Power distribution method for redundant electric drive system

Abstract

The present invention proposes a power distribution method for a redundant electric drive system. Based on the driving target on the axis and the state and parameters of the battery motor, the constraint level of the battery is determined to set the maximum power of each battery group; the optimal power configuration of the battery group is calculated, and the power matrix, disturbance term, and target power vector are set accordingly; the constraint level of each motor is determined and the power matrix and disturbance term are adjusted accordingly; according to the full rank condition of the power matrix and the rationality condition of the power distribution ratio, it is judged whether a reasonable power ratio can be obtained. If the condition is not met, the constraints are removed layer by layer from low to high until the condition is met, otherwise the upper controller or module is notified for processing; according to the obtained power ratio, the output power of each motor is calculated; the present invention provides a general method for optimizing the power distribution of a redundant electric drive system under different levels of constraint conditions, simplifies design and development, reduces development cycle and cost, and has low calculation cost, and can be implemented in a real-time control system.

Description

Technical Field

[0001] The invention relates to a power distribution method for a redundant electric drive system. Background Art

[0002] Distributed electric drive systems can be used in various types of transportation, such as airplanes, cars, ships, rail vehicles, etc. Their superiority lies in whether they can utilize the inherent redundancy of distributed electric drive systems to fully improve the driving efficiency, safety, and reliability of the transportation system. For example, multiple electric motors are connected to multiple batteries as redundant power sources to drive a rotating shaft as an output terminal.

[0003] The publication number of the application filed by the applicant on September 5, 2022 is CN115431794A, and its name is Control Architecture of Distributed Electric Drive System, in which the power distribution module adopts a redundant design; a redundant electric drive system refers to an electric drive system with redundant characteristics composed of multiple motors and multiple batteries with a pre-designed electrical connection architecture, and these motors and batteries are constructed into a power conduction network to achieve the driving goal; the redundant electric drive system is distributed in electrical structure, but not necessarily distributed in physical structure, and can be centralized; the driven object, that is, the upper system, can be distributed, that is, there are multiple independent driving targets, or it can be centralized, with only one driving target; because the system has redundant characteristics, even if some driving components fail or some driving components are restricted, as long as the control freedom is still greater than the control target, there can be multiple different power distribution schemes to achieve the same driving target; in this way, this power conduction network can not only realize fault-tolerant control, but also realize optimal drive control with efficiency balance and minimum loss as the goal; the present application aims to propose an easy-to-implement method for optimizing the power distribution of a redundant electric drive system, which can realize the optimal fault-tolerant control of a redundant electric drive system online at the lowest computational cost.

[0004] Chinese patent CN113541299A proposes a load power distribution method, device and uninterruptible power supply parallel redundant system. Although a load power distribution method is proposed, the method only proposes a method for controlling all the uninterruptible power supply devices to evenly share the load power of the load when the power supply modes of all the uninterruptible power supply devices are in the battery power supply mode. There is neither a fault-tolerant strategy nor any optimization calculation measures. It is fundamentally unable to meet the current needs of transportation electrification applications with extremely high requirements for efficiency, safety and reliability. It is not suitable for the power distribution of the redundant electric drive system targeted by the present invention. If this method is used to control the electric drive system, it will lead to poor reliability and safety of the controlled system and low efficiency. When one or more electric drive units fail or are restricted, the system cannot adjust the power distribution, not to mention considering the priority of various factors, so as to optimize the performance of the overall system in real time during operation, and cannot meet the needs of key fields such as electric vehicles and aerospace. Summary of the invention

[0005] The purpose of the present invention is to solve the problem of lack of online optimal fault-tolerant control of redundant electric drive systems in the prior art, and to propose a power distribution method for redundant electric drive systems that can be operated online in real time, aiming to improve the reliability and efficiency of the electric drive system, and to achieve dynamic power distribution between motors under various constraints by optimizing the coordinated work of multiple electric drive units (motors and batteries) in real time; and when one or more electric drive units fail or are restricted, the system can automatically adjust the power distribution to ensure that the remaining units can continue to operate efficiently and stably; it can not only improve the fault tolerance of the system, but also extend the service life of the equipment, and is suitable for key fields such as electric vehicles, aerospace, etc.

[0006] To achieve the above object, the present invention proposes a redundant electric drive system power distribution method, comprising the following steps:

[0007] a) Obtaining axis output requirements: Obtaining axis output requirements from the upper controller or control module;

[0008] b) Obtaining the state parameters of each battery and motor: Obtaining the state and various parameters of the battery and motor from the front-end control module;

[0009] c) Calculate the maximum output power of each battery group: determine the constraint level of each battery and calculate the maximum output power of each battery group;

[0010] d) Setting the maximum output power of each motor: determining the constraint level of each motor and setting the maximum output power of each motor;

[0011] e) Determine whether the motor meets the demand: Determine whether the current motor can meet the output demand on the axis under the determined constraints. If not, remove the motor constraints layer by layer from low to high until the output demand on the axis is met. Otherwise, report to the upper controller or control module: some output requirements on the axis cannot be met and the target needs to be adjusted;

[0012] f) Determine whether the power allocation ratio is reasonable: Calculate the power ratio and determine whether the power allocation ratio is reasonable. If it is unreasonable, remove the battery constraints layer by layer from low to high until the reasonable conditions are met. Otherwise, report to the upper controller or control module: the output requirements on some axes cannot be met and the target needs to be adjusted;

[0013] g) Calculate motor output power: Calculate the output power of each motor based on the power ratio.

[0014] Preferably, in step c), when calculating the maximum output power of each battery pack, within the maximum power range, the output power of each battery pack is adjusted according to the battery parameter settings to achieve global optimization; according to the optimal power configuration of the battery pack, the power matrix p, the disturbance term p0, and the target power vector A are set.

[0015] Preferably, in the step e), it is determined whether the current motor meets the on-axis output requirements under the determined constraints, by judging whether the power matrix p is full rank. If it is not full rank, it indicates that the motor constraints are too many and the on-axis output target cannot be achieved. In this case, the motor constraints are removed layer by layer from low to high until the full rank condition is met.

[0016] Preferably, the power ratio is calculated in step f) by matrix calculation to obtain the distribution control vector r=p T (pp T ) -1 (A-p0).

[0017] Preferably, in step f), whether the power allocation ratio is reasonable can be determined by judging whether the calculated power allocation ratio is in a reasonable range, such as [0,1]. If it is not in this range, it indicates that there are too many battery constraints. In this case, the battery constraints are removed layer by layer from low to high until all power allocation ratios are in the selected range.

