A sine flux axial flux motor and its optimization method

Abstract

The present invention discloses a sinusoidal magnetic flux motor and an optimization method thereof. The motor comprises a motor stator and a motor rotor symmetrically arranged on both sides of the motor stator; a plurality of fan-shaped permanent magnet units are evenly arranged on the motor rotor end disk, each permanent magnet unit comprising a sintered NdFeB permanent magnet, a bonded NdFeB permanent magnet, and a ferrite permanent magnet. The sintered NdFeB permanent magnet, the bonded NdFeB permanent magnet, and the ferrite permanent magnet are all fan-shaped and arranged in sequence according to their remanence to form sinusoidal magnetic poles. The present invention utilizes high-permeability, low-loss ferrite in combination with the NdFeB design, significantly reducing motor cost and loss.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet motors, and in particular to a magnetic flux sinusoidal axial flux motor and an optimization method thereof. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Permanent magnet synchronous motors (PMSMs) feature a simple structure, high power density, high efficiency, high power factor, and high reliability. However, the proportion of high-quality PMSMs in practical applications is relatively small. Existing motor structures often have non-sinusoidal air gap magnetic field waveforms, which contain numerous high-order harmonics and significantly impact motor performance. This increases copper, iron, and miscellaneous losses, reduces efficiency, and increases temperature rise, negating the energy-saving advantages of PMSMs. The non-sinusoidal air gap magnetic field waveform also generates pulsating electromagnetic torque, causing speed oscillations and impacting operational stability and reliability. This directly limits the use and widespread adoption of PMSMs.

[0004] Due to the shape limitations of the permanent magnets, permanent magnet synchronous motors (PMSMs) easily achieve rectangular air gap magnetic field waveforms. The width of the air gap magnetic field waveform can be varied by adjusting the angle between the center of the circle spanned by the permanent magnets in the air gap. To reduce the harmonics of the air gap magnetic flux density in PMSMs, pole clipping (equal thickness poles) is one of the most effective methods for motor optimization. This method optimizes the shape and length of the pole arcs on the rotor permanent magnets or the outer surface of the rotor core, thereby improving the sinusoidality of the radial air gap magnetic flux density, reducing the harmonic content of the air gap magnetic flux density, suppressing torque ripple, and reducing vibration and noise. Currently, the main pole clipping methods include sinusoidal, inverse cosine, and eccentric pole clipping. All of these methods require the permanent magnets to be individually machined into a specific shape, which poses numerous dimensional parameters and high process requirements.

[0005] In addition, although the permanent magnets of the existing motor skew pole scheme are sinusoidal in shape, the air gap flux density is not completely sinusoidal, and the resulting cogging torque and torque pulsation are still relatively high. In addition, the sinusoidal shape of the magnetic poles requires higher processing and magnetization accuracy of the permanent magnets, resulting in increased costs. Summary of the Invention

[0006] In order to solve the above problems, the present invention proposes a flux sinusoidal axial flux motor and an optimization method thereof. Through the sinusoidal magnetic pole design, the flux sinusoidal motor can effectively reduce the air gap magnetic density and back electromotive force harmonics, and greatly improve the sinusoidality.

[0007] In some embodiments, the following technical solutions are adopted:

[0008] A magnetic flux sinusoidal axial flux motor, comprising:

[0009] A motor stator, and a motor rotor symmetrically arranged on both sides of the motor stator;

[0010] A plurality of fan-shaped permanent magnet units are evenly arranged on the motor rotor end disk, and each permanent magnet unit includes a sintered NdFeB permanent magnet, a bonded NdFeB permanent magnet and a ferrite permanent magnet. The sintered NdFeB permanent magnet, the bonded NdFeB permanent magnet and the ferrite permanent magnet are all fan-shaped and are arranged in sequence according to different remanence to form sinusoidal magnetic poles.

