Equalization method and system of lithium ion battery pack

Abstract

The invention relates to the technical field of battery packs, in particular to an equalization method and system for a lithium ion battery pack, and the method comprises the following steps: 1, improving an equalization topological structure; 2, establishing an equalization system mathematical model; 3, formulating an equalization control strategy and an equalization control target; and 4, designing a multi-target sliding mode equalization controller, and adjusting an equalization system. According to the invention, the equalization efficiency of the battery pack can be well improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery packs, and in particular to a method and system for equalizing a lithium-ion battery pack. Background Art

[0002] As the world faces increasing challenges in ecology and energy, the energy sector is undergoing a profound transformation from traditional fossil fuels to cleaner and more efficient new energy sources. Lithium batteries have been widely used in electric vehicles and new energy power generation due to their high energy density, high reliability, low self-discharge rate, long cycle life and environmental friendliness. In order to meet the needs of high-voltage and high-capacity applications such as electric vehicles, hundreds of single cells need to be connected in series and parallel. However, due to factors such as uneven mixing of raw materials during the manufacturing process, there are inevitably certain differences between single cells within the battery pack. In addition, factors such as rapid charging and discharging, static periods and different operating temperatures introduce differences that accumulate over time, exacerbating the inconsistency of the battery pack. This so-called "barrel effect" not only reduces the energy utilization efficiency of the battery pack, but also exacerbates problems related to performance degradation and safety.

[0003] In order to improve battery pack consistency, increase capacity and extend battery pack cycle life, the application of cell balancing technology becomes crucial. As an important component of the battery management system (BMS), cell balancing technology is of great significance in ensuring the safe and efficient operation of the battery pack, improving the mileage of electric vehicles and optimizing the overall cost-effectiveness.

[0004] According to the different energy transfer methods, balancing systems can be divided into passive balancing and active balancing. Passive balancing dissipates the excess energy of high-energy batteries in the form of heat through parallel bypass resistors. This structure is simple, suitable for industrial production, cost-effective, and easy to control and implement. However, passive balancing has limitations such as low balancing current, poor energy utilization, and only applicable to charging conditions. Therefore, it faces challenges in meeting high energy storage and power requirements such as electric vehicles (EVs). In contrast, active balancing transfers energy from batteries with surplus charge to batteries with scarce energy through energy transfer elements to achieve energy balance between batteries. Whether it is a passive balancing or active balancing system, its main goal is to maintain the consistency of the state of charge (SOC) of each battery as much as possible, so that the battery pack can store and release energy to the maximum extent. In recent years, active balancing has received extensive research attention due to its faster balancing speed, lower energy consumption, and ability to balance battery packs under various working conditions.

[0005] The research on active balancing technology mainly involves two key aspects: the research on balancing topology and the research on control strategy. For balancing topology, in order to achieve the best balancing system, factors such as the number of components, hardware cost, balancing speed, energy utilization and control complexity must be considered. However, the existing balancing topology and control methods have certain limitations. Summary of the invention

[0006] The present invention provides a lithium-ion battery pack equalization method and system, which can overcome certain defects of the prior art.

[0007] The equalization method of the lithium-ion battery pack according to the present invention comprises the following steps:

[0008] 1. Improve the balanced topology structure;

[0009] 2. Establish a mathematical model of the equilibrium system;

[0010] 3. Formulate balanced control strategies and balanced control objectives;

[0011] 4. Design a multi-objective sliding mode equilibrium controller to adjust the equilibrium system.

[0012] Preferably, in step one, the improved balancing topology structure includes n series-connected single cells, n / 3 switch matrices, n MOSFETs and n / 3 flyback transformers, and every three consecutive cells form a group, which are connected to the primary winding of the flyback transformer through the switch matrix, and the secondary winding of the flyback transformer is connected to six consecutive cells through MOSFETs.

[0013] Preferably, in step 2, the method for establishing the mathematical model of the equilibrium system is:

[0014]

[0015] in, represents the change vector of SOC at time t, and u(t) represents the equilibrium current vector;

[0016] By transforming the above formula, the relationship between the balancing variable and the balancing current of each single cell is obtained:

[0017]

[0018] Among them, x(t) is the SOC vector at time t, Q x is an n×n diagonal matrix defined as the total capacity of the battery pack, where c1…c n Respectively represent the battery BAT1 ~ BAT n The actual maximum capacity of , τ is the time variable;

[0019]

[0020] T represents the energy transfer matrix, where k1=1 / 6, k2=(1 / 6)-1, k3=(1 / 3)-1, k4=1 / 3;

[0021]

[0022] The switch matrix S plays a key role in regulating the connectivity between each single cell and the flyback transformer. If the i-th single cell is connected to the flyback transformer, the corresponding element Si in the switch matrix is ​​set to 1; on the contrary, when the two are not connected, the element is set to 0;

[0023] S=diag(S1,S2,S3…S n )∈R n×n

[0024] u e (t) represents the current of the entire battery pack during the charging or discharging process in the balancing stage. During the charging cycle, u e (t) takes a positive value; during the discharge cycle, u e (t) takes a negative value; during the static period, u e (t) is always set to zero.

