Battery management system construction method for module-battery cell multi-level cooperative control

Abstract

The invention relates to a module-cell multi-level cooperative control battery management system construction method, and belongs to the technical field of electric vehicle battery management, and the method comprises the following steps: S1, designing a variable topological structure of a series-parallel circuit in a battery module, the method specifically comprises the following steps: S2, designing a power electronic structure for hierarchical self-adaptive control of a single battery cell in a battery module; s3, based on the multi-level battery management system, establishing an online parameter identification method of a battery cell level, and meanwhile, realizing a dynamic equalization process of a module level under any environmental condition; and S4, constructing a multi-level battery management system module to obtain a complete module-cell multi-level cooperative control battery management system. According to the method, high-precision estimation of battery parameters and capacity equalization in the module can be realized in parallel, and the precision of the capacity equalization process can be improved by the accurate parameter estimation result while the parameter estimation result is optimized by designing a high-information-degree signal.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle battery management and relates to a method for constructing a battery management system with multi-level coordinated control of modules and cells. Background Art

[0002] Electric vehicles are the main carrier of transportation electrification and are currently widely used in both light and heavy vehicles. However, with the development of electric vehicle technology, new electric vehicles will also face more complex and changeable application environments. For an electric vehicle's battery energy system, it is necessary to have the ability to cope with high-frequency heating in low temperature conditions, internal state detection during long-term operation, health status monitoring, and active balancing and SOC monitoring within a single cycle. Therefore, this also puts forward new requirements for the battery management system (BMS) and its corresponding battery management strategy.

[0003] Traditional battery management systems can solve one of the above problems well, designing system architecture for each requirement and proposing the optimal battery management strategy. Therefore, traditional BMS usually develops corresponding management strategies under a solidified software and hardware architecture. When the actual environment and working conditions of the battery are very different from the corresponding background in the design, the traditional BMS cannot adjust the strategy in time to provide a suitable management solution; at the same time, the structural limitations also make the existing BMS unable to meet all the above requirements at the same time in the application. Especially when the battery cell is in a large battery module, more battery management solutions currently manage the module as a whole, and can only obtain the electrical and thermal parameters of the module, making it difficult to achieve accurate status monitoring of each battery cell. Moreover, due to the limitations of the battery management system structure, the thermal management system of a large battery module usually only involves status monitoring and cannot be accurately intervened. Even if a problem with the battery cell in the module is found, it is only allowed to uniformly adjust all the cells in the module. This structural defect also limits the adaptive formulation of management strategies, greatly weakening the function of the battery management system.

[0004] Based on the current application background of battery management systems and the difficulties they face, there is an urgent need for a battery management system that can achieve multi-level coordinated control of modules and cells, and adjust the overall output of the battery module and control the local signals of the cells in the module through a high degree of freedom structure. This requires modifications to the overall architecture of the battery management system from hardware to software algorithms. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a method for constructing a battery management system with multi-level collaborative control of modules and cells, design a variable battery management system hardware structure and its related battery management algorithm, and realize signal control of any battery cell in a module with a series-parallel structure. Based on this invention, high and low frequency signals can be added simultaneously to meet the needs of low-temperature heating and parameter identification, as well as the balance of the battery cell status inside the module.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for constructing a battery management system with multi-level coordinated control of modules and cells, comprising the following steps:

[0008] Constructing a multi-level battery management system, including steps S1 to S2:

[0009] S1: Design the variable topology structure of the series-parallel circuit inside the battery module, which specifically includes the following steps:

[0010] S11: Design variable structure of series-parallel circuits;

[0011] S12: Establish a virtual model, traverse different battery series and parallel situations, build a complete series and parallel change switch strategy, and record the on and off conditions of the switches corresponding to different topologies through a list;

[0012] S13: constructing a physical circuit of the virtual model using power electronic components;

[0013] S2: Design a power electronic structure for hierarchical adaptive control of a single battery cell in a battery module, including a branch battery cell active selection structure and a battery current control unit; the branch battery cell active selection structure is a switch tube array connected to the battery module cell, and the switch tube array is used to actively select a specific battery cell in a branch and connect it to the battery current control unit; the battery current control unit is a phase-shifted full-bridge structure including a bidirectional DCDC structure;

[0014] S3: Based on the multi-level battery management system, an online parameter identification method at the battery cell level is established, and a dynamic balancing process at the module level under any environmental conditions is realized;

[0015] S4: Construct a multi-level battery management system module to obtain a complete module-cell multi-level coordinated control battery management system.

[0016] Further, the series-parallel circuit variable structure in step S11 includes a series-parallel trunk-branch switch unit, a series-parallel circuit form transformation unit and a test module;

[0017] The series-parallel trunk branch switch unit and the series-parallel circuit form conversion unit are both MOSFET or IGBT small current high frequency control switches;

[0018] The series-parallel trunk-branch switch unit is used to switch the series-parallel trunk and branches;

[0019] The series-parallel circuit form conversion unit is used to switch the series and parallel forms of the circuit;

[0020] The test module is used to test the voltage and current of each battery cell.

