An active balancing control method for lithium battery pack based on dual variables of SOC and voltage difference

Abstract

The present invention discloses an active balancing control method for a lithium battery pack based on dual variables of SOC and voltage difference, including: step 1, establishing an equivalent circuit model and parameter identification, simulating the dynamic characteristics of a lithium battery, using a single lithium battery first-order RC equivalent circuit model as the basic model of the lithium battery to analyze the characteristic parameters of the lithium battery, and identifying the parameters corresponding to each lithium battery model by combining a nonlinear least squares fitting algorithm with a pulse discharge voltage change curve of the lithium battery; step 2, building an active balancing topology structure, designing a bidirectional active balancing circuit based on inductance based on a Buck‑Boost balancing circuit; step 3, constructing a balancing control strategy for dual target variables, and completing the voltage and SOC balancing control of a series lithium battery pack through SOC estimation and voltage correction according to the OCV‑SOC relationship curve of the lithium battery. The active balancing topology circuit and the corresponding balancing control strategy constructed by the present invention are relatively efficient in improving the inconsistency between lithium batteries.

Description

Technical Field

[0001] The present invention belongs to the technical field of lithium battery pack charge equalization, and mainly relates to a lithium battery pack active equalization control method based on dual variables of SOC and voltage difference. Background Art

[0002] As the demand for power lithium batteries in the electric vehicle industry continues to increase, when lithium batteries are connected in series, due to factors such as the structure of each lithium battery and the use environment, the performance parameters such as voltage, capacity, and SOC (State of Charge) between lithium batteries in the lithium battery pack will be inconsistent, resulting in reduced energy utilization, reduced life, and a certain impact on driving safety. Therefore, it is necessary to reduce the inconsistency of lithium batteries to prevent them from overcharging or over-discharging, and balancing technology is an effective way to reduce the deterioration rate of such problems. The main parts of balancing technology are balancing topology and balancing control strategy. The optimization of balancing technology can reduce energy consumption, improve the balancing efficiency and safety of lithium battery systems. In terms of topological circuit structure, the simplicity of topological structure, the difficulty of control, and applicability are factors that need to be considered in research. In terms of control algorithm, factors such as accuracy, stability, and fit need to be considered.

[0003] At present, although passive balancing has a simple structure, it has problems such as high energy loss and low accuracy. The active balancing topology based on high-frequency switching technology transfers electricity from high-energy lithium batteries to low-energy lithium batteries through energy storage elements, with less energy dissipation and higher balancing efficiency. The first type of maximum value method commonly used in the current balancing control strategy is to transfer part of the energy of the lithium battery with the highest voltage to the lithium battery with the lowest voltage; the second type of average and difference comparison method uses the voltage or charge average in the series lithium battery group as a reference, compares the size of each lithium battery with the average, and then discharges the lithium battery with higher voltage or charge, and adopts charge balancing for the lithium battery with lower voltage.

[0004] In order to achieve energy balancing of lithium battery packs with higher efficiency, the present invention proposes to construct an active balancing topology based on inductive energy bidirectional transfer and a balancing control strategy based on adjacent SOC difference and voltage difference dual variables. Compared with the traditional balancing circuit that transmits energy between adjacent lithium batteries, this topology can transfer energy to the lithium battery that needs to be balanced, and improve the balancing speed and balancing efficiency of the SOC and voltage of the series-connected lithium battery pack, thereby improving its charging and discharging performance, reducing the inconsistency of the lithium battery, and extending its service life. Summary of the invention

[0005] In order to improve the problem of energy consistency of each lithium battery in the working process of the lithium battery pack, improve the energy utilization rate of the entire lithium battery pack, extend the service life of the lithium battery, and ensure its safety performance, the present invention provides a lithium battery pack active balancing control method based on the dual variables of SOC and voltage difference. The balancing control strategy uses the adjacent SOC value and voltage value in the series lithium-ion lithium battery pack as the dual balancing target variables, and adopts the active balancing topology of energy bidirectional flow based on inductance to balance the SOC and voltage of each lithium battery in the lithium battery pack. This structure increases the energy transfer loop, effectively reduces the balancing time, and the circuit volume is small, which can be applied to large-scale lithium battery packs.

[0006] The technical solution of the present invention is:

[0007] A lithium battery pack active balancing control method based on SOC and voltage difference dual variables includes the following steps:

[0008] Step 1, establish an equivalent circuit model and parameter identification, simulate the dynamic characteristics of lithium batteries, use the first-order RC equivalent circuit model of a single lithium battery as the basic model of the lithium battery to analyze the characteristic parameters of the lithium battery, and use the nonlinear least squares fitting algorithm combined with the pulse discharge voltage change curve of the lithium battery to identify the parameters corresponding to each lithium battery model;

[0009] Step 2: Build an active balancing topology structure and design an inductor-based bidirectional active balancing circuit based on the Buck-Boost balancing circuit;

[0010] Step 3, construct a dual-objective variable balancing control strategy. According to the OCV-SOC relationship curve of the lithium battery, when the SOC is in the two stages of 0-10% and 90%-100%, select voltage as the balancing variable, and when it is in the stage of 10%-90%, select SOC as the balancing variable. The voltage and SOC balancing control of the series lithium battery group is completed by SOC estimation and voltage correction.