[0025] Preferably, in step three, the balancing control strategy is:

[0026] (1) Predefine the threshold δ of the mean square error MSE of the SOC in the battery module 2 th; when MSE exceeds the set threshold δ 2 At th, the system starts balancing;

[0027] (2) Consider three consecutive batteries as a battery group. For each battery group, determine the maximum SOC single battery SOC max(i) and minimum SOC single cell SOC min(i) ;

[0028] (3) After starting the balance, the specific balance strategy is as follows:

[0029] a. The battery group is i≤n-1, and the status is:

[0030] soc avg(i) <soc avg(i+1) And soc avg -soc min(i) >0

[0031] The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group;

[0032] b. The battery group is i≤n-1, and the status is:

[0033] soc avg(i) <soc avg(i+1) And soc avg -soc min(i) ≤0

[0034] The direction of energy transfer is: the ith battery group does not start balancing;

[0035] c. The battery group is i≤n-1, and the status is:

[0036] soc avg(i) >soc avg(i+1) And soc max(i) -soc avg >0

[0037] The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group;

[0038] d. The battery group is i≤n-1, and the status is:

[0039] soc avg(i) >soc avg(i+1) And soc max(i) -soc avg ≤0

[0040] The direction of energy transfer is: the ith battery group does not start balancing;

[0041] e. The battery group is i=n, ​​and the status is:

[0042] soc max(i) -soc avg >0 and soc max(i) -soc avg(i) >soc avg(i) -soc min(i)

[0043] The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group;

[0044] f. The battery group is i=n, ​​and the status is:

[0045] soc avg -soc min(i) >0 and soc avg(i) -soc min(i) >soc max(i) -soc avg(i)

[0046] The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group;

[0047] g. The battery group is i=n, ​​the state is: else, and the energy transfer direction is: the ith battery group does not start balancing;

[0048] (4) Estimate the SOH of each battery and obtain the aging degree SOH of each battery i , when the average SOH of the battery pack avg -SOH i >0.5%, the balancing system will not balance the i-th battery to avoid aggravating the aging degree of the i-th battery; when the average SOH of the battery pack avg -SOH i ≤0.5%, the equilibrium system executes the strategy of step (3);

[0049] Among them, the comprehensive indicator SOC is set as the balancing variable, and the average SOC of each battery pack is calculated as follows:

[0050]

[0051] Among them, SOC j Indicates the SOC value of the jth single cell;

[0052] The average SOC of the entire battery module is expressed as:

[0053]

[0054] n represents the number of single cells in the battery module.

[0055] Preferably, in step 3, the balance control target is:

[0056] The first objective function is formulated to speed up the equilibrium as follows:

[0057] F1=u max -u(t),u min <u(t)<u max

[0058] Among them, select the maximum balancing current u max =10, minimum balancing current u min =8;

[0059] A second objective function is formulated to measure the degree of equilibrium consistency and normalized as follows:

[0060] F2=[Lx(t)]

[0061]

[0062] The third objective function is formulated to balance the energy loss of the system as follows:

[0063]

[0064] Where u is the current flowing through the inductor, L k is the inductance of the inductor, T s is the sampling time, R loss is the loss resistance, which indicates the equivalent resistance value due to loss in the circuit; u i_pmax and u i_rms They represent the peak current and effective balancing current of the ith single cell connected to the primary winding of the transformer respectively; u i_smax Represents the peak current of the secondary winding of the transformer; V cell(max) Indicates the terminal voltage of the energy-rich single cell; L p and L s Respectively represent the inductance of the primary and secondary sides of the transformer; V cell(i) represents the voltage of the ith battery.

[0065] Preferably, in step 3, the sliding mode controller is as follows:

[0066]

[0067] ε1, ε2, ε3, k1, k2, k3 are the adjustment parameters of the controller, s i represents the sliding surface, sat() represents the saturation function, at time k, u1(k), u2(k), u3(k) represent the sliding mode controller of the system, corresponding to the balancing speed, battery pack consistency and energy consumption in the balancing process respectively; by assigning different weight coefficients α1, α2, α3 to u1(k), u2(k), u3(k), a multi-objective sliding mode controller is constructed:

[0068] u(k)=α1u1(k)+α2u2(k)+α3u3(k)

[0069] st

[0070] 0≤α1≤1

[0071] 0≤α2≤1

[0072] 0≤α3≤1

[0073] α1+α2+α3=1.