[0021] Furthermore, the design steps of the branch battery cell active selection structure are as follows:

[0022] S211: Design the connection circuit between the battery and the control unit to ensure that any battery can be connected to the control unit in any working state without affecting the overall output of the battery module; the positive and negative poles of each battery are independently connected to the DCDC control center and controlled by a bidirectional DCDC circuit;

[0023] S212: traverse the situations where different batteries are connected to the control unit, build a complete active switch selection strategy, and record the on-off situations of the switches corresponding to different topologies through a list;

[0024] S213: Build a battery on the simulation platform, load virtual working conditions and sensors, verify the operation of the circuit under different battery access states, and determine the final topology solution of the active selection structure;

[0025] S214: Considering the possible current and voltage amplitudes and switching frequencies of the battery, calculate the required switch tube model and establish a complete active selection structure for the multi-level collaborative control battery management system.

[0026] Further, the design steps of the control unit of the battery current are as follows:

[0027] S221: Design the main power electronic architecture of the control unit to ensure that energy exchange between batteries in the same branch of the battery is completed and the signal control of a single battery cell is realized while the control unit itself is safe;

[0028] S222: Design specific parameters of the bidirectional DCDC inside the control unit according to the current and voltage limits, ripple limits and target signal control effects of the battery itself, including the allocation and specific parameters of the MOSFET switches, inductors, capacitors and resistors therein;

[0029] S223: Design a bidirectional DCDC control strategy and give a specific PWM output method for multiple MOSFET switches.

[0030] S224: Build a battery on the simulation platform, load virtual working conditions and sensors, and verify the control ability of the control unit on a single battery signal under different trunk road conditions.

[0031] Further, in step S222, a phase-shifted full-bridge DCDC is used as a control unit, and the parameters to be calculated include the inductance and capacitance of the secondary side; the calculated values ​​of the secondary side electrical parameters are:

[0032]

[0033] Where L represents the inductance of the secondary side, V L is the voltage across the inductor, V L =V in -V out , V in is the secondary input voltage, V out is the secondary output voltage; D is the duty cycle, T s is the PWM period, I rip is the ideal current ripple amplitude, V rip is the ideal voltage ripple amplitude, C is the secondary capacitance;

[0034] In step S223, for the phase-shifted full-bridge circuit, the phase difference Δp between the two bridge arms is controlled by a proportional-integral control method in combination with the difference between the current to be controlled and the actual current signal:

[0035] Δp=PI(ΔI,K p ,K i )

[0036] Among them, PI represents the PI control function, the input is the current difference ΔI, the proportional coefficient K p And the integration coefficient K i , where the proportional integral coefficient is determined by the Ziegler-Nichols empirical method:

[0037] K p =0.45K u

[0038] T i =0.83P u

[0039]

[0040] Where K u In order to make the control system just reach the proportional gain value of sustained oscillation, P u is the oscillation period of the system when it reaches sustained oscillation;

[0041] In order to improve the control effect of the secondary circuit, a boost ratio is given.

[0042] Further, in step S3, based on the multi-level battery management system, an online parameter identification method at the battery cell level is established, and a dynamic balancing process at the module level under any environmental conditions is realized, which specifically includes the following steps:

[0043] S31: Establishing a method for selecting the optimal battery signal, establishing an information evaluation method for different signals, and providing a signal with the optimal information degree for different parameters;

[0044] S32: Establishing a battery SOC online estimation method and a parameter online identification method;

[0045] S33: Using the Model predictive control (MPC) optimization method, with battery balancing as the goal, taking into account the estimation of the internal parameters of the battery, an online balancing control strategy for multiple batteries is established.

[0046] Further, the step S31 specifically includes the following steps:

[0047] S311: First, establish the first-order RC equivalent circuit model of the battery, including the battery's ohmic internal resistance, polarization internal resistance, and polarization capacitance:

[0048]

[0049] Where I is the battery current, V p is the battery polarization voltage, R t is the polarization resistance, C p is the battery polarization capacitance, V is the battery terminal voltage, OCV represents the open circuit voltage, and R s is the ohmic internal resistance of the battery;

[0050] S312: Derive the sensitivity of voltage to different parameters, assuming that the battery current signal is decomposed into multiple cosine functions according to Fourier transform:

[0051] I=Mcos(ωt)

[0052] Where ω is the corresponding frequency of the cosine component, the sensitivity of voltage to different parameters is:

[0053]

[0054] Where τ is the battery time constant, which is the product of polarization internal resistance and polarization capacitance;

[0055] S313: Based on the sensitivity expression of the parameters, calculate the Cramér-Rao bound corresponding to the parameter set

[0056] var(θ)≥diag(F -1 )

[0057] Among them, F represents the Fisher matrix:

[0058]

[0059] In the formula Represents the variance of the test error of the voltage; According to the Cramér-Rao bound, the size of the optimal frequency is determined by the functional relationship.

