Battery pack SOC cooperative control method and system based on multi-channel parallel equalization and rolling optimization

Abstract

The invention relates to the technical field of batteries, and provides a battery pack SOC cooperative control method and system based on multi-channel parallel equalization and rolling optimization, and the method comprises the steps: collecting the SOC of each single body in each control period, calculating an average value, screening out high and low SOC candidate single bodies, carrying out the discharge-charge pairing according to the number of channels, and carrying out the control of the SOC of each single body. Establishing a mapping relation between the monomer and the bidirectional DC-DC channel; an SOC state equation in a future prediction time domain is constructed on the basis, and an objective function is constructed by taking the equalization current of each channel as a decision variable: a deviation quadratic sum cumulant of each single SOC relative to the average SOC in the prediction time domain and a weighted sum of equalization energy consumption of each channel are minimized, and a current amplitude constraint is applied at the same time; and solving the optimization problem to obtain the optimal equalizing current, driving the corresponding monomer to execute charging and discharging operation, and then rolling to enter the next period. According to the method, overbalance and oscillation are effectively suppressed, high-precision, low-loss and self-adaptive global collaborative equalization is realized, the SOC dispersion is remarkably reduced, and the service life of the battery is prolonged.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery, in particular to a battery pack SOC cooperative control method and system based on multi-channel parallel equalization and rolling optimization. BACKGROUND

[0002] In the application of lithium ion battery, due to the factors such as manufacturing process deviation, use environment difference and inconsistent aging, the state of charge SOC of each single battery will gradually produce dispersion, which will lead to overcharge or overdischarge of some single batteries, seriously threatening the system safety and shortening the overall life. Therefore, the battery management system BMS makes the SOC of each single battery consistent through equalization control. In recent years, in order to improve the equalization speed, multi-channel parallel equalization architecture has been widely studied and applied. However, the existing parallel equalization scheme has the following problems:

[0003] (1) Most systems only set equalization action according to the current SOC value, which cannot predict future working condition changes, and is easy to cause over equalization or repeated oscillation, reducing system efficiency and life.

[0004] (2) Equalization resources are not allocated to the single battery that needs equalization most.

[0005] (3) A unified optimization framework is not constructed to minimize the SOC dispersion of the whole battery, and the equalization current energy consumption is not considered, so it is impossible to realize the balance between high-precision equalization and low energy loss.

[0006] These problems need to be solved urgently. SUMMARY

[0007] In view of the above technical problems, the purpose of the present application is to provide a battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization, which aims to solve the above technical problems.

[0008] In order to solve the above technical problems, in the first aspect, the present application provides a battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization, which comprises:

[0009] Collecting the SOC values of each single battery in the target battery pack at the current time, and calculating the average SOC value at the current time;

[0010] Selecting m single batteries with SOC values higher than the sum of the average SOC value and the first threshold value and m single batteries with SOC values lower than the difference between the average SOC value and the second threshold value from the target battery pack as candidate single batteries; wherein m is less than or equal to the number of channels of DC-DC converter;

[0011] According to the number of channels of the bidirectional DC-DC converter, the candidate monomers are paired according to a one-to-one pairing principle to form a plurality of discharge-charge monomer pairs, and a mapping relationship between the monomers and the channels of the DC-DC converter is established;

[0012] Based on the mapping relationship, a state prediction equation of each monomer SOC in a future prediction time domain is constructed;

[0013] Taking the equalization current of each DC-DC converter channel as a decision variable, based on the state prediction equation of each monomer SOC in the future prediction time domain, a time domain optimization problem is constructed, and the objective function is to minimize the weighted sum of the cumulative amount of the square sum of the deviation of each monomer SOC from the average SOC in the future prediction time domain and the equalization energy consumption of each channel, and the current constraint is applied;

[0014] Solving the time domain optimization problem, the optimal equalization current of the next control period is obtained;

[0015] Controlling each DC-DC converter channel to perform charge-discharge operation on its paired monomers according to the sign and amplitude of its optimal equalization current;

[0016] In the next control period, step S1 is returned, and the rolling execution is performed.

