness of each firefly represents the balancing index of each cell.

[0041] like Figure 3 As shown, the steps of using the firefly optimization algorithm to obtain a real-time equilibrium strategy further include: step S131, establishing an optimization model, taking the comprehensive imbalance as the objective function; step S132, initializing the fireflies, randomly setting the initial positions of the fireflies in the high-dimensional solution space, calculating the value of the objective function of the fireflies at the initial positions as the current minimum value, setting the initial absolute brightness of each firefly, defining the relative attraction between the fireflies, and the relative brightness between the fireflies; step S133, updating the positions of the fireflies, defining a position update formula, so that each firefly moves one step longer to a new position towards a firefly with a brightness higher than itself. The step size is related to the step size coefficient; step S134, adjusting the step size coefficient; step S135, evaluating the objective function, calculating the value of the objective function at the new position of the firefly; step S136, recording the optimal solution, when the new value of the objective function is less than the current minimum value, updating the current minimum value, and recording the corresponding firefly position as the current optimal solution; step S137, judging whether the iteration meets the preset number of times of the second iteration, each of the iterations includes step S133, step S134, step S135, step S136; step S138, using the current optimal solution as the real-time balancing strategy.

[0042] Referring to step S131, an optimization model is established, and the comprehensive imbalance is used as the objective function. The comprehensive imbalance is

[0043] ,

[0044] in, , and They are defined as follows:

[0045] ,

[0046] , , ,

[0047] in, , , are the voltage, state of charge and internal resistance of the ith cell respectively, and d is the number of cells.

[0048] Referring to step S132, the fireflies are initialized, the initial positions of the fireflies are randomly set in the high-dimensional solution space, the value of the objective function of the fireflies at the initial positions is calculated as the current minimum value, the initial absolute brightness of each firefly is set, and the relative attraction between the fireflies and the relative brightness between the fireflies are defined.

[0049] Wherein, the initial position of the firefly is: ,

[0050] The initial absolute brightness of each firefly is:

[0051] ,in, , , are the cell voltage, state of charge, and internal resistance of the cell in the i-th cell, respectively. 1 , k 2 , k 3 are weight coefficients of voltage, state of charge and internal resistance respectively. In order to improve the algorithm performance and convergence speed, this embodiment adopts multi-objective weighting technology by considering the inertia weight and nonlinear factors of battery characteristics.

[0052] The relative attraction between the fireflies is: ,in is the initial attraction of fireflies, is the light absorption coefficient, is the distance between the i-th firefly and the j-th firefly, and the calculation formula is: , the relative brightness between the fireflies is: .

[0053] Referring to step S133, the positions of the fireflies are updated, and a position update formula is defined so that each firefly moves one step length to a new position toward a firefly with a brightness higher than itself, and the step length is related to the step length coefficient.

[0054] The position update formula is: ,in and are the positions of the i-th and j-th fireflies at time t, is the attraction between fireflies, i.e., the step length coefficient.

[0055] Referring to step S134, the step coefficient is adjusted. According to the calculation method of the relative attraction between the fireflies defined in step S132, the step coefficient is gradually reduced as the number of iterations increases, ensuring that the algorithm can quickly explore the solution space in the early stage and gradually refine the search in the later stage.

[0056] Referring to step S135, the objective function is evaluated to calculate the value of the objective function when the firefly is in the new position. In this step, the objective function value is calculated for each new position of the firefly to evaluate the quality of the position.

[0057] Referring to step S136, the optimal solution is recorded, and when the new value of the objective function is less than the current minimum value, the current minimum value is updated, and the firefly position is recorded as the current optimal solution. That is, the firefly position corresponding to the current minimum value is the current optimal solution.

[0058] Referring to step S137, it is determined whether the iteration meets the second preset number of iterations, and each of the iterations includes step S133, step S134, step S135, and step S136. The second preset number of iterations is 50 to 100, for example, the preset number of iterations is set to 50, 55, 60, 65, 70, 80, and 100 times. In this embodiment, the second preset number of iterations is 50, and the number of iterations is increased by one after each execution of step S133, step S134, step S135, and step S136. When the number of iterations is less than 50, the iteration continues to be executed. When the number of iterations reaches 50, step S138 is executed. In some embodiments, whether the iteration stops is determined by the accuracy requirement.

[0059] Referring to step S138, the current optimal solution is used as the real-time balancing strategy. Since a current optimal solution is obtained each time steps S133, S134, S135, and S136 are executed in a loop, after the loop is completed, the current optimal solution after the last iteration is used as the real-time balancing strategy.