[0011] Preferably, in step 1, the first-order RC equivalent circuit model of the single lithium battery includes a voltage E m , resistor R 0 , resistor R 1 and capacitor C 1 ; Lithium batteries alternate between charge and discharge cycles, and the state of charge of the lithium battery model is estimated by measuring the current, voltage and temperature of the lithium battery; The discrete state space equation of a single lithium battery is:

[0012]

[0013] In the formula, C 1 (SOC t ,T) tis the polarization capacitor of the lithium battery, U1 t Represents the polarization voltage of the lithium battery at time t, C q is the rated capacity of the lithium battery, T s is the sampling period, W t is the process noise at time t, then:

[0014] U t =E m (SOC t ,T) t -U1 t -IR 0 (SOC t ,T) t +V t (2)

[0015] Where T represents the internal temperature of the lithium battery, U t and E m (SOC t ,T) t Respectively represent the terminal voltage of the lithium battery at time t and the estimated open circuit voltage OCV, R 0 (SOC t ,T) t and R 1 (SOC t ,T) t Respectively represent the internal resistance and polarization internal resistance of the lithium battery at time t, V t represents the measurement noise of the lithium battery at time t;

[0016] According to the identified parameters of the first-order RC equivalent circuit model of the single lithium battery, a first-order RC equivalent circuit Simulink model of the lithium battery is constructed; at the TE1 temperature, the parameters of the first-order RC equivalent circuit Simulink model of the lithium battery are identified; in Matlab / Simulink, the nonlinear least squares fitting algorithm is combined with the pulse discharge voltage change curve of the lithium battery to calculate the parameter value voltage E corresponding to each lithium battery model. m , resistor R 0 , resistor R 1 To identify.

[0017] Preferably, in step 2, the bidirectional active balancing circuit includes a lithium battery pack and a balancing primary and secondary loop. Each lithium battery in the lithium battery pack is connected to an inductor L and four MOS tubes to form a bidirectional energy transfer circuit. Each MOS tube is connected in parallel with a freewheeling diode. After the lithium battery is over-discharged, the protection function will be turned on and the MOS tube will be turned off. When in the charging state, the body diode of the MOS tube is used to make the circuit conductive and the system works. The main loop MOS tube Q1m (m=1,2,...,14) controls the charging circuit of the lithium battery, and the secondary loop MOS tube Q2n (n=10,11,...,124) controls the discharge circuit of the lithium battery. After a certain period of duty cycle, the MOS tube can quickly switch between the two states of conduction and shutdown. The energy transfer between each lithium battery is completed by controlling the on and off of the corresponding MOS tube. During the charging and discharging process, the energy is transferred from the lithium battery to the energy storage inductor through the main loop mode, and the energy is transferred from the inductor to the low-energy lithium battery through the main loop and the secondary loop combination mode.

[0018] Preferably, in step 3, in the stage of constructing the balancing control strategy of the dual objective variables, the difference between the SOC values ​​of adjacent lithium batteries is Δn, the threshold value of the difference between adjacent charges is δn, the SOC values ​​of each lithium battery are arranged in descending order, Δn>δn is λn (n=1,2,3,4), when the adjacent SOC difference between each lithium battery is within the set threshold range, λn is 0, and outside the threshold range, λn is 1, when λn is 0, there is no need to balance the lithium battery group, and the control has no action; when λ1=1, λ2=0, λ3=0, the lithium batteries in series are balanced. The lithium battery with the largest SOC value in the battery transfers energy to the lithium battery with the second largest SOC value, and controls the direction and size of the balancing current between the lithium batteries by controlling the on and off of the MOS tube and the time, until the difference in charge between adjacent lithium batteries is ≤3%. Similarly, the state combination of λn is determined according to the size relationship between △n and δn. The control system makes corresponding balancing actions according to different states, repeatedly switching between different state combinations until the values ​​of λn are all 0, that is, the states of λn are all 0, indicating that the SOC value balancing between the lithium batteries is completed.

[0019] Preferably, the equilibrium control strategy for constructing dual objective variables includes:

[0020] Step (1), estimating SOC using an unscented Kalman filter (UKF) algorithm and establishing a Simulink model for the estimation; sorting the SOC estimates of the lithium batteries, and then screening out lithium batteries with SOCs exceeding 90% and below 10%;

[0021] Step (2), using voltage as the balancing variable, using the maximum value method to perform voltage correction balancing on each lithium battery in the lithium battery pack, so that the lithium battery voltage is within the range of the set voltage threshold β=10mV;

[0022] Step (3): Taking SOC as the balancing target, the average value and difference comparison method are used for balancing, and the difference between adjacent charges of the lithium battery is used as the balancing criterion.

[0023] Preferably, in step (3), the difference between the SOC values ​​of adjacent lithium batteries is recorded as △n, and the threshold value of the difference between adjacent charges is δn=3%; the SOC balancing time T between each lithium battery is Equ for

[0024]

[0025] In the formula, C q is the capacity of lithium battery, I b is the balancing current;

[0026] The SOC values ​​of the lithium batteries are arranged in descending order, and △n>δn is denoted as λn (n=1, 2, 3, 4). The state combination of λn is determined based on the size relationship between △n and δn. The control system makes corresponding balancing actions based on different states, repeatedly switching between different state combinations until the values ​​of λn are all 0, that is, the states of λn are all 0, indicating that the SOC value balancing between the lithium batteries is completed.

[0027] As a preferred method, it also includes a lithium battery balancing test platform, which includes a driving power module, an electronic load module, an active balancing circuit board, a lithium battery pack module, a paperless recorder, an oscilloscope and a host computer control panel; the driving power module is used to charge each lithium battery in the lithium battery pack; the electronic load module is used to perform a constant current discharge operation on the lithium battery; the active balancing circuit board is used to balance the charge and discharge control of the lithium battery in an unbalanced state; the lithium battery pack module is composed of a number of lithium batteries connected in series; the paperless recorder is used to store and process the collected data; the oscilloscope is used to measure the parameter changes of each lithium battery during operation; the host computer control panel obtains the charging and discharging current, charge state and lithium battery voltage information of the lithium battery pack through serial communication; the test steps of the lithium battery balancing test platform include:

[0028] Step (11), according to the characteristic parameters of the lithium battery, the lithium battery pack is discharged at a constant current of 3C charge and discharge rate to the cut-off voltage of each lithium battery, the capacity and voltage value of each lithium battery are recorded, and then the battery is left to stand for T1 minute;

[0029] Step (12), continue to charge the lithium battery group with 3C current in the order of constant current and constant voltage, record the voltage change and capacity of each lithium battery, and record the voltage comparison change before and after active equalization of the lithium battery in a table;

[0030] Step (13), then continue to discharge at a constant current of 3C to the cut-off voltage, and stand for T2 minutes;

[0031] Step (14), after the discharge is completed, the lithium battery pack is continuously charged with a 3C current. At the same time, at the end of the balanced charging period, that is, when the charge SOC is greater than 0.9, the charging process is started according to the dual-variable balanced control strategy. After the control of this stage is completed, the lithium battery pack is charged to the rated voltage at a constant voltage and then left to stand for T3 minutes;

[0032] Step (15), repeat step (13) and step (14) for N times, and record the status of each lithium battery.