[0011] Specifically, the permanent magnet arrangement of the permanent magnet unit is as follows: the outermost layer is symmetrically arranged ferrite permanent magnets, the second outermost layer is symmetrically arranged bonded NdFeB permanent magnets, and the innermost layer is symmetrically arranged sintered NdFeB permanent magnets; each permanent magnet is tightly attached in sequence to form a sinusoidal magnetic pole with the same magnetization direction.

[0012] The magnetization directions of adjacent permanent magnet units are opposite.

[0013] In other embodiments, the following technical solutions are adopted:

[0014] A parameter optimization method for a sinusoidal flux axial flux motor, comprising:

[0015] The sector angle and thickness of each permanent magnet in each permanent magnet unit are used as optimization variables, and the average torque, torque ripple, cogging torque and back electromotive force harmonic rate are used as optimization targets.

[0016] Perform Latin hypercube sampling on the optimization variables. Based on the Latin hypercube sampling results, perform finite element solution on the sample data to obtain the corresponding average torque, torque ripple, cogging torque and back electromotive force harmonic rate values.

[0017] Based on the sampling results of the optimization variables and the corresponding optimization target values, the Kriging method is used to find the relationship between the optimization variables and the optimization targets, and a response surface model is established;

[0018] The particle swarm optimization algorithm is used to optimize the response surface model to obtain the optimal parameters of the sector angle and thickness of the permanent magnet.

[0019] The optimal permanent magnet sector angle and thickness parameters are brought into the finite element model for solution. The average torque, torque ripple, cogging torque and back-electromotive force harmonic rate obtained by the solution are compared with the average torque, torque ripple, cogging torque and back-electromotive force harmonic rate values ​​obtained by the response surface model to verify the performance of the response surface model.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] (1) Unlike pole clipping, which requires special processing of permanent magnets, the present invention only requires material selection for commonly used sector-shaped permanent magnets and processing of the sector angles corresponding to the sector-shaped permanent magnets. No additional permanent magnet cutting and magnetization processes are required, and the processing is convenient and easier to produce in factories. In addition, the air gap flux density is completely sinusoidal, and the torque pulsation and cogging torque are lower, which meets the design requirements of high-performance motors.

[0022] (2) The flux sinusoidal axial flux motor of the present invention, through the sinusoidal magnetic pole design, can effectively reduce the air gap flux density and back electromotive force harmonics, greatly improve the sinusoidality, minimize the cogging torque and torque pulsation; and while eliminating harmonics, it reduces the motor iron loss, helping the motor to maintain high efficiency operation. At the same time, the use of high magnetic permeability, low loss ferrite combined with the neodymium iron boron design also has a significant effect on reducing motor costs and losses. Based on the consideration of reducing torque pulsation and the cost of permanent magnet materials, a variety of specifications of permanent magnets are used to design the surface-mounted permanent magnet motor, without the need for a special permanent magnet molding process.

[0023] (3) The optimization scheme of the axial flux motor based on the sinusoidal flux linkage proposed in the present invention can make the air gap flux waveform tend to be sinusoidal through the global multi-objective optimization method combining Kriging modeling and particle swarm algorithm, effectively reduce the harmonic content, and reduce the cogging torque and torque pulsation while ensuring a higher electromagnetic torque; the establishment of an agent model through Kriging can effectively reduce the simulation calculation time, and the designed permanent magnet motor can significantly improve the sinusoidality of the stator and rotor magnetomotive force, air gap flux and back electromotive force, significantly reduce the cogging torque and torque pulsation and save the cost of permanent magnets, which is conducive to the stable operation and mass production of permanent magnet motors.