[0060] Continue reading Figure 1 Referring to step S14, it is determined whether each cell needs to be balanced according to the power system fault state, cell temperature, cell voltage and cell state. If there is a cell that needs to be balanced, the real-time balancing strategy is executed. In this step, if there is no cell that needs to be balanced, the cell parameters of each cell are continuously monitored and collected in real time.

[0061] Specifically, the battery cells that need to be balanced meet the following conditions at the same time: (1) the power control system where the battery cells are located has no faults in the current self-test; (2) the battery cells are not in the middle of charging, the late stage of discharging, or the early stage of static state; (3) the battery cell temperature is less than or equal to 50 degrees Celsius (°C); (4) the battery cell voltage is greater than or equal to 2.8 volts.

[0062] The real-time balancing strategy is executed when the power control system operates normally. Condition (1) is the primary condition for executing the real-time balancing strategy.

[0063] The state of the battery cell includes a charging state, a discharging state, and a static state. Accordingly, the battery cell being in the non-charging middle stage means that the battery cell is not in the charging middle stage; the battery cell being in the non-discharging late stage means that the battery cell is not in the discharging late stage; the battery cell being in the non-static early stage means that the battery cell is not in the static early stage. Condition (2) means that the battery cell is in a state other than the charging middle stage, the discharging late stage, and the static early stage.

[0064] In this embodiment, if the cumulative charging time of the battery cell is greater than 0.5 hours and less than 1.5 hours, the battery cell is in the middle stage of charging; if the battery cell is in a discharging state and the cumulative discharge time of the battery cell is greater than 1.5 hours, the battery cell is in the late stage of discharging; if the battery cell is in a static state and the cumulative static time of the battery cell is less than 1.5 hours, the battery cell is in the early stage of static.

[0065] When the cell temperature of the battery cell exceeds 50°C, the charging efficiency and battery life of the battery cell will be greatly reduced, corresponding to condition (3). In this embodiment, the single cell over-discharge protection of the battery cell is 2.8V. When the voltage of the battery cell is less than 2.8V, the battery cell will not start balancing, corresponding to condition (4).

[0066] In this step, when executing the real-time balancing strategy, the balancing switches of the cells that need to be balanced are turned on, and the balancing switches of the cells that do not need to be balanced are turned off. In this embodiment, the battery cluster management unit (Battery Cluster Unit, referred to as BCU) is instructed to control the battery module unit (Battery Module Unit, check BMU) circuit board in real time to turn on the balancing switches of the cells that need to be adjusted in the optimization balancing strategy, and turn off the balancing switches of the cells that do not need to be balanced, and the balancing is stopped until the balancing conditions are met.

[0067] If there are no cells that need to be balanced, the cell parameters of each cell continue to be monitored and collected in real time, and steps S11 to S14 are executed.

[0068] Based on the same inventive concept, the present invention also provides a power supply balance optimization system.

[0069] Figure 4 FIG. 1 is a schematic diagram of a power supply balancing optimization system provided by an embodiment of the present invention. Figure 4 As shown, the power supply balancing optimization system includes: a battery cell data acquisition module, a data processing and analysis module, a balancing decision generation module, a balancing condition judgment module, and a balancing condition execution module. The battery cell data acquisition module is used to monitor and collect the battery cell parameters of each battery cell in real time; the data processing and analysis module is connected to the battery cell data acquisition module, and the data processing and analysis module is used to calculate the corresponding battery cell capacity and battery cell internal resistance according to the battery cell parameters of each battery cell, and use the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each battery cell; the balancing decision generation module is connected to the battery cell data acquisition module and the data processing and analysis module, and the balancing decision generation module is used to use the battery cell parameters of each battery cell, and the charge state prediction value, and the comprehensive imbalance degree formed by the charge state consistency, battery cell voltage consistency, and battery cell internal resistance consistency as the objective function, and use the firefly optimization algorithm to obtain a real-time balancing strategy; the balancing condition judgment module The module is connected to the battery cell data acquisition module, and the balancing condition judgment module is used to judge whether each battery cell needs to be balanced according to the power system fault state, battery cell temperature, battery cell voltage and battery cell state. If there is a battery cell that needs to be balanced, the real-time balancing strategy is executed. If all the battery cells do not need to be balanced, the battery cell parameters of each battery cell continue to be monitored and collected in real time. The battery cell state includes charging state, discharging state, and static state; the balancing condition execution module is connected to the balancing decision generation module and the balancing condition judgment module. The balancing condition execution module is used to execute the real-time balancing strategy, control the circuit board of the battery cell management unit in real time through the instructions of the battery cluster management unit, turn on the balancing switch of the battery cell that needs to be balanced and turn off the balancing switch of the battery cell that does not need to be balanced.