[0033] The beneficial effects of the present invention are:

[0034] 1. The active balancing topology circuit and the corresponding balancing control strategy constructed by the present invention are relatively efficient in improving the inconsistency between lithium batteries. Compared with the traditional balancing scheme, the present invention not only solves the inconsistency problem but also greatly improves the charging time.

[0035] 2. The balancing scheme constructed by the present invention uses a bidirectional switch with low conduction loss, which can realize the bidirectional flow of energy, overcome the system loss caused by the diode in the traditional inductor balancing, take into account the balancing speed and efficiency, and realize the energy transfer between single cells in a multi-loop combination mode, effectively reducing the balancing time.

[0036] 3. The topological circuit of the present invention is small in size and easy to expand. When the number of batteries in the lithium battery pack increases, balancing can be achieved by simply increasing the number of switches. The balancing control strategy logic is simple and feasible, and can be applied to larger-scale lithium battery packs. This control strategy can be used as a battery component in energy industries such as new energy vehicles, wind power generation, and solar power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is the first-order RC equivalent circuit model of the lithium battery of the present invention;

[0038] Figure 2 The first-order RC equivalent circuit Simulink model of the lithium battery of the present invention;

[0039] Figure 3 The voltage variation curve of each lithium battery in the present invention is a dynamic pulse discharge terminal;

[0040] Figure 4 The invention is a bidirectional active balancing circuit based on inductance;

[0041] Figure 5 It is a working principle diagram of each loop mode of the present invention;

[0042] Figure 6 The figure is a relationship diagram between SOC and OCV of the lithium battery of the present invention;

[0043] Figure 7 This is a flow chart of the balance control strategy of the present invention;

[0044] Figure 8 (including Figure 8a -b) is a diagram of a balanced charge and discharge control system for a lithium battery pack according to the present invention;

[0045] Figure 9 (including Figure 9a -d) is a logical Simulink model of the balanced control system of the present invention;

[0046] Fig.10 This is a diagram showing the SOC and voltage changes during the charging simulation process of the present invention;

[0047] Fig.11 The SOC and voltage change diagram of each lithium battery during the discharge process of the present invention;

[0048] Fig.12 It is a SOC variation diagram of the static balancing simulation process under different methods of the present invention;

[0049] Fig.13 The hardware block diagram of the balanced control test system of the present invention;

[0050] Fig.14 It is an active balancing control test platform of the present invention;

[0051] Fig.15 The voltage variation curve of each lithium battery in the active equalization charging process of the present invention;

[0052] Fig.16 The voltage variation curve of each lithium battery during the active balanced discharge static process of the present invention.

[0053] Fig.14 In the middle: 1 is the driving power module, 2 is the electronic load module, 3 is the active balancing circuit board, 4 is the lithium battery pack module, 5 is the paperless recorder, 6 is the oscilloscope, and 7 is the host computer control panel. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings. The description in this part is only exemplary and explanatory and should not have any limiting effect on the protection scope of the present invention.

[0055] Embodiment 1: A lithium battery pack active balancing control method based on dual variables of SOC and voltage difference, comprising the following steps:

[0056] Step 1: Establish equivalent circuit model and parameter identification.

[0057] Simulate the dynamic characteristics of lithium batteries to establish a first-order RC equivalent circuit model of a single lithium battery. Figure 1 As shown, the first-order RC equivalent circuit model of a single lithium battery includes the voltage E m , resistor R 0 , resistor R 1 and capacitor C 1 The lithium battery alternates between charge and discharge cycles. The state of charge of the lithium battery model is estimated by measuring the current, voltage and temperature of the lithium battery. T represents the internal temperature of the lithium battery, and U t and E m (SOC t ,T) t Respectively represent the terminal voltage of the lithium battery at time t and the estimated open circuit voltage OCV, R 0 (SOC t ,T) t and R 1 (SOC t ,T) t They represent the internal resistance and polarization internal resistance of the lithium battery at time t, respectively. 1 (SOC t ,T) t is the polarization capacitor of the lithium battery, U1 t Represents the polarization voltage of the lithium battery at time t. t As the state variable of the lithium battery system, It and U t They represent the input and output respectively. Discrete state space equation of lithium battery

[0058]

[0059] Where U t is the output voltage, C q is the rated capacity of the lithium battery, T s is the sampling period, W t is the process noise at time t

[0060] Then we have:

[0061] U t =E m (SOC t ,T) t -U1 t -IR 0 (SOC t ,T) t +V t (2)

[0062] Where V t represents the measurement noise of the lithium battery at time t.

[0063] according to Figure 1 The constructed lithium battery first-order RC equivalent circuit model and equations (1) and (2) build the first-order RC equivalent circuit model Simulink model of lithium battery as shown in Figure 2 As shown. At TE1 temperature (TE1 = 20°C in this embodiment), the first-order RC model of the lithium battery is parameter identified. Specifically, the lithium battery is pulse-discharged at a current of 1C, and its charge is discharged from 100% to 0 at intervals of 10%. The voltage tends to decrease with intermittent discharge. After each 10% discharge, the lithium battery is left to stand for one hour. The open circuit voltage OCV can be considered as the voltage across the lithium battery after standing. The highest cutoff voltage of each lithium battery is 4.2V. The pulse discharge changes of each lithium battery are shown in the figure below. Figure 3 shown.