[0018] Preferably, the system involved in the method comprises n battery packs, wherein the battery pack i is composed of h i Batteries are connected in parallel, i=1,2,…,n; m motors; control l driving targets, the specific steps are:

[0019] A) Obtain a drive target A from the upper controller or control module i , i=1,2,…,l; obtain the status and parameters of each battery and motor from the front-end control program; initialize the program and assign initial values ​​to each control variable;

[0020] B) Determine the constraint level p of each battery according to its status k , k=1,2,…,a; the smaller k is, the higher the constraint level is, that is, the higher the priority is; at the same time, under the premise of meeting the power demand on the axis, the constraint level is adjusted from high to low to the optimal level, and the maximum power that each battery can provide is determined;

[0021] C) determining the maximum output power of each battery pack according to the constraint level of each battery and the combination of each battery pack;

[0022] D) According to the health status parameter SoH of each battery obtained from the front-end control program ij ,i=1,2,…,n,j=1,2,…,h i , and the battery pack's state of charge SOC i , i=1,2,…,n, adjust and optimize the output power of the battery pack to meet the power demand on the axis within the maximum output power of each battery pack while balancing the health status and charge state of each battery pack;

[0023] E) Determine the constraint level P of each motor according to its status k , k=1,2,…,b; the smaller k is, the higher the constraint level is, that is, the higher the priority is; at the same time, under the premise of meeting the power demand on the shaft, the constraint level is adjusted from high to low to the optimal level, and the maximum power that each motor can provide is determined;

[0024] F) Determine the output power of the constrained motor; then set the power matrix p and the disturbance term p0 according to the constrained motor; p is an l×(mn) matrix, and p0 is an l-dimensional term vector; and the full rank condition of the power matrix p is used to determine whether the motor constraints can meet the power requirements of each axis. If p is not full rank, it means that there are too many motor constraints, then the lowest motor constraint condition is abandoned layer by layer, that is, higher-level motor constraints are retained as much as possible until a full-rank p is obtained, and the allocation control vector r=p is calculated based on this. T (pp T ) -1 (A-p0), ;

[0025] G) Determine the power allocation ratio , j=1,2,…,mn, is it in a reasonable range, that is, between [0,1]. If there is a power allocation ratio that is not in the reasonable range, it means that there are too many battery constraints. Then, the lowest battery constraint condition is abandoned layer by layer, that is, higher-level battery constraints are retained as much as possible until all calculated power allocation ratios are in a reasonable range;

[0026] H) According to the power transmission matrix R mxnCalculate m r ji The value of , i∈{1,2,…,m}, j∈{1,2,…,n}, and r ji The output power of all unconstrained motors is calculated and combined with the output power of the constrained motors obtained in F), and then transmitted to the lower-level controller in the form of commands to control each motor to reach the required power, thereby meeting the power requirements on all axes while achieving an optimized balance between the battery and the motor.

[0027] Preferably, the battery constraint level p k , k=1,2,…,a, is a level, which can be divided into: Level 1 is failure, the highest level, that is, physical hardware limitation; Level 2 is restriction, that is, mandatory in non-emergency situations; Level 3 to level a-1 are different levels of optimization; Level a is completely free; According to the priority from high to low: p 1 >p 2 >p 3 >…>p a-1 >p a .

[0028] As a preference, the motor's constraint level P k , k=1,2,…,b, there are b levels, which can be divided into: Level 1 is failure, the highest level, that is, physical hardware limitation; Level 2 is limitation, that is, mandatory in non-emergency situations; Level 3 to level b-1 are different levels of optimization; Level b is completely free; According to the priority from high to low: P 1 >P 2 >P 3 >…>P b-1 >P b .

[0029] Preferably, the dimensions of the power matrix p and the allocation control vector r are reduced from m to mn to reduce the amount of calculation.

[0030] Preferably, the power matrix p is composed of 0 and the optimized power of each battery pack, and the disturbance term p0 is composed of a combination of 0 and the optimized power of each battery pack and the power of the constrained motor.

[0031] Beneficial effects of the present invention: The present invention solves the power allocation optimization problem of redundant electric drive systems under different priority constraints, improves the reliability and efficiency of the electric drive system, and realizes dynamic power allocation between motors under various constraints by optimizing the coordinated work of multiple electric drive units (motors and batteries) in real time: according to factors such as motor efficiency, thermal management, battery health status, battery charge balance, and load demand, and considering the priority of each factor, the overall system performance is optimized; and when one or more electric drive units fail or are restricted, the system can automatically adjust the power allocation to ensure that the remaining units can continue to operate efficiently and stably. This method not only improves the fault tolerance of the system, but also extends the service life of the equipment, and is suitable for key fields such as electric vehicles, aerospace, etc.

[0032] First, the method is designed for the purpose of online real-time operation, has low computational cost, and can be implemented in existing real-time control systems; it is not an optimization method that is only suitable for offline operation.

[0033] Second, it can handle different levels of constraints on batteries and motors, that is, it contains multiple layers of soft constraints, rather than only one level of hard constraints in usual optimization problems; and each layer of soft constraints allows dynamic changes, that is, new constraints can be added according to actual conditions, or existing constraints can be removed or changed. When all constraints cannot be met, this method adjusts the lowest level of constraints that can be met, specifically addresses different levels of safety and reliability requirements, and at the same time pursues the highest efficiency and best balance of batteries and motors, maximizing the life, reliability, and range of the electric drive system.

[0034] Third, even when some drive components fail, as long as the integrity of the power conduction network is maintained, that is, there is no inability to meet the axis requirements, there is no need for the upper-level controller or upper-level control module to intervene, that is, the upper-level control design is transparent, simplifying the design of the upper-level controller or upper-level control module.

[0035] Fourth, the present invention is a systematic method that provides a general solution for all redundant electric drive systems, rather than requiring the design of a specific algorithm to solve each specific structure. It is universal, thus simplifying the design and development process and reducing development cycle and cost.

[0036] Fifth, this method cooperates with motor control, and the power optimization process does not cause any unevenness in the output on the shaft.

[0037] The features and advantages of the present invention will be described in detail through embodiments in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is an overall flow chart of the power distribution method of the redundant electric drive system of the present invention;

[0039] Figure 2 It is a flowchart of the initialization module of step A in the power distribution method of the redundant electric drive system of the present invention;

[0040] Figure 3 It is a flow chart of a battery constraint level establishment module in step B of a redundant electric drive system power allocation method of the present invention;

[0041] Figure 4 This is a flow chart of a battery pack power boundary establishment module in step C of the redundant electric drive system power allocation method of the present invention;

[0042] Figure 5 It is a flow chart of the battery pack power optimization module in step D of the redundant electric drive system power allocation method of the present invention;

[0043] Figure 6 It is a flow chart of the module for establishing the motor constraint level and power boundary in step E of the redundant electric drive system power allocation method of the present invention;

[0044] Figure 7 It is a flow chart of the motor constraint annealing reset and module in step F of the redundant electric drive system power allocation method of the present invention;

[0045] Figure 8 It is a flow chart of the battery constraint annealing reset module in step G of the redundant electric drive system power allocation method of the present invention;

[0046] Fig. 9 It is a flow chart of the power distribution ratio and motor output power calculation module of the H step in the power distribution method of the redundant electric drive system of the present invention;