[0060] Further, the step S32 specifically includes the following steps:

[0061] S321: Determine a battery online parameter estimation method, based on the first-order RC equivalent circuit model of the battery, and use least squares to calculate the battery electrical parameters as follows:

[0062]

[0063] θ (1) represents the prior parameter vector, θ (2) is the posterior parameter vector, is a coefficient matrix given according to the first-order RC equivalent circuit model of the battery; P(t) represents the covariance matrix in the process of estimating the battery parameter matrix to be identified using the least squares principle; λ represents the forgetting factor;

[0064] S322: Determine the calculation equation and observation equation of the battery state of charge SOC, realize the automatic iteration of the battery SOC in time and establish the relationship between SOC and observable variables:

[0065]

[0066] where x t is the actual observation parameter, which is a vector containing SOC; ω t represents the process noise, that is, in x t Error value in the process of automatic update over time;υ t Represents the test noise, that is, through x t The error generated in the process of calculating the battery terminal voltage V; calculate the Kalman gain and correct the average battery SOC at the current moment:

[0067]

[0068] P t represents the covariance matrix in the SOC estimation process, is the first updated P at time t t The value of P represents the second updated t The value of P at time t tOutput value; K t is the Kalman gain; P w is the process noise ω in step S322 t The variance, R ν is the test noise υ in step S322 t The variance of t is the difference between the test terminal voltage and the terminal voltage calculated in step S322.

[0069] Further, the step S33 specifically includes the following steps:

[0070] S331: Constructing the cost function in the MPC method

[0071]

[0072] Where SOC average represents the average SOC of multiple batteries; x(k+1) and SOC average (k+1) is calculated as follows:

[0073]

[0074] SOC average (k+1)=0.25(x(k+1)+SOC2(k+1)+SOC3(k+1)+SOC4(k+1))

[0075]

[0076] Where ΔT is the time step, C max is the maximum capacity, SOC i Indicates the SOC of the non-target battery;

[0077] S332: Establish boundary constraints and control objects in the MPC method:

[0078] -1C MPC ≤2C

[0079] Among them I MPC Indicates the current calculated by the MPC algorithm. The calculated current is ultimately loaded into the actual SOC and the mean SOC. average On the battery with the largest gap;

[0080] S333: Select an optimization method to implement MPC calculation.

[0081] Further, the multi-level battery management system module in step S4 includes:

[0082] ​The signal acquisition module of the battery cell in the battery module is used to directly obtain the voltage of the battery cell and convert the battery current signal into a voltage signal, which is then transmitted to the analog-to-digital conversion module;

[0083] The driver module of the battery management system switch array is used to load the starting level suitable for the operation of different types of switch tubes;

[0084] The conversion voltage conversion module adapted by the power electronic mechanism is used to convert the voltage of the external DC power supply to the voltage required by the drive module and the voltage required by the battery management system chip;

[0085] Electrical isolation modules for power electronics, which are used to ensure that two parts of the battery management system hardware are physically isolated but still pass signals or energy;

[0086] The core chip of the battery management system is used to store algorithm programs and output control signals of the switch tube array and high-frequency PWM signals.

[0087] The beneficial effects of the present invention are as follows: In view of the problems of poor intervention capability and poor adaptability of traditional battery management systems in complex environments, the present invention develops a battery management system with multi-level collaborative control of modules and cells, designs a power electronic hardware structure with a high degree of freedom and its corresponding control algorithm, and realizes precise control of any cell monomer in a large module with a series-parallel structure.

[0088] The specific advantages of the present invention are:

[0089] 1) The present invention realizes a variable series-parallel structure in the battery module, which can transfer the batteries of adjacent branches to the branches controlled by DCDC, ensuring that the control center of the battery management system can intervene in the signal of any battery cell in the module.

[0090] 2) The present invention only sets one DCDC control unit for one battery module, transfers the cost of the battery management system to the active selection structure, enhances the flexibility of the hardware, and improves the control efficiency in the coordinated control of multiple batteries.

[0091] 3) The present invention can realize high-precision estimation of battery parameters and capacity balancing within the module in parallel. While optimizing parameter estimation results by designing high-information signal, accurate parameter estimation results can also improve the accuracy of the capacity balancing process.