[0064] In Matlab / Simulink, the nonlinear least squares fitting algorithm is combined with the pulse discharge voltage change curve of the lithium battery to calculate the parameter value voltage E corresponding to each lithium battery model. m , resistor R 0 , resistor R 1 Identify the lithium battery E at a constant temperature of 20°C m The parameter changes are shown in Table 1. The lithium battery resistance R 0 The parameter changes are shown in Table 2. The lithium battery resistance R 1 The parameter changes are shown in Table 3. Figure 2 As can be seen from the data in Tables 1 to 3, the inconsistency of lithium batteries will lead to differences in lithium battery voltages. The voltage error of the four lithium batteries decreases with the increase of discharge time, reducing by 74%. It can be seen that intermittent discharge significantly reduces the voltage inconsistency between each lithium battery.

[0065] Table 1 20℃ lithium battery voltage E m Parameter changes

[0066]

[0067] Table 2 20℃ Lithium battery resistance R 0 Parameter changes

[0068]

[0069] Table 3 20℃ Lithium battery resistance R 1 Parameter changes

[0070]

[0071] Step 2: Build an active balancing topology.

[0072] Based on the Buck-Boost equalization circuit, a bidirectional active equalization circuit based on inductance is designed, such as Figure 4 The figure shows the inductor-based balancing structure of the present invention, which is mainly composed of a lithium battery pack and a balancing primary and secondary circuit. The current and voltage in the inductor L component are out of phase. Theoretically, the loss of this component is zero. In the balancing module, the inductor L, as an energy storage component, can effectively reduce the ripple, thereby reducing the impact on the entire lithium battery pack. Therefore, in the working principle of the application L, when the current flowing through L is I, the energy E stored in the inductor is:

[0073]

[0074] The present invention transfers energy between lithium batteries by combining a primary circuit with a secondary circuit based on the Buck-Boost energy transfer mode, and has relatively high efficiency, and only requires the addition of a MOS tube and a corresponding freewheeling diode. Figure 4 Bat 1 , Bat 2 Taking the series lithium battery group balancing unit composed of Bat, Bat3, and Bat4 as an example, each lithium battery is connected to four MOS tubes to form a bidirectional energy transfer circuit. Each MOS tube is connected in parallel with a freewheeling diode. After the lithium battery is over-discharged, the protection function will be turned on and the discharge MOS tube will be turned off. When in the charging state, the body diode of the MOS tube is used to make the circuit conductive, and the control system works well. The main circuit switch MOS tube Q1m (m=1,2,…,14) controls the charging circuit of the lithium battery, and the secondary circuit switch MOS tube Q2n (n=10,11,…,124) controls the discharge circuit of the lithium battery. The switch tube can quickly switch between the two states of conduction and shutdown, and complete the energy transfer between each lithium battery by controlling the on and off of the corresponding MOS tube.

[0075] Figure 5 The working principle of the balancing circuit shown in the figure is that during the charging and discharging process, energy is transferred from the lithium battery to the energy storage inductor through the main circuit mode, and from the inductor to the low-energy lithium battery through the main circuit and the secondary circuit combination mode. The energy transfer process during balancing control includes. The final unit of the balancing topology is the lithium battery in the lithium battery pack. The balancing purpose is to keep the SOC difference between the lithium batteries within the threshold range set by the research. If the lithium battery Bat is set 1 The energy is higher than Bat 2 The energy of the two lithium batteries is beyond the set threshold range. At this time, the balancing management system sends a PWM signal to control the on and off of MOS tubes Q11 and Q111. 1 It forms a loop with MOS tube Q11, MOS tube Q111 and inductor L to realize energy transfer, such as Figure 5(a) shows. Lithium battery Bat 1 Part of the energy in the inductor is transferred to the inductor L. After a period of duty cycle, the MOS tube Q11 is turned off, and at the same time the MOS tube Q13 is turned on. At this time, the inductor L, the MOS tube Q111, and the lithium battery Bat 2 , lithium battery Q13 forms the main circuit, such as Figure 5 (b) shows the inductor L, MOS tube Q211, and lithium battery Bat 2 MOS tube Q2112 forms a secondary loop such as Figure 5 (c) The energy stored in the inductor L is transferred to the lithium battery Bat 2 In the process, the primary and secondary circuits are balanced at the same time, which can efficiently complete the lithium battery Bat 1 With lithium battery Bat 2 The energy transfer between the two batteries is repeated over and over again until the SOC difference between each two lithium batteries is controlled within the parameter threshold range, and then the balancing is cut off.

[0076] When 0<t<t 1 When the MOS tube Q11 is turned on, the energy transfer circuit of the lithium battery is as follows: Figure 5 (a) shows the lithium battery Bat 1 Discharge, inductor L stores energy. Figure 5 (d) It can be seen that the inductor current I L Increases linearly, the inductor current I L and lithium battery voltage VBat 1 The correlation between duty cycle D, PWM signal and period T is:

[0077]

[0078] Where, t 2 is the inductor current I L When it drops to 0, Voff represents the voltage across the inductor after the MOS tube is turned off.

[0079] Then, the maximum current of the inductor during the entire cycle is:

[0080]

[0081] In the formula, I L Peak represents the maximum current value of the inductor, V L represents the inductor voltage, D represents the duty cycle, and T is the control signal period.

[0082] When t 1 <t<t 2 When the MOS tube Q11 is disconnected, the energy transfer circuit is as follows: Figure 5As shown in (b) and 5(c), the energy stored in L can not only be transferred to the lithium battery Bat through the body diode of the main circuit MOS tube Q111 2 Energy can be charged through the secondary MOS tube Q211. Figure 5 (d) It can be seen that the current in the inductor L decreases linearly at this time, and the lithium battery Bat 1 Transfer to Bat 2 The energy is

[0083]

[0084] To avoid saturation hysteresis of the inductor L, make the inductor work in the critical current mode, and fully release the stored energy in the inductor L, the duty cycle needs to satisfy

[0085]

[0086] Step 3: Construct a balanced control strategy for dual objective variables.