[0047] Fig.10 Schematic diagram of hardware connection of the first embodiment of the present invention;

[0048] Fig.11 is a schematic diagram of hardware connection of the second embodiment of the present invention;

[0049] Fig.12 It is a schematic diagram of hardware connection of the third embodiment of the present invention. DETAILED DESCRIPTION

[0050] See also Figure 1, a power allocation method for a redundant electric drive system, obtaining the output requirements on the axis from the upper controller or control module; wherein, the controller refers to another hardware that transmits information through communication; the control module refers to a software module that transmits information within the program; obtaining the status and various parameters of the battery and the motor from the front-end control module; determining the constraint level of each battery, and calculating the maximum output power of each battery pack; within the maximum power range, adjusting the output power of each battery pack according to the battery parameter settings to achieve global optimization; setting the power matrix p, the disturbance term p0, and the target power vector A according to the optimal power configuration of the battery pack; determining the constraint level of each motor, and setting the maximum output power of each motor; adjusting the power matrix p and the disturbance term p0 according to the set motor constraints; judging whether the power matrix p is full rank, if not, removing the constraints layer by layer from low to high until the full rank condition is met, otherwise an alarm is given to the upper controller or control module: the output requirements on some axes cannot be met, and the target needs to be adjusted; obtaining the allocation control vector r=p T (pp T ) -1 (A-p0); Determine whether the power distribution ratio is reasonable. If not, remove the constraints layer by layer from low to high until the reasonable conditions are met. Otherwise, an alarm is given to the upper-level controller or control module: the output requirements on some axes cannot be met and the target needs to be adjusted; calculate the output power of each motor based on the final power ratio.

[0051] A redundant electric drive system consisting of n batteries and m motors needs to control l drive targets. First, an n×m matrix is ​​used to describe the power transmission from the battery to the motor as shown in Table 1.

[0052]

[0053] Table 1

[0054] In Table 1, [P1,P2,…,P m ] T represents the power achieved by m motors, [p1,p2,…,p n ] T represents the power provided by n batteries, r ij , i=1,2,…,n, j=1,2,…,m, represents the power distribution ratio of battery i to motor j, and the power transfer matrix from battery to motor R=[r ij ] n×m , so R T p=P.

[0055] The motor power satisfies the constraint MP=A, where A=[A1,A2,…,A l ] T represents l driving targets, M is an l×m motor connection matrix, and the battery power constraint satisfies R[1]m =[1] n , that is, r i1 +r i2 +…+r im =1, i=1,2,…,n; the power transmission network of this redundant electric drive system can be MR T p=A to describe.

[0056] In the presence of failures and constraints in the system, some values ​​of P and p are fixed, and some values ​​of R ij will be set to zero, so our goal is to set the remaining free r in R for P, p, and R after dynamic refresh under dynamic constraints. ij Find the optimal solution to meet the driving goal A; if R n×m Each element in the matrix tries to take different values ​​in the entire search space to find the best solution, which will be a complex and computationally expensive process. Such an indirect method obviously does not meet the needs of online real-time computing and cannot be implemented in applications with high requirements for real-time controllability, such as aviation, aerospace, vehicles, ships, etc.

[0057] Therefore, we propose a new direct matrix operation method to quickly obtain the optimal solution; at the same time, this method can simplify the power conduction network of the system according to the actual electrical connection architecture, so as to utilize some sparse matrix processing methods to reduce the system complexity and the corresponding computational cost; and by utilizing partial correlation redundancy, the dimension of the power conduction network, that is, the dimension of the R matrix, can be reduced, thereby further reducing the system complexity and the corresponding computational cost.

[0058] The entire electric propulsion system includes n independent battery packs, each of which can be composed of one or more batteries connected in parallel, and m independent motors, which jointly drive l drive shafts; according to the actual electrical connection topology, the power conduction matrix can be in the following form. The form here is only used for description, not a definite instance. The power conduction matrix is ​​a general matrix of n×m as shown in Table 2; the difference between Table 2 and Table 1 is that: 1 is a truly universal form, that is, each power allocation ratio in the n×m power conduction matrix is ​​a free variable, but the actual power conduction matrix must conform to a specific electrical connection architecture, that is, some elements in the power conduction matrix are 0, that is, there is no electrical connection between the corresponding motor and the battery, and the universal form of Table 1 can be converted into the specific structural form of Table 2.

[0059]

[0060] Table 2

[0061] It can be expressed as: M =Rp B , where P M=[P1 P2 … P m ] T , , power allocation matrix R mxn is an m×n matrix of the following form:

[0062] ;

[0063] Right now .

[0064] The topological structure of the connection between the motor and the drive shaft can be expressed as: A=D lxm P M , where A=[A1 A2 … A l ] T , D l×m is an l×m matrix, of the form D l×m =[ 0 1 1 0 … 0 0; 0 0 1 1 … 0 0; … ; 1 0 0 0 … 0 1]; thus the power transfer matrix from the battery to the drive shaft can be expressed as: A=DR T p B； Rewrite the power allocation matrix R into the power ratio vector r, which contains all the power ratio values ​​r with independent degrees of freedom. ij ,by Construct a battery power matrix p and a disturbance vector p0 generated by non-independent power ratio, so that the power conduction matrix above can be rewritten as: pr = A-p0. In this way, the logical process of the previous article is followed. First, the appropriate , to construct the battery power matrix p, and then reset the battery power matrix p and the disturbance vector p0 according to the motor constraints, and then calculate the power ratio vector r=p with the power vector A on the drive shaft obtained from the upper controller or control module T (pp T ) -1 (A- p0), and adjust the constraint level according to the full rank condition and the reasonable ratio condition, so as to optimize the power ratio, that is, the optimization of the power transmission matrix R; here GB i It is a battery pack i, which can be made up of multiple batteries connected in parallel. Let us define a battery pack i as h i The batteries are connected in parallel. i Not the same, i=1,2,…,n, its power is expressed as p ij ,i=1,2,…,n,j=1,2,…,h i； According to the above driving power conduction matrix, the distribution control vector r is determined in the form of r=[r 12 r 13 r 14 r25 r 26 … r n-1m-2 r n-1m-1 r nm r n1 ], considering the constraint that the total battery allocation is 100%, the control variable r can be reduced from an m-dimensional vector to an mn-dimensional vector to simplify the processing; after obtaining the allocation control ratio r ij Then, we can get the corresponding P j , when the motor is running at a given power P j Each battery pack will also operate at a set Provide electrical energy so that the entire electric drive system operates in an optimal way while satisfying all levels of constraints, and r ij , P j , It is updated in real time as operating conditions and constraints change, thus enabling a real-time optimization process.

[0065] Figures 2 to 9 These are the specific implementation processes of step A to step H respectively.