[0074] The present invention provides a lithium ion battery pack equalization system, which adopts the above-mentioned lithium ion battery pack equalization method.

[0075] In order to solve the problems faced by the active balancing system of lithium-ion battery packs, such as limited scalability, high cost, and poor balancing effect under complex unbalanced conditions, the present invention proposes a new balancing structure based on a flyback transformer and a switch matrix. This design effectively reduces the number of components and realizes the balancing of long series battery packs. On this basis, a multi-objective sliding mode balancing controller is designed by comprehensively considering the balancing speed, battery consistency, and energy loss during the balancing process. The controller can dynamically adjust the balancing current according to the unbalanced state of the battery pack, thereby comprehensively optimizing the system performance. The adaptability of the sliding mode controller allows different weight coefficients to be assigned to different objective functions, further improving the balancing efficiency of the battery pack. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 is a flow chart of a method for balancing a lithium-ion battery pack in an embodiment;

[0077] Figure 2 It is a schematic diagram of an improved balanced topology structure in an embodiment;

[0078] Figure 3 A schematic diagram of a circuit for managing six single cells in an embodiment;

[0079] FIG4( a ) is a schematic diagram of the energy transfer model of C2P in an embodiment;

[0080] FIG4( b ) is a schematic diagram of the energy transfer model of P2C in the embodiment; Figure 5 Schematic diagram of the structure of a 3-input, 1-output sliding mode controller based on the model in the embodiment. DETAILED DESCRIPTION

[0081] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0082] Example

[0083] like Figure 1 As shown, this embodiment provides a method for equalizing a lithium-ion battery pack, which includes the following steps:

[0084] 1. Improve the balanced topology structure;

[0085] 2. Establish a mathematical model of the equilibrium system;

[0086] 3. Formulate balanced control strategies and balanced control objectives;

[0087] 4. Design a multi-objective sliding mode equilibrium controller to adjust the equilibrium system.

[0088] Improved balanced topology

[0089] like Figure 2 As shown in FIG. 1 , the improved balancing topology includes n series-connected single cells, n / 3 switch matrices, n MOSFETs, and n / 3 flyback transformers. Every three consecutive cells (e.g., BAT1-BAT3) form a group and are connected to the primary winding of the flyback transformer through the switch matrix. In a given cycle, the switch matrix only allows one cell to be connected to the primary side of the flyback transformer. The secondary winding of the flyback transformer is connected to six consecutive cells (e.g., BAT1-BAT6) through MOSFETs. This topology facilitates energy transfer (C2P / P2C) between battery cells and battery packs to achieve interleaved cascade balancing of long series battery packs.

[0090] Take a balancing module as an example. Figure 3 The circuit shown is used to manage six single cells. In this circuit, Q p Represents the transformer primary switch control signal, connected to BATi, and Q s It is used as the secondary switch control signal of the transformer to facilitate the battery connection from BATi to BAT(i+5). a , R c and D a Together they form an RCD clamping circuit that is deployed in the primary switch matrix of the transformer to mitigate overvoltage risks. b , R d and D b Plays a similar role. a and R b Represent the equivalent resistance of the primary and secondary sides respectively.

[0091] When analyzing the energy flow from a single battery to a battery module, BAT1 is taken as an example. Assuming that the average SOC of BAT1 to BAT3 is higher than that of BAT4 to BAT6, and that BAT1 has the highest energy, discharge balancing is performed within the range of BAT1 to BAT3 to enhance the energy consistency between modules. This process turns on Qp1, so that BAT1 forms a loop with the primary side of the transformer, causing the primary current to rise. When the current reaches the maximum value Ipmax, Qp1 is turned off and Qs1 is turned on at the same time. The energy temporarily stored on the primary side of the transformer will be transferred to the secondary side, at which point the energy on the secondary side of the transformer charges the battery module. Subsequently, the secondary current gradually decreases from Ismax to 0, at which point Qs1 is turned off and Qp1 is reconnected. By repeating the cycle, the excess energy in BAT1 is effectively redistributed between BAT1 and BAT6. A schematic diagram of a single cycle is given in Figure 4(a).

[0092] Next, we take BAT2 as an example for analysis. Consider a situation where the average SOC of BAT1 to BAT3 is lower than that of BAT4 to BAT6. It is worth noting that in the subset of BAT1 to BAT3, BAT2 has the lowest energy level. In order to enhance the energy consistency within the module, a basic control strategy is adopted to ensure the charging balance of BAT2. First, Qs1 is turned on to form a closed loop involving BAT1 to BAT6 and the secondary side of the transformer, causing the secondary current to gradually increase. Once the current reaches the maximum value (Ismax), Qs1 is turned off and Qp1 is turned on at the same time. This allows the energy temporarily stored on the secondary side of the transformer to be transferred to the primary side, thereby charging BAT2. Subsequently, the primary current gradually decreases from Ipmax to 0. Finally, Qp1 is turned off and Qs1 is turned on again. By repeating the cycle, this process achieves charging balance for BAT2. As shown in Figure 4(b).