[0092] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0094] Figure 1 A simplified diagram of the steps for implementing a module-cell multi-level collaborative control battery management system according to the present invention;

[0095] Figure 2 (a) is a multi-branch module variable series-parallel power electronic structure, (b) is a schematic diagram of topology verification;

[0096] Figure 3 (a) shows the active selection architecture within the module, and (b) shows the topology verification diagram;

[0097] Figure 4 It is a schematic diagram of the DCDC control unit and its internal structure;

[0098] Figure 5 A schematic diagram of the power electronic hardware module of the module-cell multi-level collaborative control battery management system;

[0099] Figure 6 The figure is a result diagram of parameter identification and capacity balancing coordinated control, where (a) is the result of individual current control of a cell in the module and the identification result of the ohmic internal resistance and polarization internal resistance of the cell, and (b) is the result of time constant identification and the result of SOC balancing of multiple batteries. DETAILED DESCRIPTION

[0100] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0101] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0102] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0103] like Figure 1 As shown, the battery management system that can adaptively transform the topological structure within the module can realize hierarchical and multi-objective precise control from the battery module level to the battery cell level. The battery management system includes the following power electronic structure components and specific algorithm embedding implementation steps:

[0104] S1: Design of variable topology structure for series-parallel circuits inside battery modules;

[0105] S2: Selection and design of power electronic structure to achieve hierarchical adaptive control of individual cells within the module;

[0106] S3: Based on a multi-level battery management system, an online parameter identification method at the battery cell level is established, and a dynamic balancing process at the module level is realized under any environmental conditions;

[0107] S4: Embed the algorithm part to build a complete multi-level collaborative battery management system architecture.

[0108] Furthermore, step S1 specifically includes the following steps:

[0109] S11: Design variable structure of series-parallel circuits;

[0110] S12: Establish a virtual model, traverse the topology changes, and verify the rationality of different topologies of the variable structure;

[0111] S13: Select appropriate power electronic components to realize theoretical mechanism action.

[0112] Optionally, the circuit structure in step S11 includes:

[0113] S111: Figure 2 As shown in (a)-(b) in the figure, the series-parallel trunk branch switch unit, the switch here is an opposed MOSFET low-current high-frequency control switch, and a corresponding driver; the switch tubes are opposed to ensure that the switch can truly complete the circuit on-off action in the bidirectional load current;

[0114] S112: Figure 2As shown in (a)-(b) of the figure, the series-parallel circuit form conversion unit, the switch here is an opposed MOSFET low-current high-frequency control switch, and a corresponding driver; the switch tubes are opposed to ensure that the switch can truly complete the circuit on-off action in the bidirectional load current;

[0115] S113: A voltage and current test module corresponding to each battery cell, wherein the test module is an ADC module of a battery management system and an auxiliary resistance part necessary for the test current.

[0116] Optionally, the method for establishing a virtual model of the variable series-parallel mechanism in step S12 includes:

[0117] S121: By traversing different battery series and parallel situations, verify whether the designed structure is applicable to different topologies, and build a complete series and parallel change switch strategy. The switch strategy here is a list that records the on and off conditions of the switches corresponding to different topologies. Figure 2 As shown in (a)-(b) in the figure, this structure can interchange the positions of the corresponding batteries of two adjacent branches.

[0118] S122: Build a battery on the simulation platform, load virtual working conditions and sensors, and verify the rationality of different series-parallel topologies in S121.

[0119] Further, the step S2 specifically includes the following steps:

[0120] S21: A branch battery cell active selection structure, specifically including a switch tube array with a designed topology structure and a line connection between the battery module monomer, wherein the switch tube array can be an opposing MOSFET pair, or an IGBT switch, etc., and its function is to actively select a specific battery cell in a branch and connect it to the control unit described in S22;

[0121] S22: a control unit for battery current, wherein the control unit is a phase-shifted full-bridge structure including a bidirectional DCDC structure;

[0122] S23: Driving components and power supply components of all the above electronic components.

[0123] Optionally, the complete structure described in step S21 has the following composition and the following steps are required to establish the structure:

[0124] S211: Design the connection circuit between the battery and the control unit to ensure that any battery can be connected to the control unit in any working state without affecting the overall output of the battery module; Figure 3 As shown in (a)-(b), each positive and negative electrode of the battery can be independently connected to the DCDC control center and controlled by a bidirectional DCDC circuit;

[0125] S212: By traversing the situations where different batteries are connected to the control unit, verify whether the designed structure is applicable to different topologies, and build a complete active selection switch strategy. The switch strategy here is a list recording the on-off conditions of the switches corresponding to different topologies; Figure 3 As shown in (a)-(b) in the figure, there are four batteries connected in series in the branch, battery 1 is connected to the secondary side, and battery 2 is connected to the primary side. By traversing all the situations, any battery can be independently connected to the DCDC control center, and when there is energy exchange between any two batteries, they will not affect each other.