[0017] Further, the state prediction equation of each monomer SOC in the future prediction time domain is:

[0018] ;

[0019] Wherein, represents the soc prediction value of monomer i at the future t step in the current control period k, represents the soc prediction value of monomer i at the future t-1 step in the current control period k, represents the equalization current at the future t-1 step in the current control period k, is the DC-DC conversion efficiency, is a state flag, represents that monomer i is in a charging state in channel c, represents a discharge state, k represents the index of the current control period, is the rated capacity of the monomer, is the control period, i is the monomer number, and t is the prediction step index.

[0020] Further, the objective function is:

[0021] ;

[0022] Wherein, J is the weighted sum of the square sum of the deviation of each single SOC from the average SOC in the future prediction time domain and the equalization energy consumption of each channel, c is the DC-DC converter channel index, C is the number of DC-DC converter channels, N is the prediction time domain length, and λ is the weight coefficient, whose unit is , is the equivalent resistance of channel c.

[0023] Further, the current constraint includes: the equalization current amplitude of each channel does not exceed the preset maximum equalization current , and satisfies .

[0024] Further, the first threshold value is less than or equal to the second threshold value.

[0025] Further, the single body with a SOC value higher than the sum of the average SOC value and the first threshold value is regarded as a high SOC single body, and the single body with a SOC value lower than the difference between the average SOC value and the second threshold value is regarded as a low SOC single body, wherein, when screening the high SOC single body and the low SOC single body, if the number of high SOC single bodies and the number of low SOC single bodies are not equal, the smaller one of the two is taken as the pairing number m, and the single body with a larger SOC deviation from the average value is preferentially selected as the candidate single body.

[0026] In a second aspect, the embodiments of the present application provide a battery pack SOC cooperative control system based on multi-channel parallel equalization and rolling optimization, which comprises:

[0027] A collection module is configured to collect the SOC values of each single body in the target battery pack at the current time, and calculate the average SOC value at the current time;

[0028] A screening module is configured to screen m single bodies with SOC values higher than the sum of the average SOC value and the first threshold value, and m single bodies with SOC values lower than the difference between the average SOC value and the second threshold value from the target battery pack, as candidate single bodies; wherein, m is less than or equal to the number of DC-DC converter channels;

[0029] A mapping relationship establishing module is configured to establish a mapping relationship between the candidate single bodies and the DC-DC converter channels according to the number of DC-DC converter channels, and form a plurality of discharge-charge single body pairs according to the one-to-one pairing principle.

[0030] A state prediction equation constructing module is configured to construct a state prediction equation of each single body SOC in a future prediction time domain based on the mapping relationship.

[0031] a time-domain optimization problem construction module, configured to construct a time-domain optimization problem with balanced currents of each DC-DC converter channel as decision variables, based on state prediction equations of each single SOC in a future prediction time domain, and with an objective function of minimizing a weighted sum of a cumulative amount of squared deviations of each single SOC from an average SOC and balanced energy consumption of each channel in the future prediction time domain, and imposing current constraints;

[0032] a solution module, configured to solve the time-domain optimization problem to obtain optimal balanced currents of a next control period;

[0033] a control module, configured to control each DC-DC converter channel to perform charging and discharging operations on its paired single according to a sign and a magnitude of the optimal balanced current of the channel;

[0034] a rolling execution module, configured to return to step S1 in a next control period for rolling execution.

[0035] Further, the state prediction equation of each single SOC in the future prediction time domain is:

[0036]

[0037] wherein, socpred k,t represents a soc prediction value of the single i at a future tth step in the current control period k, socpred k,t-1 represents a soc prediction value of the single i at a future (t-1)th step in the current control period k, ik,t-1 represents a balanced current at a future (t-1)th step in the current control period k, is a DC-DC conversion efficiency, is a state flag, represents that the single i is in a charging state in the channel c, represents a discharging state, k represents an index of the current control period, is a single rated capacity, is a control period, i is a single number, and t is a prediction step index.

