Torque distribution control method of multilayer axial flux motor

Abstract

The invention relates to the technical field of axial magnetic flux motors of electric automobiles, and particularly discloses a torque distribution control method of a multilayer axial magnetic flux motor. The method comprises the steps that parameters are set, N motor modules are arranged, each motor module is provided with a preferred torque and a rated torque, the motor module marked with the minimum rated torque is m1, the motor module marked with the second minimum rated torque is m2, and the motor module marked with the maximum rated torque is mN, the maximum torque instruction value is set to be the sum of rated torques of all the motor modules; a torque instruction partition, wherein N + 1 partitions are divided from 0 to the maximum torque instruction value; torque instruction distribution: acquiring a torque instruction, distributing a torque instruction value to motor modules m1 to mn in a partition n according to the torque instruction value, and starting the motor modules m1 to mn; and according to the torque command value in the partition N + 1, distributing the torque command value to all the motor modules m1 to mN and starting the motor modules m1 to mN. The optimal efficiency interval of the motor can be expanded, expansion is easy, reliability is improved, and the requirement for computing resources is low.

Description

Technical Field

[0001] The invention relates to the technical field of axial flux motors for electric vehicles, and in particular to a torque distribution control method for multi-layer axial flux motors. Background Art

[0002] Electric vehicles (EVs) have become a major trend in the automotive industry, driven primarily by the need for environmentally friendly transportation solutions and higher power and range. One of the most critical components in an EV is the motor drivetrain, which directly affects the performance and efficiency of the vehicle. Conventional radial flux motors have reached their performance limits, prompting the industry to explore axial flux motors (AFMs) as a viable alternative.

[0003] Despite the potential of axial flux motors, most AFMs currently available on the market use a single-layer design containing only one set of stator windings. This design limits their power efficiency and scalability. To address these limitations, multi-layer AFM designs (containing multiple sets of stator windings) are a promising direction. Multi-layer AFMs can be conceptualized as multiple motor modules combined together, effectively connecting multiple motors in series on the same shaft. This design is only feasible within the inherently compact design of AFMs and is in stark contrast to existing designs that typically couple multiple motors through a gearbox.

[0004] From a control perspective, a multi-layer AFM can be viewed as a series of motor modules working together to produce a combined torque output at the shaft.

[0005] The existing electric vehicle torque distribution strategy has the following limitations:

[0006] It mainly focuses on the two-motor system coupled through a gearbox, which is not applicable to the structure of the multilayer AFM mentioned above;

[0007] Relying on brute force search methods to optimize torque distribution, this method is difficult to run in real time under the limited computing resources of the controller. If the pre-calculated distribution results are input into the controller by table lookup, the entire lookup table must be re-entered every time the motor module is replaced, wasting time and storage space in the controller;

[0008] Failure to consider smooth transition will cause torque output disturbance when the command changes suddenly;

[0009] If the torque command is only distributed evenly to each motor module, the motor will not be able to achieve its best energy efficiency. Summary of the invention

[0010] One technical problem to be solved by the present invention is to propose a rule-based torque distribution method applied to a multi-layer axial flux motor, which can not only solve the disadvantages of the prior art, but also maintain the advantage of the motor operating at the best efficiency at the same time.

[0011] To solve the above technical problem, a technical solution adopted by the present invention is: a torque distribution control strategy method for a multi-layer axial flux motor, which includes the following steps:

[0012] Parameter setting: The multi-layer axial flux motor is provided with N motor modules, and each motor module is provided with a preferred torque and a rated torque. The motor module marked with the minimum rated torque is m1, the motor module marked with the second smallest rated torque is m2, and so on until the motor module marked with the largest rated torque is mN. Set the maximum torque command value as the sum of the rated torques of all motor modules;

[0013] Torque command partitioning: Divide N + 1 partitions from 0 to the maximum torque command value, where: the lower boundary of partition 1 is 0 and the upper boundary is the preferred torque of m1, the lower boundary of partition 2 is the preferred torque of m1 and the upper boundary is the sum of the preferred torques of m1 and m2, and so on until the lower boundary of partition N is the sum of the preferred torques of m1, m2 to mN-1 and the upper boundary is the sum of the preferred torques of m1, m2 to mN, and the lower boundary of partition N + 1 is the sum of the preferred torques of m1, m2 to mN and the upper boundary is the maximum torque command value;

[0014] Torque command distribution: Obtain the torque command. According to the torque command value in partition n, where n is any number from 1 to N. When n is from 1 to N-1, distribute the torque command value to motor modules m1 to mn, start motor modules m1 to mn, and turn off other motor modules except m1 to mn. When n is N, distribute the torque command value to all motor modules m1 to mN and start all motor modules m1 to mN; According to the torque command value in partition N + 1, distribute the torque command value to all motor modules m1 to mN and start all motor modules m1 to mN.