[0066] When the battery constraint level is p 1 , or the motor constraint level is P 1 , the full rank condition and the reasonable power distribution ratio condition are still not met, indicating that the power on at least one axis cannot be met. Combined with the situation of insufficient overall battery power, that is, ∑ n j=1 p i <∑ l i=1 A i When the power on the shaft cannot be met, that is, the battery emergency mode or the motor emergency mode is set, it has exceeded the control scope of the controller at this level. The upper-level controller must adjust the load on the shaft to redistribute and optimize it under the premise of safe operation of the whole machine.

[0067] The power conducting MR in the present invention T p=A is also applicable to the process of driving the motor to generate electricity and charge the battery from the shaft. It is only necessary to convert all the discharge-related parameters into charging-related parameters, and the processed power is also converted from motor driving power and battery discharge power to motor generating power and battery charging power. The method proposed above can be extended to the power generation mode, that is, the motor works in generator mode and the battery works in charging mode, so that the electric energy generated by the generator can be allocated to charge different batteries in real time according to the power ratio while meeting the constraints and restrictions of the motor and battery.

[0068] Embodiment 1:

[0069] See also Fig.10Taking the redundant electric drive system as an example, we will illustrate the proposed method. The system has four drive shafts, eight motors, and eight batteries. The drive shafts are respectively the first drive shaft A1, the second drive shaft A2, the third drive shaft A3, and the fourth drive shaft A4; the motors are respectively the first motor M1, the second motor M2, the third motor M3, the fourth motor M4, the fifth motor M5, the sixth motor M6, the seventh motor M7, and the eighth motor M8; the batteries are respectively the first battery B1, the second battery B2, the third battery B3, the fourth battery B4, the fifth battery B5, the sixth battery B6, the seventh battery B7, and the eighth battery B8. Each drive shaft is driven by two independent motors, and each motor is powered by two independent batteries connected in parallel to a battery pack. Each battery pack powers two motors at the same time. The thick lines in the figure represent mechanical connections, and the thin lines represent electrical connections.

[0070] The system needs to meet the power requirements on the four drive shafts [A1, A2, A3, A4] T , and dynamically and optimally allocate the operating power of each motor under the power constraints of the motor and battery. The motor power is represented by P, which is [P1, P2, P3, P4, P5, P6, P7, P8] in this case. T , the battery power is represented by p, in this case [p1,p2,p3,p4,p5,p6,p7,p8] T , the power distribution ratio of battery n to motor m is expressed as r nm , n=1,2,…,8, m=1,2,…,8; the power conduction matrix from the battery to the motor in this example is shown in Table 3.

[0071]

[0072] Table 3

[0073] In Table 3, most of the nm , n=1,2,…,8, m=1,2,…,8, there is no electrical connection and can be directly set to zero; so the motor constraints in this example are:

[0074] P1+P2=A1P3+P4=A2P5+P6=A3P7+P8=A4 (1);

[0075] The constraint on battery power in this example can be written as:

[0076] r 12 ×p1+r 22 ×p2=P2r 13 ×p1+r 23 ×p2=P3r 34 ×p3+r 44 ×p4=P4r 35 ×p3+r45 ×p4=P5

[0077] r 56 ×p5+r 66 ×p6=P6r 57 ×p5+r 67 ×p6=P7r 78 ×p7+r 88 ×p8=P8r 71 ×p7+r 81 ×p8=P1 (2);

[0078] Among them, r nm satisfy:

[0079] r 12 +r 13 =1 r 22 +r 23 =1 r 34 +r 35 =1 r 44 +r 45 =1 r 56 +r 57 =1 r 66 +r 67 =1 r 78 +r 71 =1r 88 +r 81 =1 (3);

[0080] Among them, the motor battery power restriction conditions can be divided into different levels, such as level 1 is mandatory, level 2 is priority, and level 3 is completely free, which are expressed as P 1 , P 2 , P 3 , and p 1 , p 2 , p 3 , we set the priority of the motor to be higher than the battery, that is to say, when considering the restriction priority, the priority from high to low is: P 1 =p 1 >P 2 >p 2 >P 3 >p 3 ; Generally speaking, b restriction levels can be set for the motor and a restriction level can be set for the battery, and the lowest level can be set to no restriction.

[0081] The problem we need to deal with is to find the appropriate power allocation ratio r under the motor power constraints (1) and battery power constraints (2) and (3) listed above, and the constraints from the upper-level strategy. 12 ,r 13 ,r22 ,r 23 , r 34 ,r 35 ,r 44 ,r 45 ,r 56 ,r 57 ,r 66 ,r 67 ,r 78 ,r 71 ,r 88 ,r 81 .

[0082] Because the two batteries in the above example are connected in parallel, the power transfer matrix above can be simplified to Table 4.

[0083]

[0084] Table 4

[0085] In Table 4, GB1 is a No. 1 battery group, which is composed of No. 1 and No. 2 batteries in parallel, GB3 is a No. 3 battery group, which is composed of No. 3 and No. 4 batteries in parallel, GB5 is a No. 5 battery group, which is composed of No. 5 and No. 6 batteries in parallel, and GB7 is a No. 7 battery group, which is composed of No. 7 and No. 8 batteries in parallel. In order to simplify the description, we define , j = 1, 2, …, 4, , , , .

[0086] According to the simplified driving power conduction matrix, the distribution control vector r=[r 12 r 13 r 34 r 35 r 56 r 57 r 78 r 71 ], the original 16 control variables are reduced to 8, but in this case, because of the battery constraint (3), the control variables can be further reduced to 4 to simplify the processing, so the allocation control vector can be rewritten as r = [r 12 r 34 r 56 r 78 1], where 1 is a constant used to express r 13 =1-r 12 The constraint, that is, r 13 are not free control variables, so the true allocation control vector contains only 4 free control variables and one constant generated by the constraint.

[0087] According to different failure and constraint situations, the above problem is rewritten to suit the new method of describing direct matrix operations; we define a battery power matrix p, which is set by the current failure tolerance and optimization constraint strategy.

[0088] The drive target is obtained from the upper control module. In this case, it is the power requirement on the four drive axes, defined as a 4-dimensional vector A=[A1, A2, A3, A4] T； Obtain battery parameters from the battery management module, such as state of charge SOC, maximum output power p max_d , maximum input power p max_c , discharge impedance, charging impedance, maximum discharge current I max_d , Maximum charging current I max_c etc., as well as the battery's current state of health SoH. The battery's current limitations are defined by battery parameters such as discharge impedance, SOC, and SoH. For example, if battery pack 1 fails, p1=0, or battery pack 1 needs to limit its power to p 1_lim , then p1 must be smaller than p 1_lim , and set the limit level, and then according to the above battery parameters, the battery pack power p can be freely selected within the allowed space 12 , p 34 , p 56 , p 78 The value of 12 +p 34 +p 56 +p 78 = A1+A2+A3+A 4； To simplify the following description, define , j = 1, 2, ..., 4, theoretically there can be an infinite number of combinations can be selected, but the actual implementation requires a certain logical process to generate a reasonable set of .