[0087] Figure 6 In the OCV-SOC relationship curve of the lithium battery shown in the figure, due to the hysteresis phenomenon during the use of the lithium battery, the discharge curve of the lithium battery is always lower than the charging curve, and the OCV changes rapidly when the state of charge SOC is between 0% and 10% and between 90% and 100%. Therefore, if SOC is selected as the single equilibrium variable, when the SOC is low, under large current discharge, the lithium battery terminal voltage rises from the lower limit cut-off voltage value to the rated voltage, and its working voltage may drop suddenly, causing the lithium battery to be over-discharged; when the SOC is high, the lithium battery rises from the rated voltage to the upper limit cut-off voltage value, and the lithium battery with a higher charge is prone to overcharge. When the charge is between 10% and 90%, the OCV change of the lithium battery is relatively small. If only voltage is selected as the equilibrium variable at this time, it is easy to have a small voltage difference between lithium batteries but a large SOC difference, which will slow down the equilibrium speed. It can be seen from this that a single equilibrium variable is not sufficient to characterize the inconsistency between lithium batteries. The present invention selects the dual variables of SOC and voltage as the equilibrium criterion based on the OCV-SOC average value curve, that is, when the SOC is in the two stages of 0-10% and 90%-100%, the voltage is selected as the equilibrium variable, and when it is in the stage of 10%-90%, the SOC is selected as the equilibrium variable.

[0088] The present invention uses the SOC and voltage dual variables of adjacent charges of lithium battery packs as the balance evaluation index, and completes the voltage and SOC balance control of the series lithium battery pack by SOC estimation and voltage correction. The flow chart of the balanced charge and discharge control strategy of the lithium battery pack is as follows: Figure 7 The specific equilibrium control strategy for building dual objective variables includes:

[0089] Step (1), estimate SOC using the unscented Kalman filter (UKF) algorithm: for each lithium battery ( Figure 7 The state of charge (SOC) of the cells (referred to as monomers) is estimated and sorted, and then the lithium batteries with SOC exceeding 90% and below 10% are screened out.

[0090] Step (2): using voltage as the balancing variable, the maximum value method is used to perform voltage correction and balancing on each lithium battery in the lithium battery pack, thereby making the lithium battery voltage within the range of the set voltage threshold β=10mV.

[0091] Step (3): Taking SOC as the balancing target, the average value and difference comparison method are used for balancing, mainly taking the difference of adjacent charges of the lithium battery as the balancing judgment quantity.

[0092] The difference between the SOC values ​​of adjacent lithium batteries is △n, and the threshold value of the difference between adjacent charges is δn=3%. The SOC balancing time between each lithium battery is T Equ for

[0093]

[0094] In the formula, C q is the capacity of lithium battery, I b For balancing current.

[0095] The SOC values ​​of the lithium batteries are arranged in descending order, and △n>δn is denoted as λn (n=1, 2, 3, 4). When the SOC difference between adjacent lithium batteries is within the set threshold range, λn is 0, and outside the threshold range, λn is 1. The changes in the balancing control logic between the lithium batteries are shown in Table 4.

[0096] Table 4 Balance control logic between lithium batteries

[0097]

[0098]

[0099] It can be seen from the data in Table 4 that when λn is 0, there is no need to balance the lithium battery group and the control has no action; when λ1=1, λ2=0, λ3=0, the lithium battery with the largest SOC value in the series transfers energy to the lithium battery with the second largest SOC value; when λ1=1, λ2=0, λ3=1, the lithium battery with the largest SOC value in the series transfers energy to the lithium battery with the smallest SOC value; when λ1=1, λ2=1, λ3=0, the lithium battery with the largest SOC value in the series transfers energy to the lithium battery with the second largest SOC value; when λ1=0, When λ2=1, λ3=0, the lithium battery with the second highest SOC value in the lithium battery will transfer energy to the lithium battery with the third highest SOC value; when λ1=0, λ2=1, λ3=1, the lithium battery with the second highest SOC value in the lithium battery will transfer energy to the lithium battery with the smallest SOC value; when λ1=0, λ2=0, λ3=1, the lithium battery with the third highest SOC value in the lithium battery will transfer energy to the lithium battery with the smallest SOC value; when λ1=1, λ2=1, λ3=1, the lithium battery with the largest SOC value in the lithium battery will transfer energy to the lithium battery with the smallest SOC value. According to the relationship between △n and δn, it is determined which state combination λn is in, and the control system makes corresponding balancing actions according to different states, repeatedly switching between different state combinations until the values ​​of λn are all 0, that is, the states of λn are all 0, indicating that the SOC value balancing between the lithium batteries has been completed.

[0100] In Example 1, a simulation of the balanced control scheme for four lithium batteries connected in series was performed in Matlab / Simulink. Figure 8a -b) shows the constructed lithium battery pack balanced charge and discharge control system, which mainly includes a lithium battery model, a balanced topology circuit, a control system and a display system. Figure 9a -d is Figure 8a The "Control System" module in Figure 9 (including Figure 9a -d) shows the logic of the control system for judging the SOC size of the four groups of lithium battery packs during the charging and discharging process. According to the SOC value obtained by sorting, the Goto block in the loop receives the signal L11~L14 and sends a signal to the lithium battery MOS tube Q11~Q14, and the first switch of the main loop maximum value is turned on ( Figure 9a ), in the same circuit, the Goto block receives signals L111~L114 and sends signals to the lithium battery MOS tubes Q111~Q114, and the second switch of the main circuit maximum value is turned on ( Figure 9b ), the Goto block in the loop receives the signal L210~L213 and sends a signal to the lithium battery MOS tube Q210~Q213, and the first switch of the minimum value of the secondary loop is turned on ( Fig.9c), the Goto block in the loop receives the signal L2101~L2124 and sends a signal to the lithium battery MOS tube Q2101~Q2124, and the second switch of the secondary loop minimum is turned on ( Figure 9d ), and then control the circuit to balance the charge and discharge process. The display system in the model is used to display the SOC and voltage and other parameter changes between the four lithium batteries in the series lithium battery pack.

[0101] The simulation results show that, on the one hand, when the lithium battery pack is charged, the SOC threshold is within 3% compared with the case without equalization. The SOC range of the lithium battery is reduced from 41.5% to 2.1%, a relative reduction of 94.9%, and the voltage range is controlled within 10mV, a relative reduction of 96.7%. Fig.10 The SOC and voltage changes during the charging simulation are shown in Figure 1. (a) is without balancing, and (b) is the balancing of the present invention. On the other hand, during discharge, the SOC range of the lithium battery charge is reduced by 95% and the voltage range is reduced by 91.6% compared to the control without balancing. Fig.11 The SOC and voltage changes of each lithium battery during the discharge process are shown in (a) without balancing and (b) with balancing according to the present invention.