[0126] S213: Build a battery on the simulation platform, load virtual working conditions and sensors, verify the operation of the circuit under different battery access states in S212, and determine the final topology solution of the active selection structure;

[0127] S214: Considering the possible current and voltage amplitudes and switching frequencies of the battery, calculate the required switch tube model and establish a complete active selection structure for the multi-level coordinated control battery management system. For most operating conditions of the battery, the current amplitude flowing through the switch tube does not exceed 100A and the frequency is not higher than 100kHz. Therefore, ONSEMI's FCA47N60 model MOSFET switch tube is selected here.

[0128] Optionally, the complete structure described in step S22 has the following composition and the following steps are required to establish the structure:

[0129] S221: Design the main power electronic architecture of the control unit to ensure that the energy exchange between batteries in the same branch of the battery can be completed and the signal control of a single battery cell can be realized under the condition of the safety of the control unit itself;

[0130] S222: After determining the basic architecture of the battery control, design the specific parameters of the bidirectional DCDC inside the control unit according to the battery's own current and voltage limits, ripple limits, and target signal control effects, including the allocation and specific parameters of the MOSFET switches, inductors, capacitors, and resistors.

[0131] Among them, this patent uses a phase-shifted full-bridge DCDC as a control unit, such as Figure 4 As shown in the figure, the parameters that need to be calculated include the inductance and capacitance of the secondary side. In a general phase-shifted full bridge, the PWM duty is set to a constant value of 0.5, assuming that the PWM frequency is 50kHz. L represents the inductance of the secondary side, V L is the voltage on the inductor, D is the duty cycle, T s is the PWM period, I rip is the ideal current ripple amplitude, V rip is the ideal voltage ripple amplitude, C is the secondary capacitor. In addition, the secondary input voltage Vin =4.2V, output voltage V out =4V, then V L =V in -V out I rip Can be set to 0.3 times the nominal current, V rip It can be set to 0.1 times the output voltage. Then, the calculated value of the secondary electrical parameters is:

[0132]

[0133] Where L represents the inductance of the secondary side, V L is the voltage on the inductor, D is the duty cycle, T s is the PWM period, I rip is the ideal current ripple amplitude, V rip is the ideal voltage ripple amplitude, C is the secondary capacitance;

[0134] S223: Design a bidirectional DCDC control strategy and give a specific Pulse Width Modulation (PWM) output method for multiple MOSFET switches. For a phase-shifted full-bridge circuit, the PWM duty cycle and frequency of each MOSFET are usually determined, and what needs to be controlled is the phase difference Δp between the two bridge arms. This phase difference can be achieved by using a proportional-integral control method, combined with the difference between the current to be controlled and the actual current signal:

[0135] Δp=PI(ΔI,K p ,K i )

[0136] Among them, PI represents the PI control function, the input is the current difference ΔI, the proportional coefficient K p And the integration coefficient K i The proportional integral coefficient can be determined by the Ziegler-Nichols empirical method:

[0137] K p =0.45K u

[0138] T i =0.83P u

[0139]

[0140] Where K u In order to make the control system just reach the proportional gain value of sustained oscillation, P u is the oscillation period of the system when it reaches sustained oscillation. After simulation testing, K u =90 and P u=0.15s, so

[0141] K p =0.45K u ≈50

[0142] T i =0.83P u ≈0.01s

[0143]

[0144] In addition, the phase-shifted full bridge is a step-down DCDC. Therefore, in order to achieve a better control effect on the secondary circuit, a boost ratio can be given, such as Figure 4 The turns ratio of L2 and L3 shown is 1:1.5.

[0145] S224: Build a battery on the simulation platform, load virtual working conditions and sensors, and verify the control ability of the control unit on a single battery signal under different trunk road conditions.

[0146] Further, step S3 specifically includes the following steps:

[0147] S31: Establishing a method for selecting the optimal battery signal, establishing an information evaluation method for different signals, and providing a signal with the optimal information degree for different parameters;

[0148] S32: Establish a battery SOC online estimation method and a parameter online identification method to provide a numerical reference for battery module balancing;

[0149] S33: Using the Model predictive control (MPC) optimization method, with battery balancing as the goal, taking into account the estimation of the internal parameters of the battery, an online balancing control strategy for multiple batteries is established;

[0150] S311: First, establish the first-order RC equivalent circuit model of the battery, including the ohmic internal resistance, polarization internal resistance and polarization capacitance of the battery:

[0151]

[0152] Where I is the battery current, V p is the battery polarization voltage, R t is the polarization resistance, C p is the battery polarization capacitance, V is the battery terminal voltage, OCV represents the open circuit voltage, and R s is the ohmic internal resistance of the battery;

[0153] S312: Derive the sensitivity of voltage to different parameters under the model system. Assume that the battery current signal is decomposed into multiple cosine functions according to Fourier transform:

[0154] I=Mcos(ωt)

[0155] Where ω is the corresponding frequency of the cosine component, the sensitivity of voltage to different parameters is:

[0156]

[0157] Where τ is the battery time constant, which is the product of polarization internal resistance and polarization capacitance.