[0089] According to the restriction condition p obtained from the upper control program 1 , p 2 ,…,p k , calculate the power limit p of each battery i_lim ,i=1,2,…,8; In this example, we take k=4 as an example to illustrate a process of implementing the algorithm logic, namely p 1 , p 2 , p 3 , p 4 , where p i 1 =0, i=1,2,…,8,p i 4 =p i_max,i=1,2,…,8, that is, no constraints, p i_max The physical upper limit is obtained by the front-end control program; this example can be directly extended to the case where k>4.

[0090] The specific implementation steps in this embodiment are:

[0091] Step a): Get 4 drive targets A from the upper controller or control module i ,i=1,2,…,4;The maximum power p of each battery at each limit level is obtained from the front-end control program i k , i=1,2,…,8, k=1,2,3,4, where p i 1 =0, p i 3 = p i_max , i=1,2,…,8,p i_max is the maximum power of battery i; label the battery constraint level as p 3 ; Obtain the maximum power P of each motor at each limit level from the front-end control program i k , i=1,2,…,8, k=1,2,3,4; mark the motor constraint level as P 3 ; Get the internal resistance R of the battery from the front-end control program i ,i=1,2,…,8;Get the state of charge SOC of the battery pack from the front-end control program j ,j=1,2,…,4,battery health status SoH i , i=1,2,…,8;

[0092] Step b1): p i_lim =p i_max , i=1,2,…,8;

[0093] Step b2): determine whether battery i is invalid, i=1, 2, ..., 8; if yes, proceed to step b3); if no, continue to process the next battery until all batteries are completed and proceed to step b4);

[0094] Step b3): p i_lim =p i 1 , and mark battery i as failed, and report to the superior controller; then return to step b2);

[0095] Step b4): Determine ∑ 8 i=1 p i_lim <∑ 4 i=1 A iIs it established? If yes, go to step b5); if not, go to step b6);

[0096] Step b5): Alarm the superior controller that the power on the axis exceeds the upper limit of the battery, adjust the output power on the axis; mark the battery constraint level as p 1 , and then enter step c1) after setting the battery emergency mode;

[0097] Step b6): Determine whether the constraint level is p 1 ; If yes, proceed to step c1); If no, proceed to step b7);

[0098] Step b7): Determine whether battery i is not expired and is restricted, i=1, 2, ..., 8; if yes, proceed to step b8); if not, continue to process the next battery until all batteries are completed and proceed to step b9);

[0099] Step b8): p i_lim_tem =p i_lim , p i_lim =p i 2 ; Then return to step b7);

[0100] Step b9): Determine ∑ 8 i=1 p i_lim <∑ 4 i=1 A i Is it established? If yes, go to step b10); if not, go to step b14);

[0101] Step b10): Alarm the superior controller, the alarm content is that the power demand on the shaft is too large, the output power on the shaft is adjusted, and the battery damage alarm; then enter step b11);

[0102] Step b11): determine whether battery i is not expired and is restricted, i=1, 2, ..., 8; if yes, proceed to step b12); if not, continue to process the next battery until all batteries are completed and return to step b13);

[0103] Step b12): p i_lim =p i_lim_tem ; Then return to step b11);

[0104] Step b13): Mark the battery constraint level as p 2 ; Then proceed to step c1);

[0105] Step b14): Determine whether the constraint level is p 2; If yes, proceed to step c1); If no, proceed to step b15);

[0106] Step b15): determine whether battery i is not failed, not restricted, and optimized, i=1, 2, ..., 8; if yes, proceed to step b16); if not, continue to process the next battery until all batteries are completed and proceed to step b17);

[0107] Step b16): p i_lim_tem =p i_lim , p i_lim =p i 3 ; Then return to step b15);

[0108] Step b17): Determine ∑ 8 i=1 p i_lim <∑ 4 i=1 A i Is it established? If yes, go to step b18); if not, go to step b22);

[0109] Step b18): Report to the superior controller, the report content is to optimize the power demand on the shaft, adjust the output power on the shaft, and alarm for battery damage; proceed to step b19);

[0110] Step b19): Determine whether battery i is not failed, not restricted, and optimized, i=1, 2, ..., 8; if yes, proceed to step b20); if not, continue to process the next battery until all batteries are completed and proceed to step b21);

[0111] Step b20): p i_lim =p i_lim_tem ; Then return to step b19);

[0112] Step b21): Mark the battery constraint level as p 3 Then proceed to step c1);

[0113] Step b22): Determine whether the battery constraint level is p 3 If yes, go to step c1); if no, go to step b23);

[0114] Step b23): determine whether battery i is not failed, not restricted, and optimized, i=1, 2, ..., 8; if yes, proceed to step b24); if not, continue to process the next battery until all batteries are completed and proceed to step b25);

[0115] Step b24): p i_lim_tem =p i_lim, p i_lim =p i 3 ; Then return to step b23);

[0116] Step b25): Determine ∑ 8 i=1 p i_lim <∑ 4 i=1 A i Is it established? If yes, go to step b26); if not, go to step c1);

[0117] Step b26): Report to the superior controller, the report content is: optimize the power demand on the shaft and adjust the output power on the shaft; then enter step b27);

[0118] Step b27): Determine whether battery i is not failed, not restricted, and optimized, i=1, 2, ..., 8; if yes, proceed to step b28); if not, continue to process the next battery until all batteries are completed and proceed to step b29);

[0119] Step b28): p i_lim =p i_lim_tem ; Then return to step b27);

[0120] Step b29): Mark the constraint level as p 3 ; Then proceed to step c1);

[0121] Step c1): determine whether the battery 2j-1 in the battery pack j is marked as failed; if yes, proceed to step c2); if not, proceed to step c3);

[0122] Step c2): determine whether the battery 2j in the battery pack j is marked as failed; if so, , then go to step c3); if not, then ; Then proceed to step c3);

[0123] Step c3): Determine whether the battery 2j in the battery pack j is marked as failed; if so, , then go to step c4); if not, then , j=1,2,…,4; then go to step c4);

[0124] Step c4): Determine whether it is battery emergency mode; if so, , proceed to step g1); if not, proceed to step d1);

[0125] Step d1): Calculate the average battery power p avg :p avg =∑4 i=1 A i / 4; SOC of battery pack j j , j = 1, 2, ..., 4, calculate the mean SOC avg :∑ 4 j=1 SOC j / 4; the SoH of the worst battery in battery pack j i , i=1,2,…,8, marking the SOH of battery pack j j = min(SoH 2j-1 , SoH 2j ), j=1,2,…,4; calculate the mean SOH SOH avg : SOH avg =∑ 4 j=1 SOH j / 4;

[0126] Step d2): Determine |SOC j -SOC avg |Is it greater than a set threshold? If yes, proceed to step d3); if no, proceed to step d4);

[0127] Step d3): , ; Then proceed to step d5);

[0128] Step d4): Determine |SOH j -SOH avg |Is it greater than the set threshold? If so, then , ; if not, then dp j =0, ; Then proceed to step d5);