[0102] Table 5 Comparison data of lithium battery pack charging simulation

[0103]

[0104] Table 6 Discharge simulation data

[0105]

[0106] In order to further verify the efficiency of the proposed balancing scheme, the active balancing scheme of the lithium battery pack based on the dual variables of SOC and voltage difference of the present invention is compared with the traditional balancing scheme of inductive energy unidirectional transfer. The charge change curve under static balancing simulation is shown in FIG. Fig.12As shown, (a) is the traditional balancing method, and (b) is the balancing method proposed by the present invention. The scheme proposed by the present invention completes the balancing in 316s. Compared with the traditional balancing method of inductive energy unidirectional transfer, after adopting the balancing method proposed by the present invention, the time for the balancing of No. 1 and No. 3 lithium batteries is reduced from 254s to 169s, and the balancing speed is increased by 33%. The time for the balancing of No. 2 and No. 4 lithium batteries is reduced from 285s to 178s, and the balancing speed is increased by 37.5%. The time for all four lithium batteries to be balanced is reduced from 647s to 316s, and the balancing speed is increased by 51%. The simulation data compared with the traditional method is shown in Table 7. After the traditional topological balancing, the SOC extreme difference value of the charge of each lithium battery is reduced from the initial 33.1% to 12.1%. When the static balancing method is consistent, the SOC extreme difference value is reduced from the initial 33.1% to 1.3%. Both topological balancing reduce the inconsistency of lithium batteries to a certain extent, and under the topological balancing of the present invention, the SOC extreme difference value reaches the preset balancing condition of less than 3%. It can be seen from this that the inconsistency of each lithium battery can be improved to a certain extent in both methods, and the balancing solution proposed in the present invention can efficiently complete the balancing and improve the energy utilization rate.

[0107] Table 7 Data of static equilibrium time of 316s

[0108]

[0109]

[0110] Example 2, based on Example 1, the balancing control strategy and the balancing topology structure are integrated into the lithium battery pack balancing control test platform for charging and discharging and static balancing tests. The hardware construction of the balancing control system includes a balancing module, a collection module, a driving charging module, and a load module. The system hardware block diagram is shown in FIG. Fig.13 As shown, the topological structure of the balancing board corresponds to the balancing circuit composed of the inductor L, various MOS tubes and the freewheeling diode in FIG8 . The lithium battery 1 to the lithium battery 4 in the series-connected lithium battery pack correspond to the four lithium batteries in FIG8 . The charge SOC, voltage and current collection in the lithium battery information collection board correspond to the SOC, Voltage and Current modules in FIG8 . The main control board corresponds to the balancing control logic in FIG9 . The active balancing of the system is completed by issuing a balancing instruction to the balancing board through the balancing control strategy.

[0111] In Example 2, the lithium battery pack balancing control process needs to detect the charge and discharge status of each lithium battery. Fig.14 (a) and 14(b) are the lithium battery balanced charge and discharge system test platforms. Fig.14(b) shows an active balancing control system for a lithium battery pack consisting of a balancing circuit board and a lithium battery pack module. In the figure, 1 is a driving power module, which is used to charge each lithium battery in the lithium battery pack; 2 is an electronic load module, and the DC electronic load model N3305A0 performs constant current discharge operation on the lithium battery, and the input current range is 0A~60A, the voltage range is 0V~150V, and the measurement accuracy is 0.02%; 3 is an active balancing circuit board, which is used to balance the charge and discharge control of lithium batteries in an unbalanced state; 4 is a lithium battery pack module, which is composed of four 18650 type lithium iron phosphate batteries with a rated voltage of 3.7V and a lithium battery capacity of 3000mAh connected in series, as shown in FIG. Fig.14 (b) shows the lithium battery pack active balancing control system composed of 3 and 4; 5 is a paperless recorder for storing and processing the collected data; 6 is an oscilloscope, and the HIOKIPW3390 high-precision power analyzer is used to measure the changes in parameters such as voltage and current of each lithium battery during operation, with a reading error of ±0.04%; 7 is the host computer control panel, and the electronic computer monitoring platform obtains information such as the charging and discharging current, charge state, and lithium battery voltage of the lithium battery pack through serial communication. In order to reduce the loss of the switch, the switch signal frequency is set to a small 20kHz, the charging and discharging current is kept within the 3C range, and the test temperature is set to 20℃.

[0112] The test results show that, on the one hand, during the charging and balancing stage, an active balancing test is performed on a lithium battery pack composed of four lithium batteries connected in series to obtain the operating status of each lithium battery under the premise of balancing. The balancing evaluation index is the extreme value of the voltage of each lithium battery after charging is completed. The test steps of the lithium battery balancing test platform include:

[0113] Step (11), according to the characteristic parameters of the lithium battery, in the test, the lithium battery pack is discharged at a constant current of 3C charge and discharge rate to the cut-off voltage of each lithium battery, the capacity and voltage value of each lithium battery are recorded, and then the battery is left to stand for T1 minutes (T1=20 in this embodiment);

[0114] Step (12), continue to charge the lithium battery group with 3C current in the order of constant current and constant voltage, record the voltage change and capacity of each lithium battery, and record the voltage comparison change of the four lithium batteries before and after active equalization in Table 8;

[0115] Step (13), then continue to discharge at a constant current of 3C to the cut-off voltage, and let stand for T2 minutes (T2=20 in this embodiment);

[0116] Step (14), after the discharge is completed, the lithium battery pack is continuously charged with a 3C current. At the same time, at the end of the balanced charging period, that is, when the charge SOC is greater than 0.9, the charging process is started according to the dual-variable balanced control strategy. After the control of this stage is completed, the lithium battery pack is charged to the rated voltage at a constant voltage and then left to stand for T3 minutes (T3=20 in this embodiment);

[0117] Step (15), repeat step (13) and step (14) four times, and record the status of each lithium battery.