[0158] S313: Based on the sensitivity expression of the parameters, calculate the Cramér-Rao bound corresponding to the parameter set

[0159] var(θ)≥diag(F -1 )

[0160] Among them, F represents the Fisher matrix:

[0161]

[0162] In the formula The variance of the voltage test error is represented by . According to the above Cramér-Rao bound, the optimal frequency can be determined through the functional relationship. Finally, it can be calculated that when the battery signal has the following characteristics

[0163] I=Mcos(ωt)+Mcos(kωt)

[0164] For R t and τ, the optimal frequency is between 0.01Hz and 0.001Hz, and for R s The optimal frequency is usually greater than 0.1 Hz. Optionally, the method for establishing an online SOC estimation method and a parameter identification method of a battery in step S32 specifically includes:

[0165] S321: Determine the battery online parameter estimation method. The basic battery model has been given in S311. Based on this model, the battery electrical parameters can be calculated based on the least squares method.

[0166]

[0167] θ (1) represents the prior parameter vector, θ (2) is the posterior parameter vector, It is the coefficient matrix given by the model in S311. P(t) represents the covariance matrix in the process of estimating the battery parameter matrix to be identified using the least squares principle. Its original expression is This patent uses a recursive form. λ represents the forgetting factor, which needs to be manually selected and is generally between 0.9 and 1.

[0168] S322: Determine the calculation equation and observation equation of the battery state of charge (SOC), realize the automatic iteration of the battery SOC in time and establish the relationship between SOC and observable variables

[0169]

[0170] where x t is the actual observation parameter, which is a vector containing SOC, ω t represents the process noise, that is, in x t The error value in the process of automatically updating over time, v t It represents the test noise, that is, through x t The error generated in the process of calculating the battery terminal voltage V. Calculate the Kalman gain and correct the average battery SOC at the current moment

[0171]

[0172] P t It represents the covariance matrix in the SOC estimation process and has no actual physical meaning. Specifically, is the first updated P at time t t The value of P represents the second updated t The value of P at time t t Output value. K t In the general SOC estimation algorithm, it is called Kalman gain. w is the process noise ω in S322 t The variance, R v is the test noise v in S322 t The variance of t is the difference between the test terminal voltage and the terminal voltage calculated in S322.

[0173] Optionally, the method of using the MPC optimization method to achieve battery balancing in step S33 specifically includes:

[0174] S331: Constructing the cost function in the MPC method

[0175]

[0176] Where SOC average represents the average SOC of multiple batteries, and x(k+1) and SOC average (k+1) can be calculated as follows:

[0177]

[0178] SOC average (k+1)=0.25(x(k+1)+SOC2(k+1)+SOC3(k+1)+SOC4(k+1))

[0179]

[0180] Where ΔT is the time step, usually 1s, C max is the maximum capacity, SOC i It represents the SOC of the non-target battery, such as Figure 3 As shown in (a)-(b) in FIG. 5 , assuming that there are four batteries connected in series in a branch, the maximum value of i is 4.

[0181] S332: Establish boundary constraints and control objects in the MPC method

[0182] -1C MPC ≤2C,

[0183] Among them I MPC It represents the current calculated by the MPC algorithm. The calculated current will eventually be loaded into the actual SOC and the mean SOC. average The battery with the biggest difference.

[0184] S333: Select an optimization method to implement MPC calculation.

[0185] Further, the step S4 specifically includes the following steps:

[0186] S41: The signal acquisition module of the battery cell in the battery module is used to directly obtain the voltage of the battery cell and convert the current signal of the battery into a voltage signal, which is then transmitted to the analog-to-digital conversion module for easy reading by the chip of the battery management system. Figure 5 As shown, the OPA4277PA chip can be used to complete the task of digital-to-analog conversion.

[0187] S42: The driving module of the switch array of the battery management system is used to load the starting level suitable for the operation of different types of switch tubes. The driving principle of the switch array is usually as follows: Figure 5 As shown, TLP250(F) can be used to drive the MOSFET switch tube.

[0188] S43: A conversion voltage conversion module adapted to the power electronic mechanism, used to convert the voltage of the external DC power supply to the voltage required for driving in S42, the voltage required by the battery management system chip, etc. ​

[0189] S44: The electrical isolation module of power electronics ensures that the two parts of the battery management system hardware are physically isolated but still transmit signals or energy, avoiding the impact of high-voltage side faults on the low-voltage side and eliminating system noise caused by different grounding points of different components. Electrical isolation is also achieved using the TLP250 chip.

[0190] S45: The core chip of the battery management system with algorithm burning, used to store programs and output control signals of the switch array and high-frequency PWM signals. Since there are matrix operations in the algorithm, the DSP28335 chip can be used for calculation.