[0024] Other features and advantages of additional aspects of the present invention will be given in part in the following description and in part will become obvious from the following description or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of an exploded structure of a sinusoidal axial flux motor according to an embodiment of the present invention;

[0026] Figure 2(a)-Figure 2(b) They are schematic diagrams of two rotor structures of the magnetic flux sinusoidal axial flux motor according to an embodiment of the present invention;

[0027] FIG3(a) is a schematic diagram of the principle of a conventional flux motor; FIG3(b) is a schematic diagram of the principle of sinusoidal flux motor with sinusoidal flux linkage according to an embodiment of the present invention;

[0028] Figure 4 Schematic diagram of a parameter optimization method for a flux sinusoidal axial flux motor according to an embodiment of the present invention;

[0029] 5(a)-(c) are schematic diagrams of the final optimization results of the flux sinusoidal axial flux motor according to an embodiment of the present invention;

[0030] Among them, 1. Stator; 1-1. Stator core; 1-2. Winding; 2. Sinusoidal pole rotor I; 2-1. End plate; 2-2. N-pole ferrite permanent magnet; 2-3. N-pole bonded NdFeB permanent magnet; 2-4. N-pole sintered NdFeB permanent magnet; 2-5. S-pole ferrite permanent magnet; 2-6. S-pole bonded NdFeB permanent magnet; 2-7. S-pole sintered NdFeB permanent magnet; 3. Sinusoidal pole rotor II. DETAILED DESCRIPTION

[0031] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0032] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0033] Example 1

[0034] In one or more embodiments, a flux sinusoidal axial flux motor is disclosed, referring to Figure 1 , specifically including: a motor stator 1, and a sinusoidal magnetic pole rotor I and a sinusoidal magnetic pole rotor II symmetrically arranged on both sides of the motor stator.

[0035] Among them, the motor stator includes an annular stator core 1-1 and a flat wire winding 1-2 wound on the annular stator core; the sinusoidal pole rotor I2 and the sinusoidal pole rotor II3 have the same structure. Taking the sinusoidal pole rotor I as an example, the motor rotor end disk 2-1 is made of high-performance silicon steel sheets; a plurality of fan-shaped permanent magnet units are evenly arranged on the motor rotor end disk, and the magnetization directions of adjacent permanent magnet units are opposite, that is, each permanent magnet unit is alternately arranged with N poles and S poles; each permanent magnet unit is composed of fan-shaped sintered NdFeB permanent magnets, bonded NdFeB permanent magnets and ferrite permanent magnets. According to the different remanence, the sintered NdFeB permanent magnets, bonded NdFeB permanent magnets and ferrite permanent magnets are arranged in an order to form sinusoidal magnetic poles.

[0036] As a specific implementation scheme, combined with the structure of the two motor rotors disclosed in Figures 2(a)-(b), the arrangement of the permanent magnets in each permanent magnet unit is specifically as follows: the outermost layer is symmetrically arranged with N-pole ferrite permanent magnets 2-2 and S-pole ferrite permanent magnets 2-5, the second outermost layer is symmetrically arranged with N-pole bonded NdFeB permanent magnets 2-3 and S-pole bonded NdFeB permanent magnets 2-6, and the innermost layer is symmetrically arranged with N-pole sintered NdFeB permanent magnets 2-4 and S-pole sintered NdFeB permanent magnets 2-7; the permanent magnets are tightly fitted in sequence, and by designing the remanence of each permanent magnet, sinusoidal magnetic poles with the same magnetization direction are obtained, as shown in Figure 3(b).

[0037] Figure 3(a) is a schematic diagram of a traditional flux motor. This traditional solution has high back electromotive force and air gap magnetic field harmonic rates, large cogging torque, and high operating torque pulsation. The cost is also relatively high compared to permanent magnets, which does not meet the requirements of high-performance motors.

[0038] This embodiment uses a sinusoidal magnetic pole design to create a sinusoidal flux linkage motor that effectively reduces air gap flux density and back electromotive force harmonics, significantly improving sinusoidality and minimizing cogging torque and torque pulsation. Eliminating harmonics reduces motor iron loss, helping the motor maintain high-efficiency operation. At the same time, the use of high-permeability, low-loss ferrites in conjunction with a neodymium iron boron design significantly reduces motor cost and losses. Based on considerations of reducing torque pulsation and permanent magnet material costs, a surface-mounted permanent magnet motor is designed using a variety of permanent magnet specifications, eliminating the need for a special permanent magnet molding process.