[0118] Table 8 Lithium battery pack data before and after charge equalization

[0119]

[0120] From the data in Table 8, it can be seen that the voltage range of the lithium battery pack before charging equalization reached 235mV, and the inconsistency between the lithium batteries was high. The set equalization voltage threshold was 10mV. After active equalization, the maximum voltage difference between the lithium batteries in the lithium battery pack was reduced to 9mV, a relative decrease of 96.2%, which met the requirements of the equalization study. The entire equalization process time was relatively long compared to the simulation time. The equalization time was affected by the voltage difference and internal resistance. The smaller the voltage difference, the smaller the current and the longer the time. On the contrary, the smaller the resistance, the shorter the time. The equalization time was within an acceptable range.

[0121] The voltage change curve of each lithium battery during the active balancing process is as follows Fig.15 As shown, after charging begins, the maximum initial voltage range difference of each lithium battery in the lithium battery pack is 0.235V, which is greater than the voltage equalization start threshold of 0.01V. According to the equalization control scheme of the present invention, the equalization system starts to work and equalizes the series-connected lithium battery pack. Bat4 has the smallest initial voltage, and its voltage rise rate is relatively fast during the charging process. The charging speed of the remaining lithium batteries is relatively slow. As the voltage difference gradually decreases, the charging speed of each lithium battery decreases. When the time reaches 1120s, the equalization process is completed, the voltage range difference between the lithium batteries is less than 10mV, and the inconsistency of the lithium battery pack is reduced, which is consistent with the simulation results. Fig.10 (b) The voltage change trends of each lithium battery show good consistency. However, while maintaining relative synchronization, since the actual balancing process is affected by temperature and internal resistance, there may be a certain degree of error between the test results and the simulation results.

[0122] Furthermore, in order to verify the effectiveness of the balancing control strategy proposed in the present invention in the discharge stage, a series-connected lithium battery pack consisting of 4 lithium batteries was selected as the research object, and a discharge static test was carried out. After each lithium battery was charged and static, it was discharged at a constant current of 3C. When the system detected that the voltage was outside the threshold range, the balancing between the lithium batteries was turned on. The voltage comparison values ​​of each lithium battery in the lithium battery pack before and after discharge balancing are shown in Table 9.

[0123] Table 9 Data of lithium battery pack before and after discharge equalization

[0124]

[0125] The voltage change curve of each lithium battery is as follows: Fig.16 As shown in Table 5-3, when the lithium battery pack is in a discharge static state, the change trend is consistent with the simulation results. When the balanced charge and discharge control system detects lithium batteries with inconsistent initial energy, the voltage extreme difference is greater than the balanced voltage threshold. According to the active balanced control strategy, the lithium battery with higher energy stores the excess energy in the inductor, and transfers it to the lithium battery with lower voltage through the inductor until the voltage extreme difference in the lithium battery pack is less than 0.01V, and the balance is turned off. From the data in Table 5-3, it can be seen that the initial voltage extreme difference of each lithium battery in the lithium battery pack is 0.226V, and the inconsistency before discharge static is relatively high. After the active balance is completed, the voltage extreme difference is 0.006V, which is relatively reduced by 97.3%, meeting the voltage difference set by the balance requirement of less than 10mV.

[0126] The results of Example 2 show that the changing trends of the test results and the simulation results show good consistency. Therefore, the dual-variable active balancing control method proposed in the present invention can greatly improve the inconsistency between the lithium batteries. The balancing efficiency of the lithium battery pack during charging, discharging and static states is high, and the service life is effectively extended.

[0127] The embodiments described above are merely descriptions of preferred implementations of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A lithium battery pack active balancing control method based on dual variables of SOC and voltage difference, characterized in that: The following steps are involved: Step 1, establish an equivalent circuit model and parameter identification, simulate the dynamic characteristics of lithium batteries, use the first-order RC equivalent circuit model of a single lithium battery as the basic model of the lithium battery to analyze the characteristic parameters of the lithium battery, and use the nonlinear least squares fitting algorithm combined with the pulse discharge voltage change curve of the lithium battery to identify the parameters corresponding to each lithium battery model; Step 2: Build an active balancing topology structure and design an inductor-based bidirectional active balancing circuit based on the Buck-Boost balancing circuit; Step 3: Construct a dual-objective variable balancing control strategy. According to the OCV-SOC relationship curve of the lithium battery, when the SOC is in the 0-10% and 90-100% stages, select the voltage as the balancing variable. When the SOC is in the 10%-90% stage, select the SOC as the balancing variable. The voltage and SOC balancing control of the series lithium battery pack is completed by SOC estimation and voltage correction. In step 3, in the stage of constructing the balancing control strategy of the dual objective variables, the difference between the SOC values ​​of adjacent lithium batteries is Δn, the threshold value of the difference between adjacent charges is δn, the SOC values ​​of each lithium battery are arranged in descending order, Δn>δn is λn, n=1,2,3,4, when the adjacent SOC difference between each lithium battery is within the set threshold range, λn is 0, and outside the threshold range, λn is 1. When λn is 0, there is no need to balance the lithium battery group, and the control has no action; when λ1=1, λ2=0, λ3=0, the lithium batteries in series The lithium battery with the largest SOC value transfers energy to the lithium battery with the second largest SOC value, and controls the direction and size of the balancing current between the lithium batteries by controlling the on and off of the MOS tube and the time, until the difference in charge between adjacent lithium batteries is ≤3%. Similarly, the state combination of λn is determined based on the size relationship between △n and δn. The control system makes corresponding balancing actions based on different states, repeatedly switching between different state combinations until the values ​​of λn are all 0, that is, the states of λn are all 0, indicating that the SOC value balancing between the lithium batteries is completed.