[0129] Step d5): Calculate the total power adjustment value dp sum =∑ 4 j=1 dp j , then determine dp sum =0 is true, if true, then go to step e1), if not true, then go to step d6);

[0130] Step d6): Determine dp sum >0 is true; if true, proceed to step d7); if not, proceed to step d8);

[0131] Step d7): Calculate the adjustment deviation: dp j_mrgn= 0 is true, if true, then adjust the bias: bias j =0; if not true, calculate the adjustment bias: bias j =min[max 4 k=1 (SOC k )+dSOC ad -SOC j ,dp j_mrgn / dp sum ]; Calculate adjustment ratio: rate j_ad = bias j / ∑ 4 j=1 bias j ;dp j =dp j + rate j_ad ×dp sum ; Then proceed to step d9);

[0132] Step d8): Calculate the adjustment bias: bias j =min[SOC j -min 4 k=1 (SOC k )-dSOC ad , 0]; Calculate adjustment ratio: rate j_ad = bias j / ∑ 4 j=1 bias j ;dp j =dp j + rate j_ad ×dp sum ; Then proceed to step d9);

[0133] Step d9): Add the adjustment value to ; Then proceed to step e1);

[0134] Step e1): P i_lim =P i_max , i=1,2,…,8;

[0135] Step e2): determine whether motor i is failed, i=1, 2, ..., 8; if yes, proceed to step e3); if no, proceed to step e4);

[0136] Step e3): P i_lim =P i 1 , and mark motor i as constrained, report motor i failure to the superior controller, and go to step e4);

[0137] Step e4): Determine ∑ 8 i=1 P i_lim <∑ 4 i=1 A i Is it true? If so, an alarm is sent to the upper controller: the shaft power exceeds the upper limit of the motor, and the shaft output power P is adjusted. i =P i_lim ; Mark the constraint level as P 1 ; Set the motor emergency mode and the process ends and returns; if not, go to step e5);

[0138] Step e5): Determine whether the motor constraint level is P 1 , if yes, proceed to step f1); if no, proceed to step e6);

[0139] Step e6): Determine whether motor i is not failed and is limited, i=1,2,…,8; if yes, then P i_lim_tem =P i_lim , P i_lim =P i 2 , and mark motor i as constrained, and go to step e7); if not, go to step e7);

[0140] Step e7): Determine ∑ 8 i=1 P i_lim <∑ 4 i=1 A i Is it true? If yes, go to step e8); if no, go to step e9);

[0141] Step e8): Alarm the superior controller: the power demand on the shaft is too large, adjust the output power on the shaft, and alarm the motor damage; and determine whether the motor i is not failed and is limited, i=1,2,…,8; if so, then P i_lim =P i_lim_tem ; Mark the motor constraint level as P 2 Then, go to step f1); if not, mark the motor constraint level as P 2 Then, go to step f1);

[0142] Step e9): Determine whether the constraint level is P 2 ; If yes, proceed to step f1); If no, proceed to step e10);

[0143] Step e10): Determine whether motor i is not failed, not restricted, and optimized, i=1,2,…,8; if yes, then Pi_lim_tem =P i_lim , P i_lim =P i 3 , and mark motor i as constrained and then go to step e11); if not, go directly to step e11);

[0144] Step e11): Determine ∑ 8 i=1 P i_lim <∑ 4 i=1 A i Is it true? If so, go to step e12), if not, go directly to step f1);

[0145] Step e12): Report to the superior controller: optimize the power demand on the shaft, adjust the output power value on the shaft, and the motor efficiency loss value; and determine whether the motor i is not failed, not restricted, and is optimized, i=1,2,…,8; if so, then P i_lim =P i_lim_tem ; Mark the motor constraint level as P 3 Then, go to step f1); if not, mark the motor constraint level as P 3 Then, go to step f1);

[0146] Step f1): determine whether motor i is constrained; if yes, proceed to step f2); if no, proceed to the next motor until all motors are determined and proceed to step f5);

[0147] Step f2): Determine P i_lim ≥P i 4 Is it true? If true, then P i =P i 4 , then go to step f3); if not, then P i =P i_lim , then proceed to step f3);

[0148] Step f3): According to the P of the i-th motor i Set p and p0;

[0149] Step f4): count the restricted motors count = count + 1; then continue with the next motor until all motors are judged and enter step f5);

[0150] Step f5): Determine whether count=0; if so, P1=P1 of the first motor 4Set p and p0, and mark motor 1 as constrained; then go to step f6); if not, go directly to step f6);

[0151] Step f6): Determine whether rank (p) is equal to 4; if yes, proceed to step f8); if no, proceed to step f7);

[0152] Step f7): Determine whether the motor constraint level is P 1 ; If yes, then alarm the superior controller: the power demand on the shaft is too large, adjust the output power on the shaft, set the motor emergency mode, and then end; if no, then check the current motor constraint level, set the current motor constraint level P k Promoted to P k-1 , then return to step e1);

[0153] Step f8): Through the matrix p and the vector p0, the power requirement vector A on the axis is obtained from the upper controller, and the distribution control vector r = p is solved by solving the equation p×r=A-p0. -1 (A-p0); then proceed to step g1);

[0154] Step g1): Determination , j=1,2,…,4, is it between [0,1]; if yes, proceed to step h); if no, proceed to step g2);

[0155] Step g2): Determine whether the battery constraint level is p 1 ; If yes, then alarm the superior controller: the power demand on the shaft is too large, adjust the output power on the shaft, set the battery emergency mode, and then end; if no, then check the current battery constraint level, set the current battery constraint level p k Raised to p k-1 , then return to step b1);

[0156] Step h): Determine whether motor i is constrained; if not, then ,in , , then end; if yes, then end the process and return directly.

[0157] Embodiment 2:

[0158] See also Fig.11 The system has six independent drive shafts, each of which is driven by two independent motors. Each motor is powered by two independent batteries connected in parallel to a battery pack. Each battery pack powers three motors at the same time. The first battery pack (batteries B1 and B2) powers motors M2, M4, and M6; the second battery pack (batteries B3 and B4) powers motors M3, M5, and M7; the third battery pack (batteries B5 and B6) powers motors M6, M7.A 、M C Power supply; the fourth battery group (batteries B7, B8) is for motors M9, M B , M1 supplies power; after simplifying the parallel batteries, the power conduction matrix from the battery pack to the motor is shown in Table 5.