[0070] The comprehensive imbalance degree is:

[0071] ,in, BCI is the comprehensive imbalance degree, For the charge state consistency, For cell voltage consistency, The internal resistance of the battery cell is consistent, and , , , ,in, , , are the voltage, state of charge and internal resistance of the ith cell respectively, and d is the number of cells.

[0072] It should be noted that, in this document, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion. The various embodiments in this specification are described in a related manner, and the same and similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0073] The above is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. It should be noted that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for optimizing the balance of a power supply, wherein the power supply is composed of a plurality of battery cells connected in series, characterized in that: include: The cell parameters of each cell are monitored and collected in real time; the corresponding cell capacity and cell internal resistance are calculated according to the cell parameters of each cell, and the state of charge prediction value of each cell is obtained by using the improved adaptive Kalman filter algorithm; according to the cell parameters of each cell and the state of charge prediction value, the comprehensive imbalance degree formed by the consistency of state of charge, cell voltage consistency, and cell internal resistance consistency is used as the objective function, and the firefly optimization algorithm is used to obtain a real-time balancing strategy, and the comprehensive imbalance degree is: ,in, BCI is the comprehensive imbalance degree, For the charge state consistency, For cell voltage consistency, The internal resistance of the battery cell is consistent, and , , , ,in, , , are respectively the voltage, state of charge and internal resistance of the ith battery cell, and d is the number of batteries; whether each battery cell needs to be balanced is determined according to the power system fault state, battery cell temperature, battery cell voltage and battery cell state, and if there is a battery cell that needs to be balanced, the real-time balancing strategy is executed; the battery cell that needs to be balanced meets the following conditions at the same time: the power control system where the battery cell is located has no fault in the current self-test, the battery cell is in the non-charging middle stage, the non-discharging late stage, and the non-static early stage, the battery cell temperature of the battery cell is less than or equal to 50 degrees Celsius, and the battery cell voltage of the battery cell is greater than or equal to 2.8 volts; wherein, if the battery cell is in a charging state, the cumulative charging time of the battery cell is greater than 0.5 hours and less than 1.5 hours, then the battery cell is in the middle charging stage; if the battery cell is in a discharging state, the cumulative discharge time of the battery cell is greater than 1.5 hours, then the battery cell is in the discharging late stage; if the battery cell is in a static state, the cumulative static time of the battery cell is less than 1.5 hours, then the battery cell is in the static early stage.

2. The method according to claim 1, characterized in that The battery cell parameters include battery cell voltage, battery cell temperature, charging current, charging time, discharging current, and discharging time.

3. The method according to claim 1, characterized in that The power supply has more than 50 cells.

4. The method according to claim 1, characterized in that: The step of using the improved adaptive Kalman filter algorithm to obtain the state of charge prediction value of each battery cell further includes: establishing a state space model, the state space model includes state variables, a state transfer matrix, and an observation matrix, wherein the state of charge is used as the state variable, and the charging current, the discharging current, the battery cell voltage, and the battery cell internal resistance are used as input / output variables; initializing the improved adaptive Kalman filter, setting the initial state of charge estimation value, as well as the initial estimation error covariance matrix and the initial system noise covariance matrix; predicting the state of charge estimation value and updating the state of charge estimation value, wherein the state of charge estimation value of the next time step is predicted by the state transfer matrix, and the state of charge estimation value is updated by the observation matrix according to the real-time measurement value and the state of charge estimation value, and the real-time measurement value includes the battery cell voltage, the charging current, and the discharging current; cyclically executing the steps of predicting the state of charge estimation value and updating the state of charge estimation value until the preset number of iterations is met; and using the state of charge estimation value of each battery cell as the state of charge prediction value of each battery cell.

5. The method according to claim 4, characterized in that In the step of predicting the state of charge estimation value and updating the state of charge estimation value, the estimation error covariance matrix and the system noise covariance matrix are updated.

6. The method according to claim 1, characterized in that In the step of obtaining the predicted value of the state of charge of each battery cell using the improved adaptive Kalman filter algorithm, the maximum voltage difference and the maximum temperature difference are also obtained, wherein the maximum voltage difference is used to calculate the balancing accuracy, and the maximum temperature difference is used to determine the power supply failure.

7. The method according to claim 1, characterized in that In the step of obtaining the predicted value of the state of charge of each battery cell using the improved adaptive Kalman filter algorithm, the relationship between the open circuit voltage and the state of charge of the battery cell is fitted into a linear relationship.