[0015] Preferably, when the torque command value is within partition 1, it means that the torque command value is greater than the lower boundary of partition 1 and less than or equal to the upper boundary of partition 1. At this time, the torque command value is assigned to m1, and m1 is started, while other motor modules except m1 are turned off. When the torque command value is within partition 2, it means that the torque command value is greater than the lower boundary of partition 2 and less than or equal to the upper boundary of partition 2. At this time, the torque command value is assigned to m1 and m2, and m1 and m2 are started, while other motor modules except m1 and m2 are turned off. When the torque command value is within partition n, where n ranges from 1 to N - 1, it means that the torque command value is greater than the lower boundary of partition n and less than or equal to the upper boundary of partition n. The torque command value is assigned to motor modules m1 to mn, and motor modules m1 to mn are started, while other motor modules except m1 to mn are turned off. When n is N, the torque command value is assigned to all motor modules m1 to mN, and all motor modules m1 to mN are started. When the torque command value is within partition N + 1, it means that the torque command value is greater than the lower boundary of partition N + 1 and less than or equal to the upper boundary of partition N + 1. At this time, the torque command value is assigned to all motor modules m1 to mN, and all motor modules m1 to mN are started.

[0016] Preferably, in the torque command distribution step: when the torque command value is within partitions 1 to N, the torque command value is evenly distributed to each started motor module. If the distributed torque command value is greater than or equal to the preferred torque of a certain motor module, the torque command value assigned to this motor module is limited to the preferred torque of this motor module, and the remaining undistributed torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all torque command value balances are distributed to the motor modules. When the torque command value is within partition N + 1, the torque command value is evenly distributed to N motor modules. If the distributed torque command value is greater than or equal to the rated torque of a certain motor module, the torque command value assigned to this motor module is limited to the rated torque of this motor module, and the remaining undistributed torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all torque command value balances are distributed to the motor modules.

[0017] Preferably, in the torque command distribution step, when the torque command value exceeds the upper boundary of partition N + 1, that is, when the torque command value is greater than the maximum torque command value, the rated torque of each motor module in the N motor modules is assigned to each motor module.

[0018] Preferably, in the torque command distribution step, after the controller receives two commands T cmd,t and T cmd,t+1 , it calculates the difference ΔT cmd = T cmd,t+1 - T cmd,t , and uses the interpolated torque command T inter.cmd,a each time the torque command distribution is executed:

[0019]

[0020] A = f inter.cmd / f cmd update ,

[0021] where t represents the time series, T cmd,t represents the instruction received by the controller at time t, T cmd,t+1 represents the instruction received by the controller at time t + 1, f inter.cmd is the frequency allocated by the controller to execute the torque instruction T inter.cmd,a f cmd update is the frequency at which the controller receives the instruction T cmd,t a increases from 1 to A - 1 to generate A - 1 interpolated torque instructions T inter.cmd,a to correspond to A - 1 interpolation periods between t and t + 1.

[0022] Preferably, in the torque instruction partitioning step, N + 1 partitions are divided from 0 to the maximum torque instruction value, and each partition is bounded by the torque instruction boundary T cmd,sec n For the lower boundary of partition n is T cmd,sec n-1 , and the upper boundary of partition n is T cmd,sec n When n is from 2 to N, a hysteresis zone T buffer is added before and after the partition boundary. The lower boundary of partition n is bounded by T cmd,sec n-1 + T buffer and T cmd,sec n-1 - T buffer When the torque instruction T cmd exceeds T cmd,sec n-1 + T buffer the motor module mn is started after adding the hysteresis zone. When the torque instruction T cmd is lower than T cmd,sec n-1 - T buffer the motor module mn is turned off after reducing the hysteresis zone.