[0159]

[0160] Table 5

[0161] The power transmission matrix from the battery pack to the motor is listed in a similar manner to that in the first embodiment and simplified to obtain the motor power constraint in this example:

[0162] P1+P2=A1P3+P4=A 2 5 +P6=A3P7+P8=A4P9+P A =A5P B +P C =A6(4);

[0163] The battery power constraint can be written as:

[0164] (5);

[0165] Among them, r nm satisfy:

[0166] r 12 +r 14 +r 16 =1 r 33 +r 35 +r 37 =1 r 58 +r 5A +r 5C =1 r 79 +r 7B +r 71 =1 (6);

[0167] The motor power constraint and battery power constraint can be divided into different levels, such as level 1 is mandatory, level 2 is priority, and level 3 is completely free, which are expressed as P 1 , P 2 , P 3 , and p 1 , p 2 , p 3 , we can set the priority of the motor to be higher than the battery, that is to say, when considering the restriction priority, the priority from high to low is: P 1 = p 1 > P 2 > p 2 >P3 > p 3 ; Similarly, the priority of the battery can be set higher than that of the motor, which does not hinder the implementation of the present method and should be regarded as a direct promotion of the present invention; generally, b restriction levels can be set for the motor and a restriction level can be set for the battery, and the lowest level can be set to an unconstrained situation.

[0168] Because of the battery constraint (6), the number of independent control variables is reduced to 8, namely r 12 , r 14 , r 33 , r 35 , r 58 ,r 5A , r 79 , r 7B The problem we need to deal with is to find the appropriate power allocation ratio r under the motor power constraints (4) and battery power constraints (5) and (6) listed above, as well as the constraints from the upper-level strategy. 12 , r 14 , r 33 , r 35 ,r 58 , r 5A , r 79 , r 7B .

[0169] The power requirements on the six drive axes are obtained from the upper control module as the drive target, which is defined as a 6-dimensional vector A=[A1, A2, A3, A4, A5, A6] T , determine the allocation control vector r=[ r 12 r 14 r 33 r 35 r 58 r 5A r 79 r 7B ], and then according to different failure tolerance and optimization restriction strategies, determine the battery power matrix p and the disturbance vector p0 generated by constraint (3), so that p×r=A-p0, since p here is a 6×8 matrix, p0 is a 6-dimensional vector, such as:

[0170] ;

[0171] So r=p T (pp T ) -1 (A-p0).

[0172] Obtain battery parameters from the battery management module, such as state of charge SOC, maximum output power p max_d, maximum input power p max_c , discharge impedance, charging impedance, maximum discharge current I max_d , Maximum charging current I max_c etc., as well as the battery's current state of health SoH. The battery's current limitations are defined by battery parameters such as discharge impedance, SOC, and SoH. For example, if battery No. 1 fails, p1=0, or battery No. 1 needs to limit its power to p 1_lim , then p1 must be smaller than p 1_lim , and set the restriction level, and then according to the above battery parameters, you can freely choose the battery power within the allowed space , , , As long as the value of Specific implementation steps refer to steps a) to h) in the content of the invention, select according to the design process , , , Then, based on the constraint level, full rank and rationality conditions of each motor, the power matrix p and the disturbance term p0 are determined to obtain the distribution control vector r=p T (pp T ) -1 (A-p0), and thereby calculate the output power of each motor; such logic can be implemented by different processes, and the present invention is not only aimed at the process a) to h), but also includes protection for different processes but implementing the same logic.

[0173] Embodiment three:

[0174] See also Fig.12 The system has six independent drive shafts, each of which is driven by two independent motors. Each motor is powered by two independent batteries connected in parallel to a battery pack. Each battery pack powers three motors at the same time. The first battery pack (batteries B1 and B2) powers motors M2, M6, and M7. A Power supply; the second battery group (batteries B3, B4) is for motors M3, M7, M B Power supply; the third battery group (batteries B5, B6) is for motors M4, M8, M C Power supply; the fourth battery group (batteries B7, B8) supplies power to motors M5, M9, and M1; after simplifying the parallel batteries, the power conduction matrix from the battery group to the motor is shown in Table 6.

[0175]

[0176] Table 6

[0177] In this example, the power transmission matrix from the battery pack to the motor is simplified by using a method similar to that of Embodiments 1 and 2, and the motor power constraint is obtained as follows:

[0178] P1+P2=A1P3+P4=A2P5+P6=A3 P7+P8=A4P9+P A =A5P B +P C =A6(7);

[0179] In this example, the battery power constraint can be written as:

[0180] (8);

[0181] Among them, r nm satisfy:

[0182] r 12 +r 16 +r 1A =1 r 33 +r 37 +r 3B =1 r 54 +r 58 +r 5C =1 r 75 +r 79 +r 71 =1 (9);

[0183] The motor and battery power restrictions can be divided into different levels, such as level 1 for mandatory, level 2 for priority, and level 3 for complete freedom, expressed as P 1 , P 2 , P 3 , and p 1 , p 2 , p 3 , we can set the priority of the motor to be higher than the battery, that is to say, when considering the restriction priority, the priority from high to low is: P 1 = p 1 > P 2 > p 2 > P 3 > p 3 ; Similarly, the priority of the battery can be set higher than that of the motor, which does not hinder the implementation of the present method and should be regarded as a direct promotion of the present invention; generally, b restriction levels can be set for the motor and a restriction level can be set for the battery, and the lowest level can be set to an unconstrained situation.

[0184] Because of the battery power constraint (9), the number of independent control variables is reduced to 8, namely r 12 , r16 , r 33 , r 37 ,r 54 , r 58 , r 75 , r 79 The problem we need to deal with is to find the appropriate power allocation ratio r under the motor power constraints (7) and battery power constraints (8) and (9) listed above, as well as the constraints from the upper-level strategy. 12 , r 16 , r 33 , r 37 ,r 54 , r 58 , r 75 , r 79 .

[0185] The power requirements on the six drive axes are obtained from the upper control module as the drive target, which is defined as a 6-dimensional vector A=[A1, A2, A3, A4, A5, A6] T , determine the allocation control vector r=[ r 12 r 16 r 33 r 37 r 54 r 58 r 75 r 79 ], and then according to different failure tolerance and optimization restriction strategies, determine the battery power matrix p and the disturbance vector p0 generated by constraint (3), so that p×r=A-p0, since p here is a 6×8 matrix, p0 is a 6-dimensional vector, such as:

[0186] ;

[0187] So r=p T (pp T ) -1 (A-p0).

[0188] Obtain battery parameters from the battery management module, such as state of charge SOC, maximum output power p max_d , maximum input power p max_c , discharge impedance, charging impedance, maximum discharge current I max_d , Maximum charging current I max_c etc., as well as the battery's current state of health SoH. The battery's current limitations are defined by battery parameters such as discharge impedance, SOC, and SoH. For example, if battery No. 1 fails, p1=0, or battery No. 1 needs to limit its power to p 1_lim , then p1 must be smaller than p 1_lim, and set the restriction level, and then according to the above battery parameters, you can freely choose the battery power within the allowed space , , , As long as the value of Specific implementation steps refer to steps a) to h) in the content of the invention, select according to the design process , , , Then, based on the constraint level, full rank and rationality conditions of each motor, the power matrix p and the disturbance term p0 are determined to obtain the distribution control vector r=p T (pp T ) -1 (A-p0), and thereby calculate the output power of each motor; such logic can be implemented by different processes, and the present invention is not only aimed at the process a) to h), but also includes protection for different processes but implementing the same logic.