[0093] Establishing a mathematical model of the equilibrium system

[0094] The method to establish the mathematical model of the equilibrium system is:

[0095]

[0096] in, represents the change vector of SOC at time t, and u(t) represents the equilibrium current vector;

[0097] By transforming the above formula, the relationship between the balancing variable and the balancing current of each single cell is obtained:

[0098]

[0099] Among them, x(t) is the SOC vector at time t, Q x is an n×n diagonal matrix defined as the total capacity of the battery pack, where c1…c n Respectively represent the battery BAT1 ~ BAT n The actual maximum capacity of , τ is the time variable;

[0100]

[0101] T represents the energy transfer matrix, where k1=1 / 6, k2=(1 / 6)-1, k3=(1 / 3)-1, k4=1 / 3;

[0102]

[0103] The switch matrix S plays a key role in regulating the connectivity between each single cell and the flyback transformer. If the i-th single cell is connected to the flyback transformer, the corresponding element Si in the switch matrix is ​​set to 1; on the contrary, when the two are not connected, the element is set to 0;

[0104] S=diag(S1,S2,S3…S n )∈R n×n

[0105] u e (t) represents the current of the entire battery pack during the charging or discharging process in the balancing stage. During the charging cycle, u e (t) takes a positive value; during the discharge cycle, u e (t) takes a negative value; during the static period, u e (t) is always set to zero.

[0106] SOC is a comprehensive indicator that includes the voltage and internal characteristics of the battery. Its value represents the ratio between the current battery capacity and the actual maximum capacity, and its value range is between [0,1]. Compared with using open circuit voltage as the balancing variable, using SOC as the balancing variable can eliminate the impact of the maximum capacity difference between single cells. This feature enables each battery to reach the full charge or discharge cut-off state at the same time, thereby improving the overall energy utilization efficiency of the battery pack. Therefore, SOC is defined as the balancing variable in the balancing system.

[0107] Balance control strategy

[0108] (1) Predefine the threshold δ of the mean square error MSE of the SOC in the battery module 2 th; when MSE exceeds the set threshold δ 2 At th, the system starts balancing;

[0109] (2) Consider three consecutive batteries as a battery group. For each battery group, determine the maximum SOC single battery SOC max(i) and minimum SOC single cell SOC min(i) ;

[0110] (3) After starting the balance, the specific balance strategy is as follows:

[0111] a. The battery group is i≤n-1, and the status is:

[0112] soc avg(i) <soc avg(i+1) And soc avg -soc min(i) >0

[0113] The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group;

[0114] b. The battery group is i≤n-1, and the status is:

[0115] soc avg(i) <soc avg(i+1) And soc avg -soc min(i) ≤0

[0116] The direction of energy transfer is: the ith battery group does not start balancing;

[0117] c. The battery group is i≤n-1, and the status is:

[0118] soc avg(i) >soc avg(i+1) And soc max(i) -soc avg >0

[0119] The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group;

[0120] d. The battery group is i≤n-1, and the status is:

[0121] soc avg(i) >soc avg(i+1) And soc max(i) -soc avg ≤0

[0122] The direction of energy transfer is: the ith battery group does not start balancing;

[0123] e. The battery group is i=n, ​​and the status is:

[0124] soc max(i) -soc avg >0 and soc max(i) -soc avg(i) >soc avg(i) -soc min(i)

[0125] The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group;

[0126] f. The battery group is i=n, ​​and the status is:

[0127] soc avg -soc min(i) >0 and soc avg(i) -soc min(i) >soc max(i) -soc avg(i)

[0128] The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group;

[0129] g. The battery group is i=n, ​​the state is: else, and the energy transfer direction is: the ith battery group does not start balancing;

[0130] As shown in Table 1:

[0131] Table 1 Inter-group connection balance control strategy table

[0132]

[0133] (4) Estimate the SOH of each battery and obtain the aging degree SOH of each battery i , when the average SOH of the battery pack avg -SOH i >0.5%, the balancing system will not balance the i-th battery to avoid aggravating the aging degree of the i-th battery; when the average SOH of the battery pack avg -SOH i ≤0.5%, the equilibrium system executes the strategy of step (3).

[0134] This embodiment proposes a component-connected balancing strategy, which aims to increase the balancing speed by incorporating as many batteries as possible into the balancing process.