[0023] The present invention has the following beneficial effects:

[0024] 1. The method of the present invention can expand the optimal efficiency range of the motor. Compared with blindly averaging the torque instructions to each motor module, the efficiency is higher when the motor operates in the low torque region using the method of the present invention.

[0025] 2. The hysteresis loop and torque command interpolation process of the method of the present invention ensure smooth torque transition during "gear shifting".

[0026] 3. The method of the present invention makes it easier to expand when the motor has more modules, and is also easier to implement and adjust.

[0027] 4. The method of the present invention considers the countermeasures for motor module failure and improves its redundancy.

[0028] 5. The method of the present invention has low requirements for computing resources. The amount of calculation required during execution is small, and most parameters are determined before operation. This is beneficial for implementation on low-end devices and high-frequency operation control loops. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of the present invention.

[0030] Figure 2 It is a schematic diagram of the structure of a multi-layer axial flux motor used in the present invention.

[0031] Figure 3 It is a torque command partition diagram of the present invention.

[0032] Figure 4 It is a schematic diagram of preferred torque selection of the present invention.

[0033] Figure 5 It is the torque command distribution process of the present invention Figure 1 。

[0034] Figure 6 It is the torque command distribution process of the present invention Figure 2 。

[0035] Figure 7 It is the torque command distribution process of the present invention Figure 3 。

[0036] Figure 8 It is a schematic diagram of torque command distribution for three motors.

[0037] Figure 9 It is a total energy efficiency diagram of three motors, where (a) is the energy efficiency diagram of the three motors using the average torque command distribution method, and (b) is the energy efficiency diagram of the three motors using the torque command distribution method of the present invention.

[0038] Figure 10 It is a schematic diagram of torque command interpolation.

[0039] Figure 11 It is a schematic diagram of partitioned hysteresis loop.

[0040] Figure 12 It is a schematic diagram of torque command distribution countermeasures when a module fails. Detailed implementation mode

[0041] The detailed description and technical content of the present invention are described below in conjunction with the accompanying drawings. However, the accompanying drawings are only provided for reference and illustration purposes and are not used to limit the present invention.

[0042] Based on the structure of a multi-layer axial flux motor (multi-layer AFM), when the controller receives a torque command, the present invention distributes the total command to each motor module. By optimizing the distribution of the torque command, each motor module can operate within its most efficient power output region, making full use of the better energy efficiency torque section of each motor module to meet the torque instruction requirements. At the same time, it avoids the motor module operating in the low torque range (because the energy efficiency of the motor module is very low at this time), and allows the motor module to operate in the torque range with better energy efficiency for as much time as possible.

[0043] As Figure 1 shown, the torque distribution control method of the multi-layer axial flux motor of the present invention includes the following steps:

[0044] Parameter setting: The multi-layer axial flux motor is provided with N motor modules, and each motor module is provided with a preferred torque and a rated torque. The motor module marked with the smallest rated torque is m1, the motor module marked with the second smallest rated torque is m2, and so on until the motor module marked with the largest rated torque is mN. Set the maximum acceptable torque instruction value (maximum torque instruction value) of the motor as the sum of the rated torques of all motor modules;

[0045] Torque instruction partitioning: Divide N + 1 partitions from 0 to the maximum torque instruction value, where: the lower boundary of partition 1 is 0 and the upper boundary is the preferred torque of m1, the lower boundary of partition 2 is the preferred torque of m1 and the upper boundary is the sum of the preferred torques of m1 and m2, and so on until the lower boundary of partition N is the sum of the preferred torques of m1, m2 to mN-1 and the upper boundary is the sum of the preferred torques of m1, m2 to mN, and the lower boundary of partition N + 1 is the sum of the preferred torques of m1, m2 to mN and the upper boundary is the maximum torque instruction value;

[0046] Torque instruction distribution: Obtain the torque instruction. According to the torque instruction value in partition n, where n is any number from 1 to N. When n is from 1 to N-1, distribute the torque instruction value to motor modules m1 to mn, start motor modules m1 to mn, and turn off other motor modules except m1 to mn. When n is N, distribute the torque instruction value to all motor modules m1 to mN and start all motor modules m1 to mN; According to the torque instruction value in partition N + 1, distribute the torque instruction value to all motor modules m1 to mN and start all motor modules m1 to mN.