[0189] It should be particularly noted that the above embodiments are for illustration of the present invention, not for limitation of the present invention, and any solution obtained by simply changing the present invention belongs to the protection scope of the present invention.

Claims

1. A redundant electric drive system power distribution method, characterized in that: The following steps are involved: a) Obtaining axis output requirements: Obtaining axis output requirements from the upper controller or control module; b) Obtaining the state parameters of each battery and motor: Obtaining the state and various parameters of the battery and motor from the front-end control module; c) Calculate the maximum output power of each battery group: determine the constraint level of each battery and the combination of each battery group, and calculate the maximum output power of each battery group; when calculating the maximum output power of each battery group, adjust the output power of each battery group according to the battery parameter settings within the maximum output power range to achieve the global optimum; according to the optimal power configuration of the battery group, set the power matrix p, the disturbance term p0, and the target power vector A; d) Setting the maximum output power of each motor: Determine the constraint level of each motor, set the maximum output power of each motor; determine the output power of the constrained motor; then set the power matrix p and the disturbance term p0 according to the constrained motor; e) Determine whether the motor meets the demand: Determine whether the current motor can meet the on-axis output demand under the determined constraints. If not, remove the motor constraints layer by layer from low to high until the on-axis output requirements are met, otherwise report to the upper-level controller or control module: some on-axis output requirements cannot be met and the target needs to be adjusted; Determine whether the current motor meets the on-axis output demand under the determined constraints by determining whether the power matrix p is full rank. If it is not full rank, it indicates that the motor constraints are too many and the on-axis output target cannot be achieved. Then remove the motor constraints layer by layer from low to high until the full rank condition is met; f) Determine whether the power allocation ratio is reasonable: Calculate the power allocation ratio and determine whether the power allocation ratio is reasonable. If it is unreasonable, remove the battery constraints layer by layer from low to high until the reasonable conditions are met. Otherwise, report to the upper controller or control module: the output requirements on some axes cannot be met and the target needs to be adjusted; calculate the power allocation ratio and obtain the allocation control vector r=p through matrix calculation. T (pp T ) -1 (A-p0); is the power allocation ratio, j = 1, 2, …, mn; g) Calculate motor output power: Calculate the output power of each motor based on the power distribution ratio.

2. A redundant electric drive system power distribution method as claimed in claim 1, characterized in that: In the step f), whether the power allocation ratio is reasonable is determined by judging whether the calculated power allocation ratio is within a reasonable range. If it is not within this range, it indicates that there are too many battery constraints. In this case, the battery constraints are removed layer by layer from low to high until all power allocation ratios are within the selected range.

3. A redundant electric drive system power distribution method as claimed in claim 1, characterized in that: The system involved in the method includes n battery packs, and battery pack i is composed of h i Batteries are connected in parallel, i = 1, 2, ..., n; m motors; control l driving targets, the specific steps are: A) Obtain a drive target A from the upper controller or control module i , i=1,2,…,l; obtain the status and parameters of each battery and motor from the front-end control program; initialize the program and assign initial values ​​to each control variable; B) Determine the constraint level p of each battery according to its status k , k = 1, 2, …, a; the smaller k is, the higher the constraint level is, that is, the higher the priority is; at the same time, under the premise of meeting the power demand on the axis, the constraint level is adjusted from high to low to the optimal level, and the maximum power that each battery can provide is determined; C) determining the maximum output power of each battery pack according to the constraint level of each battery and the combination of each battery pack; D) According to the health status parameter SoH of each battery obtained from the front-end control program ij , i=1,2,…,n, j=1,2,…,h i , and the battery pack's state of charge SOC i , i = 1, 2, ..., n, adjust and optimize the output power of the battery pack to meet the power demand on the axle within the maximum output power of each battery pack while balancing the health status and charge state of each battery pack; E) Determine the constraint level P of each motor according to its status k , k=1,2,…,b; the smaller k is, the higher the constraint level is, that is, the higher the priority is; at the same time, under the premise of meeting the power demand on the shaft, the constraint level is adjusted from high to low to the optimal level, and the maximum power that each motor can provide is determined; F) determining the output power of the constrained motor; then setting the power matrix p and the disturbance term p0 according to the constrained motor; p is an l×(mn) matrix, p0 is an l-dimensional term vector; and the full rank condition of the power matrix p is used to determine whether the motor constraints can meet the power requirements of each axis. If p is not full rank, it means that there are too many motor constraints, and the lowest motor constraint condition is abandoned layer by layer, that is, higher-level motor constraints are retained as much as possible until a full-rank p is obtained, and the allocation control vector r=p is calculated based on this. T (pp T ) -1 (A-p0), G) Determine the power allocation ratio j=1,2,…,mn, is it in a reasonable range, that is, between [0,1]? If there is a power allocation ratio that is not in the reasonable range, it means that there are too many battery constraints. Then, the lowest battery constraint condition is abandoned layer by layer, that is, higher-level battery constraints are retained as much as possible until all calculated power allocation ratios are in a reasonable range; H) Based on the power transfer matrix R from battery to motor mxn Calculate m r ji The value of r ji represents the power allocation ratio of battery j to motor i, i∈{1,2,…,m}, j∈{1,2,…,n}, and r ji The output power of all unconstrained motors is calculated and combined with the output power of the constrained motors obtained in F), and then transmitted to the lower-level controller in the form of commands to control each motor to reach the required power, thereby meeting the power requirements on all axes while achieving an optimized balance between the battery and the motor.

4. A redundant electric drive system power distribution method as claimed in claim 3, characterized in that: Battery constraint level p k , k = 1, 2, ..., a, is a level, divided into: Level 1 is failure, the highest level, that is, physical hardware restrictions; Level 2 is restriction, that is, mandatory in non-emergency situations; Level 3 to level a-1 are different levels of optimization; Level a is completely free; According to the priority from high to low: p 1 >p 2 >p 3 >…>p a-1 >p a .

5. A redundant electric drive system power distribution method as claimed in claim 3, characterized in that: Motor constraint level P k , k = 1, 2, ..., b, is b levels, divided into: Level 1 is failure, the highest level, that is, physical hardware restrictions; Level 2 is restriction, that is, mandatory in non-emergency situations; Level 3 to level b-1 are different levels of optimization; Level b is completely free; According to the priority from high to low: P 1 >P 2 >P 3 >…>P b-1 >P b .

6. A redundant electric drive system power distribution method as claimed in claim 3, characterized in that: The dimensions of the power matrix p and the allocation control vector r are reduced from m to mn, which reduces the amount of calculation.

7. A redundant electric drive system power distribution method according to any one of claims 3 to 6, characterized in that: The power matrix p is composed of 0 and the optimized power of each battery pack, and the disturbance term p0 is composed of a combination of 0 and the optimized power of each battery pack and the power of the constrained motor.

Images