[0038] Further, the objective function is:

[0039]

[0040] wherein, J is a weighted sum of a cumulative amount of squared deviations of each single SOC from an average SOC and balanced energy consumption of each channel in the future prediction time domain, c is a DC-DC converter channel index, C is a DC-DC converter channel number, N is a prediction time domain length, and λ is a weight coefficient with a unit of , is an equivalent resistance of the channel c.

[0041] ​​Further, the current constraint comprises: equalizing the current amplitude of each channel not exceeding a preset maximum equalizing current , and satisfying .

[0042] The embodiment of the application builds a state prediction equation of each single SOC in a future prediction time domain, takes the equalizing current of each DC-DC converter channel as a decision variable, builds a target function of minimizing the weighted sum of the cumulative amount of the square sum of the deviation of each single SOC from the average SOC in the future prediction time domain and the equalizing energy consumption of each channel based on the state prediction equation of each single SOC in the future prediction time domain, comprehensively considers the evolution trend of the SOC in the future multiple steps, effectively suppresses the over-equalization, repeated oscillation or insufficient equalization phenomenon caused by lag response, and improves the stability and safety of system operation. The optimization problem of taking the target function of "minimizing the weighted sum of the cumulative amount of the square sum of the deviation of each single SOC from the average SOC in the future prediction time domain and the equalizing energy consumption of each channel" is built, the current constraint is applied, the battery pack is considered as a whole for collaborative control, and the balance of high-precision equalization and low-energy loss is realized. By dynamically screening high / low SOC single bodies and real-time pairing according to the number of channels every cycle, it is ensured that the equalization task is only assigned to the single bodies that really need it; and by the optimization problem, it is automatically determined how much current each channel uses, and the single body with greater deviation obtains higher equalization power, so that the limited equalization hardware capacity can play the maximum benefit. By rolling the above process every control cycle, i.e. "perception → screening → pairing → prediction → optimization → execution → re-perception", the application realizes the closed-loop, adaptive and global collaborative active equalization control, which not only significantly reduces the SOC dispersion of the battery pack and prolongs the service life of the battery, but also effectively reduces unnecessary energy conversion loss and improves the overall energy efficiency of the system. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creating any creative labor.

[0044] Figure 1 is a flowchart of the battery pack SOC collaborative control method based on multi-channel parallel equalization and rolling optimization provided by the embodiment of the application;

[0045] Figure 2 is a structural schematic diagram of the battery pack SOC collaborative control system based on multi-channel parallel equalization and rolling optimization provided by the embodiment of the application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0047] Those skilled in the art can understand that the singular forms "a", "an" and "the" used herein include plural forms unless specifically stated otherwise. It should be further understood that the use of the term "include" in the specification of the present application means that a feature, integer, step, operation, element, module and / or assembly exists, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, assemblies and / or their combinations. It should be understood that when we say that an element is "connected" or "coupled" to another element, it can be directly connected or coupled to the other element, or there can be intermediate elements. In addition, "connected" or "coupled" used herein can include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any module and all combinations of the associated listed items.

[0048] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as that generally understood by those skilled in the art to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have meanings consistent with those in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as such.

[0049] Please refer to Figure 1 The embodiment of the present application provides a battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization, which comprises the following steps:

[0050] S1, collecting the SOC values of each single battery in the target battery pack at the current time, and calculating the average SOC value at the current time;

[0051] S2, selecting m single batteries with SOC values higher than the sum of the average SOC value and a first threshold value, and m single batteries with SOC values lower than the difference between the average SOC value and a second threshold value from the target battery pack as candidate single batteries; wherein m is less than or equal to the number of channels of the DC-DC converter;

[0052] S3, according to the number of channels of the bidirectional DC-DC converter, the candidate single batteries are paired according to the one-to-one pairing principle to form a plurality of discharge-charge single battery pairs, and a mapping relationship between the single batteries and the channels of the DC-DC converter is established;

[0053] S4, constructing a state prediction equation of each monomer SOC in a future prediction time domain based on the mapping relationship;

[0054] S5, taking the equalization current of each DC-DC converter channel as a decision variable, constructing a time domain optimization problem based on the state prediction equation of each monomer SOC in the future prediction time domain, and taking the objective function as: minimizing the weighted sum of the cumulative amount of the square sum of the deviation of each monomer SOC from the average SOC and the equalization energy consumption of each channel in the future prediction time domain, and imposing a current constraint;

[0055] S6, solving the time domain optimization problem to obtain the optimal equalization current of the next control period;

[0056] S7, controlling each DC-DC converter channel to perform charging and discharging operations on its paired monomer according to the sign and amplitude of its optimal equalization current;

[0057] S8, returning to step S1 in the next control period and rolling execution.