[0047] The torque command value in partition 1 means that the torque command value is greater than the lower boundary of partition 1 and less than or equal to the upper boundary of partition 1. At this time, the torque command value is assigned to m1, m1 is started, and other motor modules except m1 are turned off; the torque command value in partition 2 means that the torque command value is greater than the lower boundary of partition 2 and less than or equal to the upper boundary of partition 2. At this time, the torque command value is assigned to m1 and m2, m1 and m2 are started, and other motor modules except m1 and m2 are turned off; the torque command value in partition n means that the torque command value is greater than the lower boundary of partition n and less than or equal to the upper boundary of partition n, where n ranges from 1 to N - 1. The torque command value is assigned to motor modules m1 to mn, motor modules m1 to mn are started, and other motor modules except m1 to mn are turned off. When n is N, the torque command value is assigned to all motor modules m1 to mN, and all motor modules m1 to mN are started.

[0048] The torque command value in partition N + 1 means that the torque command value is greater than the lower boundary of partition N + 1 and less than or equal to the upper boundary of partition N + 1. At this time, the torque command value is assigned to all motor modules m1 to mN, and all motor modules m1 to mN are started.

[0049] In the torque command allocation step:

[0050] When the torque command value is within partitions 1 to N, the torque command value is evenly distributed to each started motor module. If the allocated torque command value is greater than or equal to the preferred torque of a certain motor module, the torque command value allocated to that motor module is limited to the preferred torque of that motor module, and the remaining unallocated torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all torque command value balances are allocated to the motor modules;

[0051] When the torque command value is within partition N + 1, the torque command value is evenly distributed to N motor modules. If the allocated torque command value is greater than or equal to the rated torque of a certain motor module, the torque command value allocated to that motor module is limited to the rated torque of that motor module, and the remaining unallocated torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all torque command value balances are allocated to the motor modules.

[0052] In the torque command allocation step, when the torque command value exceeds the upper boundary of partition N + 1, that is, when the torque command value is greater than the maximum torque command value, the rated torque of each motor module in the N motor modules is assigned to each motor module.

[0053] Such as Figure 2As shown, the structure of the multi-layer axial flux motor used in the present invention. Assume that the multi-layer axial flux motor has N motor modules. The motor module with the smallest rated torque is marked as m1, the motor module with the second smallest rated torque is marked as m2, and so on up to the motor module mN.

[0054] The maximum acceptable motor torque command is the sum of the rated torques of all motor modules:

[0055] T cmd,max = T m1,rated + T m2,rated +…T mN,rated .

[0056] As Figure 3 shown, the torque command value of the present invention ranges from 0 to T cmd,max , and N + 1 partitions are divided:

[0057] Each partition is bounded by the torque command T cmd,sec n cmd,sec . The lower boundary of partition n (where n is any number from 1 to N) is T n-1 , and the upper boundary of partition n is T cmd,sec n ;

[0058] When the torque command value is in partition 1, only module m1 is started and the other modules are turned off;

[0059] When the torque command value is in partition 2, modules m1 and m2 are started and the other modules are turned off;

[0060] When the torque command value is in partition n, modules m1 to mn are started and the other modules are turned off;

[0061] …

[0062] And so on, until when the torque command value is in partition mN or in partition mN + 1, all modules are started.

[0063] Among them, partitions 1 to N are for the motor modules to operate within its 0 to preferred torque; partition N + 1 is for the motor modules to operate within its preferred torque to rated torque.

[0064] A preferred torque can be selected from the energy efficiency map of each motor module. The range of the preferred torque is: a torque that can achieve better efficiency at most speeds; the energy efficiency generally rises from 0 to this torque; it cannot exceed the rated torque of the motor module.

[0065] As Figure 4As shown in the example of the energy efficiency map of the Emrax 228 motor, it can be seen that at 115 Nm, the motor can achieve an efficiency of 94-96% at most speeds. Moreover, before this torque point, the energy efficiency increases with the torque. Also, 115 Nm is less than the rated torque of the motor, 130 Nm. Therefore, its preferred torque is taken as T Emrax 228,optimal = 115 Nm.

[0066] Similarly, the preferred torque of module mn is T mn,optimal (where n is any number from 1 to N).

[0067] When selecting the preferred torque, the order of the rated torques between modules needs to be followed for convenience of calculation (usually this is also the case in actual situations, the larger the rated torque, the larger the preferred torque).