[0135] Among them, the comprehensive indicator SOC is set as the balancing variable, and the average SOC of each battery pack is calculated as follows:

[0136]

[0137] Among them, SOC j Indicates the SOC value of the jth single cell;

[0138] The average SOC of the entire battery module is expressed as:

[0139]

[0140] n represents the number of single cells in the battery module.

[0141] Although the above balancing strategies are suitable for balancing long series battery packs, the operating conditions of electric vehicles are usually complex and dynamic. Under different environmental conditions, the relative priorities of different balancing objectives may change. Relying solely on the above balancing strategies to balance long series battery packs cannot always achieve the desired effect. Therefore, it is necessary to improve the balancing performance of the system by designing a multi-objective sliding mode controller.

[0142] Balanced control target

[0143] The balancing of a battery pack is a multifaceted dynamic process. In this example, we take a comprehensive approach to design a balancing controller that considers the balancing speed, the consistency of the battery pack, and the energy consumption during the balancing process. The subsequent analysis will be carried out in a systematic order.

[0144] First, the balancing speed is a key indicator to measure the balancing effect. In the case of an unbalanced battery pack, the most important goal is to quickly restore the individual batteries in the battery pack to a balanced state. This requires the battery balancing system to provide the maximum achievable balancing current within the specified range. Therefore, the first objective function is formulated to speed up the balancing speed as shown below:

[0145] F1=u max -u(t),u min <u(t)<u max

[0146] Among them, select the maximum balancing current u max =10, minimum balancing current u min =8.

[0147] Secondly, the main goal of the balancing system is to reduce the differences between the individual cells in the battery pack and ensure that they remain as consistent as possible during their service life. In the balancing control system that uses battery SOC as the balancing variable, the state of charge (SOC i ) and SOC avg The internal consistency of the battery pack is evaluated by calculating the difference between the two or the mean square error of the battery pack SOC. When the inconsistency within the series battery pack is small or close to the end of charging (discharging), it is recommended to use a relatively low balancing current to avoid over-balancing. On the contrary, in the case of large battery inconsistency or stable operating conditions, it is recommended to adjust the balancing current to a higher level. The goal is to enable the balancing current to dynamically adapt according to the state of the battery pack during the balancing process. In summary, a second objective function is formulated to measure the degree of balancing consistency and normalized as follows:

[0148] F2=[Lx(t)]

[0149]

[0150] Finally, when building a balancing system, it is necessary not only to consider the speed of system balancing and the consistency of the battery pack, but also the energy consumption generated during the entire balancing process. Although minimizing energy loss is a very important balancing goal, overemphasizing this goal may prolong the balancing process. In addition, during the balancing process, changes in the SOC of the single cells will cause their output voltages to change accordingly. Even if the balancing current remains unchanged, the energy loss per unit time will fluctuate with changes in the battery voltage. Therefore, the balancing process requires adjustments to the balancing current to effectively solve the problem of energy loss. Therefore, a third objective function is formulated to weigh the energy loss of the balancing system, as shown below:

[0151]

[0152]

[0153] Where u is the current flowing through the inductor, L k is the inductance of the inductor, T s is the sampling time, R loss is the loss resistance, which indicates the equivalent resistance value due to loss in the circuit; u i_pmax and u i_rms They represent the peak current and effective balancing current of the ith single cell connected to the primary winding of the transformer respectively; u i_smax Represents the peak current of the secondary winding of the transformer; V cell(max) Indicates the terminal voltage of the energy-rich single cell; L p and L s Respectively represent the inductance of the primary and secondary sides of the transformer; V cell(i) represents the voltage of the ith battery.

[0154] Model-Based Multi-Objective Sliding Mode Controller

[0155] The goal of traditional balancing systems when designing balancing topology and balancing strategy is to optimize the convergence speed of balancing variables in the balancing process. However, existing designs are insufficient in comprehensively considering key factors such as balancing speed, consistency, and energy loss. Therefore, this embodiment introduces multi-objective sliding mode control (SMC) to adjust the balancing system to ensure that the entire system can remain stable when certain parameters are disturbed.

[0156] Figure 5 The structure of a model-based 3-input, 1-output sliding mode controller is presented.

[0157] 1) In order to facilitate the solution by computer, the forward Euler method must be used to discretize the energy flow model. Finally, the following expression is obtained:

[0158] x(k+1)=x(k)+T S QX -1 [TSu(k)+u e (k)]

[0159] st

[0160] u min ≤u(k)≤u max

[0161] 0≤x(k)≤I

[0162] Wherein, Ts represents the sampling time. In this embodiment, we set it to a constant value of 2.