[0058] The embodiment of the application adopts an existing battery equalization parallel architecture to support dynamic energy scheduling between any high / low SOC monomers. For step S3, according to the number of channels of the bidirectional DC-DC converter, the candidate monomers are paired according to the one-to-one pairing principle to form a plurality of discharge-charge monomer pairs, and a mapping relationship between the monomers and the DC-DC converter channels is established. Specifically, monomers with a SOC higher than the sum of the average SOC value and the first threshold value are taken as a discharge candidate set and arranged in descending order of SOC; monomers with a SOC lower than the difference between the average SOC value and the second threshold value are taken as a charging candidate set and arranged in ascending order of SOC; the first m pairs of monomers are selected from the two candidate sets in turn to form discharge-charge monomer pairs, wherein the cth channel is connected to the cth high SOC monomer and the cth low SOC monomer, thereby establishing a one-to-one mapping relationship between the monomers and the DC-DC converter channels; wherein m is less than or equal to C, C is the number of channels, and c is the channel number. In step S6, the time domain optimization problem is solved, specifically, a quadratic programming (QP) method is used for solving. Specifically, an open solver such as QSQP can be used for solving. OSQP (Operator Splitting Quadratic Program) is a high-performance quadratic convex optimization solver with high efficiency, robustness and open source characteristics, and is widely used in embedded systems and control optimization fields.

[0059] The embodiment of the application constructs a state prediction equation of each single SOC in a future prediction time domain, takes the balanced current of each DC-DC converter channel as a decision variable, constructs a target function of minimizing the weighted sum of the cumulative amount of the deviation square sum of each single SOC relative to the average SOC in the future prediction time domain and the balanced energy consumption of each channel based on the state prediction equation of each single SOC in the future prediction time domain, comprehensively considers the evolution trend of the SOC in the future multiple steps, effectively suppresses the over-balancing, repeated oscillation or insufficient balancing phenomenon caused by lag response, and improves the stability and safety of system operation. The optimization problem of taking the target function of "minimizing the weighted sum of the cumulative amount of the deviation square sum of each single SOC relative to the average SOC in the future prediction time and the balanced energy consumption of each channel" is constructed, and the current constraint is applied, the battery pack is regarded as a whole for collaborative control, and the balance of high-precision balancing and low-energy loss is realized. By dynamically screening high / low SOC monomers and real-time pairing according to the number of channels every cycle, it is ensured that the balancing task is only assigned to the monomers that really need it; and by the optimization problem, it is automatically determined how much current each channel uses, and the monomers with greater deviation obtain higher balancing power, so that the limited balancing hardware capacity can play the maximum benefit. By rolling the above process in each control cycle, that is, "perception -> screening -> pairing -> prediction -> optimization -> execution -> re-perception", the application realizes the closed-loop, adaptive and global collaborative active balancing control, which not only significantly reduces the SOC dispersion of the battery pack and prolongs the service life of the battery, but also effectively reduces unnecessary energy conversion loss and improves the overall energy efficiency of the system.

[0060] In an embodiment, the state prediction equation of each single SOC in the future prediction time domain is:

[0061] ;

[0062] Wherein, represents the soc prediction value of the single i in the future t step at the current control cycle k, represents the soc prediction value of the single i in the future t-1 step at the current control cycle k, represents the balanced current of the future t-1 step at the current control cycle k, is the DC-DC conversion efficiency, is a state flag, represents that the single i is in the charging state in the channel c, represents the discharging state, k represents the index of the current control cycle, is the single rated capacity, is the control cycle, i is the single number, and t is the prediction step index.