[0191] S46: Cooling element for battery management system.

[0192] It should also be noted that the single control result of the above battery management system is as follows: Figure 6 As shown in (a)-(b) in the figure, (a) shows the result of the battery management system's individual control of the current of a battery in any branch, as well as the parameter identification of the ohmic internal resistance and polarization internal resistance under the current condition; (b) shows the result of the identification of the battery time constant and the result of SOC balancing of multiple batteries in the branch at the same time as the parameter identification.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A method for constructing a battery management system with multi-level coordinated control of modules and cells, characterized in that: The following steps are involved: Constructing a multi-level battery management system, including steps S1 to S2: S1: Design the variable topology structure of the series-parallel circuit inside the battery module, which specifically includes the following steps: S11: Design variable structure of series-parallel circuits; S12: Establish a virtual model, traverse different battery series and parallel situations, build a complete series and parallel change switch strategy, and record the on and off conditions of the switches corresponding to different topologies through a list; S13: constructing a physical circuit of the virtual model using power electronic components; S2: Design a power electronic structure for hierarchical adaptive control of a single battery cell in a battery module, including a branch battery cell active selection structure and a battery current control unit; the branch battery cell active selection structure is a switch tube array connected to the battery module cell, and the switch tube array is used to actively select a specific battery cell in a branch and connect it to the battery current control unit; the battery current control unit is a phase-shifted full-bridge structure including a bidirectional DCDC structure; S3: Based on the multi-level battery management system, an online parameter identification method at the battery cell level is established, and a dynamic balancing process at the module level under any environmental conditions is realized; S4: Construct a multi-level battery management system module to obtain a complete module-cell multi-level coordinated control battery management system.

2. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 1, characterized in that: Step S11: the series-parallel circuit variable structure includes a series-parallel trunk-branch switch unit, a series-parallel circuit form transformation unit and a test module; The series-parallel trunk branch switch unit and the series-parallel circuit form conversion unit are both MOSFET or IGBT small current high frequency control switches; The series-parallel trunk-branch switch unit is used to switch the series-parallel trunk and branches; The series-parallel circuit form conversion unit is used to switch the series and parallel forms of the circuit; The test module is used to test the voltage and current of each battery cell.

3. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 1, characterized in that: The design steps of the branch battery cell active selection structure are as follows: S211: Design the connection circuit between the battery and the control unit to ensure that any battery can be connected to the control unit in any working state without affecting the overall output of the battery module; the positive and negative poles of each battery are independently connected to the DCDC control center and controlled by a bidirectional DCDC circuit; S212: traverse the situations where different batteries are connected to the control unit, build a complete active switch selection strategy, and record the on-off situations of the switches corresponding to different topologies through a list; S213: Build a battery on the simulation platform, load virtual working conditions and sensors, verify the operation of the circuit under different battery access states, and determine the final topology solution of the active selection structure; S214: Considering the possible current and voltage amplitudes and switching frequencies of the battery, calculate the required switch tube model and establish a complete active selection structure for the multi-level collaborative control battery management system.

4. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 1, characterized in that: The design steps of the battery current control unit are as follows: S221: Design the main power electronic architecture of the control unit to ensure that energy exchange between batteries in the same branch of the battery is completed and the signal control of a single battery cell is realized while the control unit itself is safe; S222: Design specific parameters of the bidirectional DCDC inside the control unit according to the current and voltage limits, ripple limits and target signal control effects of the battery itself, including the allocation and specific parameters of the MOSFET switches, inductors, capacitors and resistors therein; S223: Design a bidirectional DCDC control strategy and give a specific PWM output method for multiple MOSFET switches. S224: Build a battery on the simulation platform, load virtual working conditions and sensors, and verify the control unit's ability to control a single battery signal under different trunk road conditions.

5. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 4, characterized in that: In step S222, a phase-shifted full-bridge DCDC is used as a control unit, and the parameters to be calculated include the inductance and capacitance of the secondary side; the calculated values ​​of the secondary side electrical parameters are: Where L represents the inductance of the secondary side, V L is the voltage across the inductor, V L =V in -V out , V in is the secondary input voltage, V out is the secondary output voltage; D is the duty cycle, T s is the PWM period, I rip is the ideal current ripple amplitude, V rip is the ideal voltage ripple amplitude, C is the secondary capacitance; In step S223, for the phase-shifted full-bridge circuit, the phase difference Δp between the two bridge arms is controlled by a proportional-integral control method in combination with the difference between the current to be controlled and the actual current signal: Δp=PI(ΔI,K p ,K i ) Among them, PI represents the PI control function, the input is the current difference ΔI, the proportional coefficient K p And the integration coefficient K i , where the proportional integral coefficient is determined by the Ziegler-Nichols empirical method: K p =0.45K u T i =0.83P u Where K u In order to make the control system just reach the proportional gain value of sustained oscillation, P u is the oscillation period of the system when it reaches sustained oscillation; In order to improve the control effect of the secondary circuit, a boost ratio is given.

6. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 1, characterized in that: Step S3 establishes an online parameter identification method at the battery cell level based on the multi-level battery management system, and realizes a dynamic balancing process at the module level under any environmental conditions, which specifically includes the following steps: S31: Establishing a method for selecting the optimal battery signal, establishing an information evaluation method for different signals, and providing a signal with the optimal information degree for different parameters; S32: Establishing a battery SOC online estimation method and a parameter online identification method; S33: Using the MPC optimization method, with battery balancing as the goal, taking into account the estimation of the internal parameters of the battery, an online balancing control strategy for multiple batteries is established.

7. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 6, characterized in that: The step S31 specifically includes the following steps: S311: First, establish the first-order RC equivalent circuit model of the battery, including the battery's ohmic internal resistance, polarization internal resistance, and polarization capacitance: Where I is the battery current, V p is the battery polarization voltage, R t is the polarization resistance, C p is the battery polarization capacitance, V is the battery terminal voltage, OCV represents the open circuit voltage, and R s is the ohmic internal resistance of the battery; S312: Derive the sensitivity of voltage to different parameters, assuming that the battery current signal is decomposed into multiple cosine functions according to Fourier transform: I=Mcos(ωt) Where ω is the corresponding frequency of the cosine component, the sensitivity of voltage to different parameters is: Where τ is the battery time constant, which is the product of polarization internal resistance and polarization capacitance; S313: Based on the sensitivity expression of the parameters, calculate the Cramér-Rao bound corresponding to the parameter set var(θ)≥diag(F -1 ) Among them, F represents the Fisher matrix: In the formula Represents the variance of the test error of the voltage; According to the Cramér-Rao bound, the size of the optimal frequency is determined by the functional relationship.

8. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 6, characterized in that: The step S32 specifically includes the following steps: S321: Determine a battery online parameter estimation method, based on the first-order RC equivalent circuit model of the battery, and use least squares to calculate the battery electrical parameters as follows: θ (1) represents the prior parameter vector, θ (2) is the posterior parameter vector, is a coefficient matrix given according to the first-order RC equivalent circuit model of the battery; P(t) represents the covariance matrix in the process of estimating the battery parameter matrix to be identified using the least squares principle; λ represents the forgetting factor; S322: Determine the calculation equation and observation equation of the battery state of charge SOC, realize the automatic iteration of the battery SOC in time and establish the relationship between SOC and observable variables: where x t is the actual observation parameter, which is a vector containing SOC; ω t represents the process noise, that is, in x t Error value in the process that is automatically updated over time; v t Represents the test noise, that is, through x t The error generated in the process of calculating the battery terminal voltage V; calculate the Kalman gain and correct the average battery SOC at the current moment: P t represents the covariance matrix in the SOC estimation process, is the first updated P at time t t The value of P represents the second updated t The value of P at time t t Output value; K t is the Kalman gain; R w is the process noise ω in step S322 t The variance, R v is the test noise v in step S322 t The variance of t is the difference between the test terminal voltage and the terminal voltage calculated in step S322.

9. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 6, characterized in that: The step S33 specifically includes the following steps: S331: Constructing the cost function in the MPC method Where SOC average represents the average SOC of multiple batteries; x(k+1) and SOC average (k+1) is calculated as follows: SOC average (k+1)=0.25(x(k+1)+SOC2(k+1)+SOC3(k+1)+SOC4(k+1)) Where ΔT is the time step, C max is the maximum capacity, SOC i Indicates the SOC of the non-target battery; S332: Establish boundary constraints and control objects in the MPC method: -1C<I MPC ≤2C Among them I MPC Indicates the current calculated by the MPC algorithm. The calculated current is ultimately loaded into the actual SOC and the mean SOC. average On the battery with the largest gap; S333: Select an optimization method to implement MPC calculation.

10. The method for constructing a battery management system with multi-level coordinated control of modules and cells according to claim 1, characterized in that: The multi-level battery management system module in step S4 includes: The signal acquisition module of the battery cell in the battery module is used to directly obtain the voltage of the battery cell and convert the battery current signal into a voltage signal, which is then transmitted to the analog-to-digital conversion module; The driver module of the battery management system switch array is used to load the starting level suitable for the operation of different types of switch tubes; The conversion voltage conversion module adapted by the power electronic mechanism is used to convert the voltage of the external DC power supply to the voltage required by the drive module and the voltage required by the battery management system chip; Electrical isolation modules for power electronics, which are used to ensure that two parts of the battery management system hardware are physically isolated but still pass signals or energy; The core chip of the battery management system is used to store algorithm programs and output control signals of the switch tube array and high-frequency PWM signals.

Images