[0163] 2) When designing a sliding mode controller, the main consideration is the construction of the sliding surface. The sliding surface is usually expressed in s i Represented, and follows a standardized form, usually described as:

[0164]

[0165] Where λ is a matrix related to the system, and its size depends on the dimension of the system. r represents relativity (in this embodiment, its value is set to 1, that is, s i =e i ), where (i=1,2,3). e i Represents the tracking error of the control target.

[0166] 3) The objective functions are expressed as F1, F2, and F3 respectively. Therefore, the control target tracking error e in this embodiment is i The definition is as follows:

[0167]

[0168] 4) In the above equation, F 1,ref 、F 2,ref and F 3,ref In this embodiment, the control target reference value F 1,ref 、F 2,ref and F 3,ref are set to 0, 0, 0 respectively. Therefore, the sliding surface expression of the system is as follows:

[0169]

[0170] 5) The sliding mode reaching law is selected as:

[0171]

[0172] st

[0173] ε1,ε2,ε3>0

[0174] k1,k2,k3>0

[0175] Among them, ε1, ε2, ε3, k1, k2, and k3 are the adjustment parameters of the controller.

[0176] 6) Discretize the above equation and get the exponential reaching law as follows:

[0177]

[0178] st

[0179] ε1,ε2,ε3>0

[0180] k1,k2,k3>0

[0181] To meet the convergence conditions, the following requirements need to be met when selecting parameters:

[0182]

[0183] Therefore, we can get:

[0184]

[0185] 7) Substituting the sliding surface into the reaching law, we get the following results:

[0186]

[0187] The sliding mode control rate can be obtained as follows:

[0188]

[0189] However, using the exponential reaching law to determine the controller u will cause the system to jitter. In order to alleviate this problem, the reaching law is improved, that is, in the structure of the exponential reaching law, the saturation function sat(s(t)) is used to replace the sign function sgn(s(t)).

[0190]

[0191] This method introduces a boundary layer, applies continuous control within the boundary layer, and implements normal sliding mode control outside the boundary layer. This design effectively eliminates the adverse effects of jitter. Here, △ represents the boundary of the established boundary layer, and the thickness of the boundary layer can be adjusted to alleviate the jitter phenomenon caused by discontinuity. For this system, a specific value of 0.005 is selected to achieve this purpose.

[0192] 8) Therefore, the final sliding mode controller is as follows:

[0193]

[0194] Using the exponential reaching law to determine the controller will cause the system to jitter. In order to alleviate this problem, the reaching law is improved in the above formula, that is, in the structure of the exponential reaching law, the saturation function sat(s(t)) is used to replace the sign function sgn(s(t)).

[0195] At time k, u1(k), u2(k), and u3(k) represent the sliding mode controllers of the system, corresponding to the balancing speed, battery pack consistency, and energy consumption during the balancing process, respectively; in the long series battery pack balancing system, we expect the values ​​output by these controllers to converge at time k. However, due to different balancing objectives, these controllers may not tend to be equal at a specific time point. Therefore, by giving different weight coefficients α1, α2, and α3 to u1(k), u2(k), and u3(k), a multi-objective sliding mode controller is constructed:

[0196] u(k)=α1u1(k)+α2u2(k)+α3u3(k)

[0197] st

[0198] 0≤α1≤1

[0199] 0≤α2≤1

[0200] 0≤α3≤1

[0201] α1+α2+α3=1

[0202] This embodiment also provides a lithium-ion battery pack equalization system, which adopts the above-mentioned lithium-ion battery pack equalization method.

[0203] Simulation and comparison

[0204] This simulation uses 12 single cells connected in series as the research. The battery module consists of 4 battery groups, each of which contains 3 single cells. The system is modeled and simulated using MATLAB / Simulink 2023b. The simulation parameters of the battery and balancing circuit are detailed in Table 2. In order to speed up the simulation, the nominal capacity is set to 10% of the rated capacity of the single cell, while other parameters remain unchanged. In the balancing system, the initial SOC of each single cell is set to [60, 62, 42, 48, 64, 76, 84, 62, 80, 78, 56, 70]%. The weight coefficient ratio of this embodiment is used to indicate the importance of the system to a specific balancing target: the values ​​of α1, α2, and α3 respectively represent the weights of the balancing speed, the consistency of the battery pack SOC, and the energy consumption during the balancing process. To verify the balancing effect, four different weight coefficient combinations were selected: α1 = 1, α2 = 0, α3 = 0 (fast balancing target - FBT), α1 = 0, α2 = 1, α3 = 0 (improved consistency target - ICT), α1 = 0, α2 = 0, α3 = 1 (minimum energy consumption target - MECT), and α1 = 0.36, α2 = 0.48, α3 = 0.16 (comprehensive balancing target - CBT). The simulation experiment was carried out in the static state of the battery pack. When the mean square error (δ 2 When (tf))≤3.2%, the balancing system stops balancing.