[0068] The calculation of the torque command boundary is as follows:

[0069] T cmd,sec 1 = T m1,optimal ,

[0070] T cmd,sec 2 = T cmd,sec 1 + T m2,optimal ,

[0071] T cmd,sec 3 = T cmd,sec 2 + T m3,optimal ,

[0072] …,

[0073] T cmd,sec n = T cmd,sec n-1 + T mn,optimal ,

[0074] …,

[0075] T cmd,sec N = T cmd,sec N-1 + T mN,optimal .

[0076] The gist of the logic of the torque command distribution in the present invention is that the maximum torque command of each started motor module is its own preferred torque, and the remaining unmet commands are evenly shared by the remaining modules.

[0077] As Figure 5 、 Figure 6 、 Figure 7 shown, for example, in partition n, there are n started motor modules;

[0078] When T cmd / n Less than T m1,optimal When it is less than T, since according to the selection conditions for the previously selected preferred torque, the motor module m1 has the smallest preferred torque, T can be determined cmd / n All are less than the preferred torques of the remaining motor modules, and the torque command values assigned to these n started motor modules are all T cmd / n;

[0079] However, when T cmd / n is greater than or equal to T m1,optimal , according to the foregoing gist, the command of the motor module m1 can be at most T m1,optimal ;

[0080] Then start the loop judgment process:

[0081] k starts to increment from 2 to n, and for each k, calculate the average value of the unsatisfied torque commands as follows:

[0082] T cmd,remain,k =(T cmd -T m1,optimal …-T mk-2,optimal -T mk-1,optimal ) / (n -

[0083] (k - 1));

[0084] If T cmd,remain,k >T mk,optimal , then the torque command value of mk is its own preferred torque T mk,optimal , k is incremented by 1, and the loop continues;

[0085] If T cmd,remain,k ≤T mk,optimal , then the torque commands for the modules mk to mn are T cmd,remain,k .

[0086] Exit the loop;

[0087] If the torque command value is equal to the partition boundary T cmd,sec n , then the torque command values of all motor modules are their own preferred torques.

[0088] In the last partition N + 1, the calculation logic for the commands of each module is the same as that in partition N, except that all the preferred torques are replaced with rated torques. Specifically as follows:

[0089] When T cmd / N is less than T m1,rated , since according to the previous order of module labels, the motor module m1 has the smallest rated torque, T cmd / N can be determined. All are less than the rated torques of the remaining motor modules. Therefore, the torque command values assigned to all N motor modules are all Tcmd / N.

[0090] However, when T cmd / N is greater than or equal to T m1,rated , according to the foregoing gist, the command of the motor module m1 can be at most T m1,rated .

[0091] Then start the loop judgment process:

[0092] k increments from 2 to n, and for each k, calculate the average value of the unsatisfied torque commands as follows:

[0093] T cmd,remain,k =(T cmd -T m1,rated …-T mk-2,rated -T mk-1,rated ) / (n-(k - 1))

[0094] If T cmd,remain,k >T mk,rated , the torque command value of mk is its rated torque T mk,rated , k is incremented by 1, and the loop continues.

[0095] If T cmd,remain,k ≤T mk,rated , the torque commands of the modules from mk to mn are T cmd,remain,k . Jump out of the loop.

[0096] If the torque command is greater than or equal to the maximum command value T cmd,max , the torque command values of all motor modules are their own rated torques.

[0097] Taking three Emrax228 motors as an example (assuming their motor shafts are connected together), their torque command distributions are as Figure 8 shown: As can be seen from the figure, when the motor torque command value is within partition 1, only module 1 needs to output. The commands assigned to the remaining modules are 0; when the motor torque command value crosses partition 2, modules 1 and 2 are activated, and each module is evenly assigned half of the motor torque command value; when the motor torque command value is within partitions 3 and 4, all modules are activated. Each module is evenly assigned one-third of the command; since the 3 modules are the same, there is no need to check whether the assigned command exceeds the preferred torque of the module.

[0098] The total energy efficiency diagrams of the three Emrax228 motors are as Figure 9 shown, where (a) is the energy efficiency diagram of the three motors using the average distribution method of torque commands, and (b) is the energy efficiency diagram of the three motors using the torque command distribution method of the present invention. It can be seen that after using the method of the present invention, the motor can maintain the best energy efficiency in most working conditions.