8. The method according to claim 1, characterized in that The step of using the firefly optimization algorithm to obtain a real-time equilibrium strategy further includes: establishing an optimization model, taking the comprehensive imbalance as the objective function; initializing the fireflies, randomly setting the initial positions of the fireflies in the high-dimensional solution space, calculating the value of the objective function of the fireflies at the initial positions as the current minimum value, setting the initial absolute brightness of each firefly, defining a method for calculating the relative attraction between fireflies, and the relative brightness between fireflies; updating the positions of the fireflies, defining a position update formula, so that each firefly moves one step toward a firefly with a brightness higher than itself to a new position, wherein the step length is related to a step length coefficient; adjusting the step length coefficient; evaluating the objective function, calculating the value of the objective function of the firefly at the new position; recording the optimal solution, and when the new value of the objective function is less than the current minimum value, updating the current minimum value, and recording the corresponding firefly position as the current optimal solution; looping iterations until the number of iterations or accuracy requirements are met, each iteration including the steps of updating the positions of the fireflies, adjusting the step length coefficient, evaluating the objective function, and recording the current optimal solution; using the current optimal solution as the real-time equilibrium strategy.

9. The method according to claim 8, characterized in that The positions of the fireflies are: , the initial absolute brightness of each firefly is: ,in, , , are the cell voltage, state of charge, and internal resistance of the cell in the ith section, respectively; k1, k2, and k3 are the weight coefficients of voltage, state of charge, and internal resistance, respectively; the relative attraction between the fireflies is: ,in, is the initial attraction of fireflies, is the light absorption coefficient, is the distance between the i-th firefly and the j-th firefly, The calculation formula is: , the relative brightness between the fireflies is: , the position update formula is: ,in and are the positions of the i-th and j-th fireflies at time t, is the step size coefficient, i.e. the attraction between fireflies.

10. The method according to claim 1, characterized in that When the real-time balancing strategy is executed, the balancing switches of the cells that need to be balanced are turned on, and the balancing switches of the cells that do not need to be balanced are turned off.

11. The method according to claim 1, characterized in that: If there are no battery cells that need to be balanced, the battery cell parameters of each battery cell continue to be monitored and collected in real time.

12. A power supply balancing and optimization system, characterized in that: include: Cell data acquisition module: the cell data acquisition module is used to monitor and collect the cell parameters of each cell in real time; data processing and analysis module: connected to the cell data acquisition module, the data processing and analysis module is used to calculate the corresponding cell capacity and cell internal resistance according to the cell parameters of each cell, and use the improved adaptive Kalman filter algorithm to obtain the charge state prediction value of each cell; Balance decision generation module: connected to the cell data acquisition module and the data processing and analysis module, the balance decision generation module is used to obtain a real-time balance strategy using the firefly optimization algorithm based on the cell parameters of each cell and the predicted value of the state of charge, the comprehensive imbalance degree formed by the consistency of the state of charge, the consistency of the cell voltage, and the consistency of the cell internal resistance as the objective function, and the comprehensive imbalance degree is: ,in, BCI is the comprehensive imbalance degree, For the charge state consistency, For cell voltage consistency, The internal resistance of the battery cell is consistent, and , , , ,in, , , are the voltage, state of charge and internal resistance of the ith battery cell respectively, and d is the number of batteries; a balancing condition judgment module: connected to the battery cell data acquisition module, the balancing condition judgment module is used to judge whether each battery cell needs to be balanced according to the power system fault state, battery cell temperature, battery cell voltage and battery cell state. If there is a battery cell that needs to be balanced, the real-time balancing strategy is executed. If all batteries do not need to be balanced, the battery cell parameters of each battery cell continue to be monitored and collected in real time. The battery cell state includes charging state, discharging state, and static state. The battery cell that needs to be balanced meets the following conditions at the same time: the power control system where the battery cell is located currently has no faults in self-test, the battery cell is in the non-charging middle stage, the non-discharging late stage, and the non-static early stage, the battery cell temperature of the battery cell is less than or equal to 50 degrees Celsius, and the battery cell voltage of the battery cell is The voltage is greater than or equal to 2.8 volts; wherein, if the battery cell is in a charging state and the cumulative charging time of the battery cell is greater than 0.5 hours and less than 1.5 hours, the battery cell is in a middle charging state; if the battery cell is in a discharging state and the cumulative discharging time of the battery cell is greater than 1.5 hours, the battery cell is in a late discharging state; if the battery cell is in a static state and the cumulative static time of the battery cell is less than 1.5 hours, the battery cell is in an early static state; a balancing condition execution module: connected to the balancing decision generation module and the balancing condition judgment module, the balancing condition execution module is used to execute the real-time balancing strategy, and control the circuit board of the battery cell management unit in real time through the instructions of the battery cluster management unit, turn on the balancing switch of the battery cell that needs to be balanced and turn off the balancing switch of the battery cell that does not need to be balanced.

Images