[0205] Table 2 Parameters of single battery and balancing circuit components

[0206]

[0207]

[0208] The quantitative comparison of the experimental results is shown in Table 3.

[0209] Table 3 Comparison of active equalization performance

[0210] method Balanced deadline (s) SOC after equilibrium (%) MSE(%) Energy loss (J) FBT 88.3 59.084 3.15 363.022 ICT 138.5 58.8661 2.39 348.426 MECT 164.9 59.8818 3.05 311.937 CBT 124.5 59.1355 2.97 323.218 FLC 177.4 57.9907 3.76 383.665

[0211] After analyzing the results, it can be concluded that emphasizing FBT or MECT will cause the output balancing current of the balancing system to reach the maximum and minimum boundary values ​​of the set balancing current, respectively. When FBT is given priority, the shortest balancing time can be achieved, but the energy loss is high due to the excessive balancing current. On the contrary, when MECT is considered, although the energy loss is minimal, the required time is the longest. When only ICT is considered, the mean square error (MSE) of the battery pack SOC is the smallest when the balancing is completed, but this comes at the expense of balancing time and energy loss during the balancing process. However, when all three indicators are considered comprehensively, higher consistency is obtained when the balancing is completed compared with FBT and MECT. In addition, CBT is superior to ICT in terms of balancing time and energy consumption indicators. The comprehensive comparison shows that CBT is the best choice because it considers balancing speed, battery pack consistency, and energy efficiency during the balancing process. In addition, the comparative analysis in Table III shows that the proposed battery pack balancing method is superior to the fuzzy logic control (FLC) method in terms of balancing time, consistency of battery pack SOC after balancing, and energy loss.

[0212] In this embodiment, a bidirectional balancing circuit with a flyback transformer and a switch matrix is ​​introduced to alleviate the energy imbalance between single cells in an electric vehicle battery pack. The topology of the circuit adopts a switch matrix to minimize the number of components, thereby reducing cost and space requirements. In addition, this embodiment also utilizes a grouping and cascading method to promote mutual energy transfer within a series battery pack. In addition, this embodiment proposes a multi-objective active balancing controller for a C2P / P2C balancing system. The controller adopts sliding mode control, which comprehensively considers the balancing speed, battery consistency and energy consumption. It dynamically adjusts the balancing current according to different operating conditions and the unbalanced state of the battery pack within a predefined range. Simulation results show that our multi-objective balancing controller has superior performance compared with the FLC method. By adjusting the weighting coefficient of each objective function, the controller effectively manages the relationship between the balancing speed, battery consistency and energy efficiency. Therefore, the method of this embodiment not only improves the cycle life of the battery pack, but also optimizes the energy absorption and release during use.

[0213] The present invention and its embodiments are described schematically above, and the description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it and designs a structural method and an embodiment similar to the technical solution without creativity without departing from the purpose of the invention, they shall all fall within the protection scope of the present invention.

Claims

1. Equalization method of lithium-ion battery pack, Features: The following steps are involved:

1. Improve the balanced topology structure; 2. Establish a mathematical model of the equilibrium system; 3. Formulate balanced control strategies and balanced control objectives; 4. Design a multi-objective sliding mode equilibrium controller to adjust the equilibrium system.

2. The equalization method of a lithium-ion battery pack according to claim 1, Features: In step 1, the improved balancing topology includes n series-connected single cells, n / 3 switch matrices, n MOSFETs and n / 3 flyback transformers. Every three consecutive cells form a group and are connected to the primary winding of the flyback transformer through the switch matrix. The secondary winding of the flyback transformer is connected to six consecutive cells through MOSFETs.

3. The equalization method of a lithium-ion battery pack according to claim 2, Features: In step 2, the method for establishing the mathematical model of the equilibrium system is: in, represents the change vector of SOC at time t, and u(t) represents the equilibrium current vector; By transforming the above formula, the relationship between the balancing variable and the balancing current of each single cell is obtained: Among them, x(t) is the SOC vector at time t, Q x is an n×n diagonal matrix defined as the total capacity of the battery pack, where c 1 …c n Respectively represent battery BAT 1 ~BAT n The actual maximum capacity of , τ is the time variable; T represents the energy transfer matrix, where k 1 =1 / 6, k 2 =(1 / 6)-1, k 3 =(1 / 3)-1, k 4 =1 / 3; The switch matrix S plays a key role in regulating the connectivity between each single cell and the flyback transformer. If the i-th single cell is connected to the flyback transformer, the corresponding element Si in the switch matrix is ​​set to 1; on the contrary, when the two are not connected, the element is set to 0; S=diag(S 1 ,S 2 ,S 3 …S n )∈R n×n u e (t) represents the current of the entire battery pack during the charging or discharging process in the balancing stage. During the charging cycle, u e (t) takes a positive value; during the discharge cycle, u e (t) takes a negative value; during the static period, u e (t) is always set to zero.