[0099] Considering the discontinuity caused by changing the torque command (especially when the command crosses the boundary), as Figure 10 shown, linearly interpolate the change between commands:

[0100] The controller receives the command T cmd,t at a relatively low frequency f cmd update , such as 100 Hz (so the period is Prd cmd upadte . In this example, it is 10 ms). t represents the time series, and T cmd,t represents the command received by the controller at time t, and T cmd,t+1 represents the command received by the controller at time t + 1. The difference between t + 1 and t is Prd cmd upadte .

[0101] The controller executes the torque command T inter.cmd,a at a relatively high frequency f inter.cmd , such as 1000 Hz (so the period is Prd inter.cmd . In this example, it is 1 ms).

[0102] After the controller receives two commands (T cmd,t and T cmd,t+1 ), that is, after one receiving period Prd cmd udpate has passed, the difference ΔT cmd between the two commands can be calculated as ΔT cmd,t+1 = T cmd,t .

[0103] In the second receiving period, ΔT cmd will be subdivided and added to T cmd,t . Each time the torque command is allocated and executed, the interpolated torque command T inter.cmd,a will be used. The calculation is as follows:

[0104]

[0105] A = f inter.cmd / f cmd update

[0106] where a increments from 1 to A - 1 to generate A - 1 interpolated torque commands T inter.cmd,a to correspond to the A - 1 interpolation periods between t and t + 1. This interpolation method slows down the command change between torque command allocation periods. This approach will cause the controller to always be delayed by one Prd cmd udpate in its response to the torque command. But as long as fcmd update If it is relatively high, there will be no obvious delay in responding to the torque command.

[0107] To prevent the torque command from oscillating at the boundary and causing frequent and unnecessary startup / shutdown of the motor module, a design with a hysteresis loop is adopted at the boundary as Figure 11 shown:

[0108] When n ranges from 2 to N, a hysteresis zone is added before and after the partition boundary. The lower boundary of partition n is defined by T cmd,sec n-1 +T buffer and T cmd,sec n-1 -T buffer as the boundary;

[0109] When T cmd exceeds the increased hysteresis zone (T cmd,sec n-1 +T buffer ), the motor module mn will be started;

[0110] Similarly, when T cmd is lower than the decreased hysteresis zone (T cmd,sec n-1 -T buffer ), the motor module mn will be shut down.

[0111] When one of the started modules fails, due to the structure of the motor (see Figure 2 ), if one of the started modules fails, the controller can immediately turn off the faulty module and redistribute its torque command to the remaining modules that are still operating normally, or immediately start other idle modules to fill the torque gap caused by the faulty module. This will affect the energy efficiency but enhance the redundancy of the motor.

[0112] Take Figure 12 as an example. Originally, the motor modules m1 to mN are all operating normally and outputting relative torque commands T cmd,m1 …T cmd,mN . However, when the module mn suddenly fails (such as overcurrent, overheating...), the controller immediately shuts down the module mn and evenly distributes its command to the remaining modules (it can be seen that the torque commands of the modules other than mn are all increased by ).

[0113] The above-described embodiments are merely preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Any equivalent substitution or transformation made by those skilled in the art on the basis of the present invention shall fall within the protection scope of the present invention.

Claims

1. A torque distribution control method for a multi-layer axial flux motor, characterized in that, It includes the following steps: Parameter setting: The multi-layer axial flux motor is provided with N motor modules, and each motor module is provided with a preferred torque and a rated torque. The motor module marked with the minimum rated torque is m1, the motor module marked with the second smallest rated torque is m2, and so on until the motor module marked with the maximum rated torque is mN. Set the maximum torque command value as the sum of the rated torques of all motor modules; Torque command partitioning: Divide N + 1 partitions from 0 to the maximum torque command value. Among them: The lower boundary of partition 1 is 0, and the upper boundary is the preferred torque of m1. The lower boundary of partition 2 is the preferred torque of m1, and the upper boundary is the sum of the preferred torques of m1 and m2. And so on until the lower boundary of partition N is the sum of the preferred torques of m1, m2 to mN - 1, and the upper boundary is the sum of the preferred torques of m1, m2 to mN. The lower boundary of partition N + 1 is the sum of the preferred torques of m1, m2 to mN, and the upper boundary is the maximum torque command value; Torque command distribution: Obtain the torque command. According to the torque command value in partition n, where n is any number from 1 to N. When n is from 1 to N - 1, distribute the torque command value to motor modules m1 to mn, start motor modules m1 to mn, and turn off the other motor modules except m1 to mn. When n is N, distribute the torque command value to all motor modules m1 to mN, and start all motor modules m1 to mN; According to the torque command value in partition N + 1, distribute the torque command value to all motor modules m1 to mN, and start all motor modules m1 to mN.