[0039] In this embodiment, the sector angle and thickness of each permanent magnet in each permanent magnet unit are used as optimization variables, a reasonable optimization variable range is determined, and the average torque, torque ripple, cogging torque and back electromotive force harmonic rate are used as optimization targets; an optimization scheme is designed to optimize the sector angle and thickness of each permanent magnet.

[0040] In order to improve optimization efficiency and save optimization costs, combined Figure 4 , the optimization scheme of this embodiment is as follows:

[0041] (1) Taking the sector angle and thickness of each permanent magnet in each permanent magnet unit as optimization variables, determine the reasonable optimization variable range, and take the average torque, torque ripple, cogging torque and back electromotive force harmonic rate as optimization targets;

[0042] Since there are constraints between multiple optimization objectives that cannot be optimized at the same time, and the importance of the four optimization objectives is also different, the weight coefficient λ is introduced to establish the optimization model (optimization objective function). The model includes four response surface models of average torque, torque ripple, cogging torque and back electromotive force harmonic rate, as shown below:

[0043]

[0044] Where x i is the optimization variable, f(x i ) is the optimization variable x i The corresponding optimization objective function value when changing, i = 1, 2, 3... T av0 、T r0 、T c0 are the initial values ​​of average torque, torque ripple, cogging torque and back EMF harmonic rate, T av (x i ), T r (x i ), T c (x i ), E THD (x i ) are the optimization variables x i The corresponding average torque, torque pulsation, cogging torque and back electromotive force harmonic rate, λ1, λ2, λ3, λ4 are the weight coefficients of average torque, torque pulsation, cogging torque and back electromotive force harmonic rate respectively, and satisfy λ1+λ2+λ3+λ4=1. Since the flux sinusoidal axial flux motor pays more attention to the output torque performance and the reduction of torque pulsation, the weight coefficient λ1 of the average torque is set to 0.4, the weight coefficient λ2 of the torque pulsation is set to 0.3, the weight coefficient λ3 of the cogging torque is set to 0.2, and the weight coefficient λ4 of the back electromotive force harmonic rate is set to 0.1.

[0045] (2) Perform Latin hypercube sampling on the optimization variables and perform finite element solution on the sample data based on the Latin hypercube sampling results;

[0046] In this embodiment, the initial sample of optimization variables is obtained by Latin hypercube sampling technology, namely, the sample data: angle and thickness values ​​of sector ferrite, bonded NdFeB, and sintered NdFeB permanent magnets;

[0047] Latin hypercube sampling divides the range of each input optimization variable (the angle and thickness of the fan-shaped ferrite, bonded NdFeB, and sintered NdFeB permanent magnet) into intervals, and a random sampling observation is performed on the input variable within each interval; the specific implementation method of Latin hypercube sampling is prior art and will not be described in detail in this embodiment. In practical applications, Latin hypercube sampling points can be randomly generated, with the advantages of high efficiency and strong representativeness, and can more accurately map the relationship between input and output, and the amount of sample data is relatively small.

[0048] The angles and thicknesses of the sampled sector ferrites, bonded NdFeB magnets, and sintered NdFeB permanent magnets were brought into the finite element simulation tool to establish models with different parameters and solve the results to obtain the average torque, torque ripple, cogging torque, and back electromotive force harmonic rate outputs.

[0049] (3) Based on the sampling results of the optimization variables and the corresponding optimization target values, the Kriging method is used to find the relationship between the optimization variables and the optimization targets, and a response surface model is established, which is the optimization model; the model includes four response surface models of average torque, torque pulsation, cogging torque and back electromotive force harmonic rate.