4. The equalization method of a lithium-ion battery pack according to claim 3, Features: In step three, the balance control strategy is: (1) Predefine the threshold δ of the mean square error MSE of the SOC in the battery module 2 th; when MSE exceeds the set threshold δ 2 At th, the system starts balancing; (2) Consider three consecutive batteries as a battery group. For each battery group, determine the maximum SOC single battery SOC max(i) and minimum SOC single cell SOC min(i) ; (3) After starting the balance, the specific balance strategy is as follows: a. The battery group is i≤n-1, and the status is: soc avg(i) <soc avg(i+1) and soc avg -soc min(i) >0 The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group; b. The battery group is i≤n-1, and the status is: soc avg(i) <soc avg(i+1) and soc avg -soc min(i) ≤0 The direction of energy transfer is: the ith battery group does not start balancing; c. The battery group is i≤n-1, and the status is: soc avg(i) > soc avg(i+1) and soc max(i) -soc avg > 0 The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group; d. The battery group is i≤n-1, and the status is: soc avg(i) > soc avg(i+1) and soc max(i) -soc avg ≤ 0 The direction of energy transfer is: the ith battery group does not start balancing; e. The battery group is i=n, ​​and the status is: soc max(i) -soc avg > 0 and soc max(i) -soc avg(i) > soc avg(i) -soc min(i) The direction of energy transfer is: discharging and balancing the single battery with the maximum SOC in the i-th battery group; f. The battery group is i=n, ​​and the status is: soc avg -soc min(i) > 0 and soc avg(i) -soc min(i) > soc max(i) -soc avg(i) The direction of energy transfer is: charging and balancing the single battery with the minimum SOC in the i-th battery group; g. The battery group is i=n, ​​the state is: else, and the energy transfer direction is: the ith battery group does not start balancing; (4) Estimate the SOH of each battery and obtain the aging degree SOH of each battery i , when the average SOH of the battery pack avg -SOH i >0.5%, the balancing system will not balance the i-th battery to avoid aggravating the aging degree of the i-th battery; when the average SOH of the battery pack avg -SOH i ≤0.5%, the equilibrium system executes the strategy of step (3); Among them, the comprehensive indicator SOC is set as the balancing variable, and the average SOC of each battery pack is calculated as follows: Among them, SOC j Indicates the SOC value of the jth single cell; The average SOC of the entire battery module is expressed as: n represents the number of single cells in the battery module.

5. The equalization method of a lithium-ion battery pack according to claim 4, Features: In step 3, the equilibrium control target is: The first objective function is formulated to speed up the equilibrium as follows: F 1 =u max -u(t),u min <u(t)<u max Among them, select the maximum balancing current u max =10, minimum balancing current u min =8; A second objective function is formulated to measure the degree of equilibrium consistency and normalized as follows: F 2 =[Lx(t)] The third objective function is formulated to balance the energy loss of the system as follows: Where u is the current flowing through the inductor, L k is the inductance of the inductor, T s is the sampling time, R loss is the loss resistance, which indicates the equivalent resistance value due to loss in the circuit; u i_pmax and u i_rms They represent the peak current and effective balancing current of the ith single cell connected to the primary winding of the transformer respectively; u ismax Represents the peak current of the secondary winding of the transformer; V cell(max) Indicates the terminal voltage of the energy-rich single cell; L p and L s Respectively represent the inductance of the primary and secondary sides of the transformer; V cell(i) represents the voltage of the ith battery.

6. The equalization method of a lithium-ion battery pack according to claim 5, Features: In step 3, the sliding mode controller is as follows: ε 1 , ε 2 , ε 3 , k 1 , k 2 , k 3 is the adjustment parameter of the controller, s i represents the sliding surface, sat() represents the saturation function, at time k, u 1 (k),u 2 (k),u 3 (k) represents the sliding mode controller of the system, corresponding to the balancing speed, battery pack consistency and energy consumption during the balancing process; by giving u 1 (k),u 2 (k),u 3 (k) Assign different weight coefficients to α 1 , α 2 , α 3 , construct a multi-objective sliding mode controller: u(k)=a 1 you 1 (k)+a 2 you 2 (k)+a 3 you 3 (k) st 0≤α 1 ≤1 0≤α 2 ≤1 0≤α 3 ≤1 α 1 +α 2 +α 3 ＝1。 7.Balance system for lithium-ion battery packs, Features: It adopts the equalization method of the lithium-ion battery pack as described in any one of claims 1-6.

Images