2. The torque distribution control method of the multi-layer axial flux motor according to claim 1, characterized in that The torque command value in partition 1 means that the torque command value is greater than the lower boundary of partition 1 and less than or equal to the upper boundary of partition 1. At this time, distribute the torque command value to m1, start m1, and turn off the other motor modules except m1. The torque command value in partition 2 means that the torque command value is greater than the lower boundary of partition 2 and less than or equal to the upper boundary of partition 2. At this time, distribute the torque command value to m1 and m2, start m1 and m2, and turn off the other motor modules except m1 and m2. The torque command value in partition n means that the torque command value is greater than the lower boundary of partition n and less than or equal to the upper boundary of partition n. When n is from 1 to N - 1, distribute the torque command value to motor modules m1 to mn, start motor modules m1 to mn, and turn off the other motor modules except m1 to mn. When n is N, distribute the torque command value to all motor modules m1 to mN, and start all motor modules m1 to mN. The torque command value in partition N + 1 means that the torque command value is greater than the lower boundary of partition N + 1 and less than or equal to the upper boundary of partition N + 1. At this time, distribute the torque command value to all motor modules m1 to mN, and start all motor modules m1 to mN.

3. The torque distribution control method of the multi-layer axial flux motor according to claim 1 or 2, characterized in that In the said torque command distribution step: When the torque command value is within partitions 1 to N, the torque command value is evenly distributed to each activated motor module. If the distributed torque command value is greater than or equal to the preferred torque of a certain motor module, the torque command value assigned to this motor module is limited to the preferred torque of this motor module, and the remaining undistributed torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all the torque command value balances are distributed to the motor modules; When the torque command value is within partition N + 1, the torque command value is evenly distributed to N motor modules. If the distributed torque command value is greater than or equal to the rated torque of a certain motor module, the torque command value assigned to this motor module is limited to the rated torque of this motor module, and the remaining undistributed torque command value balance is evenly distributed to the remaining motor modules. Repeat the above steps until all the torque command value balances are distributed to the motor modules.

4. The torque distribution control method for a multi-layer axial flux motor as claimed in claim 3, wherein in the torque command distribution step, when the torque command value exceeds the upper boundary of partition N + 1, the rated torque of each motor module among the N motor modules is assigned to each motor module.

5. The torque distribution control method for a multi-layer axial flux motor as claimed in claim 1 or 2, wherein In the torque command distribution step, after the controller receives two commands T cmd,t and T cmd,t+1 , it calculates the difference ΔT cmd = T cmd,t+1 - T cmd,t . Each time the torque command distribution is executed, the interpolated torque command T inter.cmd,a is used: A = f inter.cmd / f cmdupdate , Among them, t represents the time series, and T cmd,t represents the instruction received by the controller at time t, and T cmd,t+1 represents the instruction received by the controller at time t + 1, and f inter.cmd is the frequency allocated by the controller to execute the torque instruction T inter.cmd,a and f cmdupdate is the frequency at which the controller receives the instruction T cmd,t a increases from 1 to A - 1 to generate A - 1 interpolated torque instructions T inter.cmd,a to correspond to the A - 1 interpolation periods between t and t + 1.

6. The torque distribution control method of the multi-layer axial flux motor according to claim 1 or 2, characterized in that In the torque command partitioning step, N + 1 partitions are divided from 0 to the maximum torque command value, and each partition is bounded by a torque command boundary T cmd,secn as the boundary. The lower boundary of partition n is T cmd,secn-1 , and the upper boundary of partition n is T cmd,secn . When n is from 2 to N, a hysteresis zone T buffer is added before and after the partition boundary. The lower boundary of partition n is bounded by T cmd,secn-1 +T buffer and T cmd,secn-1 -T buffer . When the torque command T cmd exceeds T cmd,secn-1 +T buffer , the motor module mn is started. When the torque command T cmd is lower than T cmd,secn-1 -T buffer , the motor module mn is turned off.

Images