[0063] In the embodiments of the present application, it should be understood that the SOC (state of charge) of the battery is essentially "current remaining capacity ÷ rated capacity", so the change in SOC is equal to "the capacity of charging / discharging in the balancing process" divided by "the rated capacity". is the SOC prediction value of the previous step, is the change in SOC (increase / decrease in SOC caused by the balancing current). The core logic of the state prediction equation of each single SOC in the future prediction time domain is to gradually calculate the prediction value of each step in the future from the currently known SOC value. When t = 1, represents the currently known SOC value.

[0064] In an embodiment, the target function is:

[0065] ;

[0066] where J is the weighted sum of the square sum of the deviation of each single SOC in the future prediction time domain from the average SOC and the energy consumption of each channel, c is the index of the DC-DC converter channel, C is the number of DC-DC converter channels, N is the length of the prediction time domain, and λ is the weight coefficient, whose unit is , is the equivalent resistance of channel c.

[0067] It should be noted that the target function integrates the SOC deviation of each step in the future and the energy consumption of each step in the future into a "total cost", and the amplitude of the balancing current of each channel is the decision variable of this "total cost"; by minimizing this "total cost", the balancing current that has good balancing effect and low energy consumption in the future time domain can be obtained, which is the optimal balancing current of the next control cycle.

[0068] In the embodiments of the present application, it should be understood that the energy consumption weight coefficient λ is designed to have a unit of inverse joule , which is used to convert the energy loss (unit: J) in the balancing process into a cost quantity of the same dimension as the SOC deviation square term (dimensionless), to meet the requirement of dimensional consistency. In addition, the energy consumption weight coefficient is also used to adjust the relative importance of the SOC deviation cost and the energy consumption cost. In an embodiment, can be designed as follows: first, select a typical initial SOC distribution and fix other parameters (e.g. , , ); then try different λ (e.g. ), and observe the balancing time and total energy consumption to select the Pareto optimal λ.

[0069] In one embodiment, the current constraint includes: equalizing the current amplitude of each channel. Not exceeding the preset maximum equalization current And satisfy .

[0070] In this embodiment of the application, the large equalization current The ratings are determined by the power device ratings of the DC-DC converter and the safe current limits of individual battery cells, and are used to prevent hardware overload and system instability.

[0071] In one embodiment, the first threshold is less than or equal to the second threshold.

[0072] In this embodiment, introducing dual thresholds effectively suppresses oscillations and improves system stability. Furthermore, setting the first threshold to be lower than the second threshold indicates a higher risk of SOC (State of Charge), allowing for earlier intervention and balancing safety and stability. The first and second thresholds are determined based on battery safety characteristics.

[0073] In one embodiment, monomers with a SOC value higher than the sum of the average SOC value and a first threshold are designated as high SOC monomers, and monomers with a SOC value lower than the difference between the average SOC value and a second threshold are designated as low SOC monomers. When screening high SOC and low SOC monomers, if the number of high SOC monomers and the number of low SOC monomers are not equal, the smaller of the two is taken as the pairing quantity m, and monomers with a large SOC deviation from the average value are preferentially selected as candidate monomers.

[0074] This application provides a battery pack SOC collaborative control system based on multi-channel parallel equalization and rolling optimization, the system comprising:

[0075] Acquisition module 1 is used to acquire the SOC value of each cell in the target battery pack at the current moment and calculate the average SOC value at the current moment;

[0076] The screening module 2 is used to screen m cells from the target battery pack whose SOC value is higher than the sum of the average SOC value and the first threshold, and m cells whose SOC value is lower than the difference between the average SOC value and the second threshold, as candidate cells; wherein m is less than or equal to the number of channels of the DC-DC converter.

[0077] The mapping relationship establishment module 3 is used to form several discharge-charge unit pairs according to the number of channels of the bidirectional DC-DC converter, and establish the mapping relationship between the unit and the DC-DC converter channel.

[0078] The state prediction equation construction module 4 is used to construct the state prediction equation of each individual SOC in the future prediction time domain based on the mapping relationship.