2. The active equalization control method for a lithium battery pack based on dual variables of SOC and voltage difference according to claim 1, characterized in that: In step 1, the first-order RC equivalent circuit model of the single lithium battery includes a voltage E m , resistor R0, resistor R1 and capacitor C1; the lithium battery alternates between charge and discharge cycles, and the state of charge of the lithium battery model is estimated by measuring the current, voltage and temperature of the lithium battery; Discrete state space equation of single lithium battery: In the formula, C1(SOC t ,T) t is the polarization capacitor of the lithium battery, U1 t Represents the polarization voltage of the lithium battery at time t, C q is the rated capacity of the lithium battery, T s is the sampling period, W t is the process noise at time t, then: U t =E m (SOC t ,T) t -U1 t -IR0(SOC t ,T) t +V t (2) Where T represents the internal temperature of the lithium battery, U t and E m (SOC t ,T) t Respectively represent the terminal voltage of the lithium battery at time t and the estimated open circuit voltage OCV, R0 (SOC t ,T) t and R1(SOC t ,T) t Respectively represent the internal resistance and polarization internal resistance of the lithium battery at time t, V t represents the measurement noise of the lithium battery at time t; According to the identified parameters of the first-order RC equivalent circuit model of the single lithium battery, a first-order RC equivalent circuit Simulink model of the lithium battery is constructed; at the TE1 temperature, the parameters of the first-order RC equivalent circuit Simulink model of the lithium battery are identified; in Matlab / Simulink, the nonlinear least squares fitting algorithm is combined with the pulse discharge voltage change curve of the lithium battery to calculate the parameter value voltage E corresponding to each lithium battery model. m , resistor R0, and resistor R1 for identification.

3. The active equalization control method for a lithium battery pack based on dual variables of SOC and voltage difference according to claim 2, characterized in that: In step 2, the bidirectional active balancing circuit includes a lithium battery pack and a balancing primary and secondary circuit. Each lithium battery in the lithium battery pack is connected to an inductor L and four MOS tubes to form a bidirectional energy transfer circuit. Each MOS tube is connected in parallel with a freewheeling diode. After the lithium battery is over-discharged, the protection function will be turned on and the MOS tube will be turned off. When in the charging state, the body diode of the MOS tube is used to make the circuit conductive and the system works. The main circuit MOS tube Q1m, m=1,2,...,14, controls the charging circuit of the lithium battery, and the secondary circuit MOS tube Q2n, n=10,11,...,124, controls the discharge circuit of the lithium battery. After a certain period of duty cycle, the MOS tube can quickly switch between the two states of conduction and shutdown. The energy transfer between each lithium battery is completed by controlling the on and off of the corresponding MOS tube. During the charging and discharging process, the energy is transferred from the lithium battery to the energy storage inductor through the main circuit mode, and the energy is transferred from the inductor to the low-energy lithium battery through the main circuit and the secondary circuit combination mode.

4. The active equalization control method for a lithium battery pack based on dual variables of SOC and voltage difference according to claim 3, characterized in that: The equilibrium control strategy for constructing dual objective variables includes: Step (1), estimating SOC using an unscented Kalman filter (UKF) algorithm and establishing a Simulink model for the estimation; sorting the SOC estimates of the lithium batteries, and then screening out lithium batteries with SOCs exceeding 90% and below 10%; Step (2), using voltage as the balancing variable, using the maximum value method to perform voltage correction balancing on each lithium battery in the lithium battery pack, so that the lithium battery voltage is within the range of the set voltage threshold β=10mV; Step (3): Taking SOC as the balancing target, the average value and difference comparison method are used for balancing, and the difference between adjacent charges of the lithium battery is used as the balancing criterion.

5. The active equalization control method for a lithium battery pack based on dual variables of SOC and voltage difference according to claim 4, characterized in that: In step (3), the difference between the SOC values ​​of adjacent lithium batteries is recorded as △n, and the threshold value of the difference between adjacent charges is δn=3%; the SOC balancing time between each lithium battery is T Equ for In the formula, C q is the capacity of lithium battery, I b is the balancing current; The SOC values ​​of the lithium batteries are arranged in descending order, and △n>δn is denoted as λn, where n=1, 2, 3, 4. The state combination of λn is determined based on the size relationship between △n and δn. The control system makes corresponding balancing actions based on different states, repeatedly switching between different state combinations until the values ​​of λn are all 0, that is, the states of λn are all 0, indicating that the SOC value balancing between the lithium batteries is completed.

6. The active equalization control method for a lithium battery pack based on dual variables of SOC and voltage difference according to claim 5, characterized in that: It also includes a lithium battery balancing test platform, which includes a driving power module, an electronic load module, an active balancing circuit board, a lithium battery pack module, a paperless recorder, an oscilloscope and a host computer control panel; the driving power module is used to charge each lithium battery in the lithium battery pack; the electronic load module is used to perform a constant current discharge operation on the lithium battery; the active balancing circuit board is used to perform balanced charging and discharging control on the lithium battery in an unbalanced state; the lithium battery pack module is composed of a number of lithium batteries connected in series; the paperless recorder is used to store and process the collected data; the oscilloscope is used to measure the parameter changes of each lithium battery when it is working; The host computer control panel obtains the charging and discharging current, state of charge and lithium battery voltage information of the lithium battery pack through serial communication; the test steps of the lithium battery balancing test platform include: Step (11), according to the characteristic parameters of the lithium battery, the lithium battery pack is discharged at a constant current of 3C charge and discharge rate to the cut-off voltage of each lithium battery, the capacity and voltage value of each lithium battery are recorded, and then the battery is left to stand for T1 minute; Step (12), continue to charge the lithium battery group with 3C current in the order of constant current and constant voltage, record the voltage change and capacity of each lithium battery, and record the voltage comparison change before and after active equalization of the lithium battery in a table; Step (13), then continue to discharge at a constant current of 3C to the cut-off voltage, and stand for T2 minutes; Step (14), after the discharge is completed, the lithium battery pack is continuously charged with a 3C current. At the same time, at the end of the balanced charging period, that is, when the charge SOC is greater than 0.9, the charging process is started according to the dual-variable balanced control strategy. After the control of this stage is completed, the lithium battery pack is charged to the rated voltage at a constant voltage and then left to stand for T3 minutes; Step (15), repeat step (13) and step (14) for N times, and record the status of each lithium battery.

Images