[0079] The time-domain optimization problem construction module 5 is used to construct a time-domain optimization problem based on the state prediction equation of each individual SOC in the future prediction time domain, with the equalization current of each DC-DC converter channel as the decision variable. Its objective function is to minimize the weighted sum of the sum of squares of the deviation of each individual SOC from the average SOC and the equalization energy consumption of each channel in the future prediction time domain, and to apply current constraints.

[0080] Solver module 6 is used to solve the time-domain optimization problem to obtain the optimal equalization current for the next control cycle;

[0081] Control module 7 is used to control each DC-DC converter channel to perform charging and discharging operations on its paired cells according to the sign and amplitude of its optimal equalization current.

[0082] The rolling execution module 8 is used to return to step S1 in the next control cycle and perform rolling execution.

[0083] In one embodiment, the state prediction equation for each individual SOC in the future prediction time domain is as follows:

[0084] ;

[0085] in, This represents the predicted soc value of monomer i at step t in the future, given the current control period k. This represents the predicted soc value of monomer i in the (t-1)th step of the future, given the current control period k. This represents the equalization current in the current control period k at the (t-1)th future step. For DC-DC conversion efficiency, For status flags, This indicates that cell i is in a charging state in channel c. This indicates the discharge state, and k represents the index of the current control cycle. This refers to the rated capacity of a single unit. To control the cycle, i is the individual unit number, and t is the prediction step size index.

[0086] In one embodiment, the objective function is:

[0087] ;

[0088] Where J is the weighted sum of the squared deviations of each unit's SOC relative to the average SOC within the future prediction time domain, and the equalization energy consumption of each channel; c is the DC-DC converter channel index; C is the number of DC-DC converter channels; N is the prediction time domain length; and λ is the weighting coefficient, with units of 1000 kJ / m². , Let be the equivalent resistance of channel c.

[0089] In one embodiment, the current constraint includes: equalizing the current amplitude of each channel. Not exceeding the preset maximum equalization current And satisfy .

[0090] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0091] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0092] The above description is only a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization, characterized in that, The method comprises: collecting SOC values of each single battery in the target battery pack at the current time, and calculating an average SOC value at the current time; selecting m single batteries with SOC values higher than the sum of the average SOC value and a first threshold value and m single batteries with SOC values lower than the difference between the average SOC value and a second threshold value from the target battery pack as candidate single batteries; wherein m is less than or equal to the number of channels of the DC-DC converter; according to the number of channels of the bidirectional DC-DC converter, pairing the candidate single batteries according to the one-to-one pairing principle to form a plurality of discharge-charge single battery pairs, and establishing a mapping relationship between the single batteries and the channels of the DC-DC converter; based on the mapping relationship, constructing a state prediction equation of each single battery SOC in a future prediction time domain; taking the equalization current of each DC-DC converter channel as a decision variable, based on the state prediction equation of each single battery SOC in the future prediction time domain, constructing a time domain optimization problem, and taking the weighted sum of the cumulative amount of the square sum of the deviation of each single battery SOC from the average SOC in the future prediction time domain and the equalization energy consumption of each channel as the objective function, and imposing a current constraint; solving the time domain optimization problem to obtain the optimal equalization current of the next control period; controlling each DC-DC converter channel to perform charge-discharge operation on the paired single batteries according to the sign and amplitude of the optimal equalization current of the channel; returning to step S1 in the next control period and rolling execution.

2. The battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization according to claim 1, characterized in that, The state prediction equation of each single battery SOC in the future prediction time domain is: ； wherein, denotes the soc prediction value of the monomer i at future step t in the current control period k, denotes the soc prediction value of the monomer i at future step t-1 in the current control period k, denotes the balancing current at future step t-1 in the current control period k, is the DC-DC conversion efficiency, is the state flag, denotes that the monomer i is in the charging state in the channel c, denotes the discharging state, k denotes the index of the current control period, is the monomer rated capacity, is the control period, i is the monomer number, and t is the prediction step index.

3. The battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization according to claim 2, characterized in that, the objective function is: ； Wherein J is the weighted sum of the square deviation of each single SOC from the average SOC in the future prediction time domain and the energy consumption of each channel, c is the DC-DC converter channel index, C is the number of DC-DC converter channels, N is the prediction time domain length, and λ is the weight coefficient, whose unit is , is the equivalent resistance of channel c.

4. The battery pack SOC cooperative control method based on multi-channel parallel equalization and rolling optimization according to claim 3, characterized in that, The current constraint includes: each channel equalizes current amplitude No more than the preset maximum equalization current , and satisfies .

5. The method of claim 1, wherein, the first threshold value is less than or equal to the second threshold value.

6. The method of claim 1, wherein, The single battery with an SOC value higher than the sum of the average SOC value and the first threshold value is taken as a high SOC single battery, and the single battery with an SOC value lower than the difference between the average SOC value and the second threshold value is taken as a low SOC single battery, wherein when selecting the high SOC single battery and the low SOC single battery, if the number of high SOC single batteries and the number of low SOC single batteries are not equal, the smaller number of the two is taken as the pairing number m, and the single battery with a larger SOC deviation from the average value is preferentially selected as the candidate single battery.

7. A battery pack SOC cooperative control system based on multi-channel parallel equalization and rolling optimization, characterized in that, The system comprises: a collection module configured to collect SOC values of each single battery in a target battery pack at the current time, and calculate an average SOC value at the current time; a selection module configured to select m single batteries with SOC values higher than the sum of the average SOC value and a first threshold value and m single batteries with SOC values lower than the difference between the average SOC value and a second threshold value from the target battery pack as candidate single batteries; wherein m is less than or equal to the number of channels of the DC-DC converter; a mapping relationship establishment module configured to pair the candidate single batteries according to the one-to-one pairing principle to form a plurality of discharge-charge single battery pairs according to the number of channels of the bidirectional DC-DC converter, and establish a mapping relationship between the single batteries and the channels of the DC-DC converter; a state prediction equation construction module configured to construct a state prediction equation of each single battery SOC in a future prediction time domain based on the mapping relationship; a time-domain optimization problem construction module, configured to construct a time-domain optimization problem with balanced currents of each DC-DC converter channel as decision variables, based on state prediction equations of each single SOC in a future prediction time domain, and with an objective function of minimizing a weighted sum of a cumulative amount of squared deviations of each single SOC from an average SOC and balanced energy consumption of each channel in the future prediction time domain, and imposing current constraints; a solution module, configured to solve the time-domain optimization problem to obtain optimal balanced currents of a next control period; a control module, configured to control each DC-DC converter channel to perform charging and discharging operations on its paired single according to a sign and a magnitude of the optimal balanced current of the DC-DC converter channel; a rolling execution module, configured to return to step S1 in a next control period for rolling execution.

8. The multi-channel parallel equalization and rolling optimization based battery pack SOC cooperative control system according to claim 7, wherein, the state prediction equations of each single SOC in the future prediction time domain are: ； wherein, denotes the soc prediction value of the monomer i at future step t in the current control period k, denotes the soc prediction value of the monomer i at future step t-1 in the current control period k, denotes the balancing current at future step t-1 in the current control period k, is the DC-DC conversion efficiency, is the state flag, denotes that the monomer i is in the charging state in the channel c, denotes the discharging state, k denotes the index of the current control period, is the monomer rated capacity, is the control period, i is the monomer number, and t is the prediction step index.

9. The multi-channel parallel equalization and rolling optimization based battery pack SOC cooperative control system according to claim 8, wherein, the objective function is: ； Wherein J is the weighted sum of the square deviation of each single SOC from the average SOC in the future prediction time domain and the energy consumption of each channel, c is the DC-DC converter channel index, C is the number of DC-DC converter channels, N is the prediction time domain length, and λ is the weight coefficient, whose unit is , is the equivalent resistance of channel c.

10. The multi-channel parallel equalization and rolling optimization based battery pack SOC cooperative control system according to claim 9, wherein, The current constraint includes: each channel equalizes current amplitude No more than the preset maximum equalization current , and satisfies .

Images