[0050] We have already collected sample data using the Latin hypercube sampling method. We only need to select an appropriate proxy model to fit the sample data. To ensure consistency between the response surface model and the actual motor output objective function, this embodiment uses Kriging to identify the relationship between the optimization variables and the optimization objective and establish a response surface model. Kriging is a semi-parametric interpolation method that uses known sample information to estimate the unknown information of unknown points. It has better fit for high-order nonlinear problems and can more accurately estimate sample trends and dynamics.

[0051] The established response surface model is trained and repeatedly verified to ensure that the response surface model results and the finite element results are within the allowable goodness of fit range.

[0052] (4) The particle swarm algorithm is used to optimize the response surface model to obtain the optimal fan angle and thickness parameters of the permanent magnet.

[0053] In this embodiment, the response surface model is optimized by particle swarm optimization, and the size parameters of the optimal response surface model (optimal angles and thicknesses of sector ferrite, bonded NdFeB, and sintered NdFeB permanent magnets) are obtained.

[0054] In order to verify the accuracy of the optimization of the surrogate model, the dimensional parameters of the optimal surrogate model were substituted into the finite element model for simulation analysis. The average torque, torque ripple, cogging torque and back-electromotive force harmonic rate of the finite element simulation were obtained, and compared with the average torque, torque ripple, cogging torque and back-electromotive force harmonic rate calculated in the response surface model to check whether the performance matches.

[0055] Figure 5(a) shows the no-load back electromotive force of the flux-sinusoidal axial flux motor of this embodiment. It can be seen that the motor back electromotive force has a high sinusoidal degree, and the harmonic rate of the back electromotive force is significantly reduced compared to general axial motors. Figure 5(b) shows the no-load air gap magnetic flux sampling results of the flux-sinusoidal axial flux motor designed in this embodiment. It can be seen that the motor air gap magnetic flux has a very high sinusoidal degree, and the harmonic rate of the air gap magnetic flux is significantly reduced compared to the square wave magnetic flux of general axial motors. Figure 5(c) shows the rated load operating torque results of the flux-sinusoidal axial flux motor designed in this embodiment. Compared to general axial motors, it can be seen that the motor torque pulsation is very small, and the output average torque is relatively effectively improved, meeting high performance requirements.

[0056] The parameter optimization method of this embodiment has the advantages of high optimization accuracy, strong algorithm adaptability, and short calculation time. The rotor magnetomotive force, air gap magnetic flux, and back electromotive force obtained in the optimization results are highly sinusoidal, which significantly reduces the cogging torque and torque pulsation and can save the cost of permanent magnets.

[0057] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A magnetic flux sinusoidal axial flux motor, characterized in that: include: A motor stator, and a motor rotor symmetrically arranged on both sides of the motor stator; A plurality of fan-shaped permanent magnet units are evenly arranged on the motor rotor end disk, each permanent magnet unit includes a sintered NdFeB permanent magnet, a bonded NdFeB permanent magnet and a ferrite permanent magnet, and the sintered NdFeB permanent magnet, the bonded NdFeB permanent magnet and the ferrite permanent magnet are all fan-shaped and arranged in sequence according to different remanence to form sinusoidal magnetic poles; The permanent magnets of each permanent magnet unit are arranged as follows: the outermost layer is symmetrically arranged ferrite permanent magnets, the second outermost layer is symmetrically arranged bonded NdFeB permanent magnets, and the innermost layer is symmetrically arranged sintered NdFeB permanent magnets; each permanent magnet is closely attached in sequence to form a sinusoidal magnetic pole with the same magnetization direction; The magnetization directions of adjacent permanent magnet units are opposite.

2. The flux sinusoidal axial flux motor according to claim 1, characterized in that: The sector angle and thickness of each permanent magnet in each permanent magnet unit are used as optimization variables, and the average torque, torque ripple, cogging torque and back electromotive force harmonic rate are used as optimization targets. An optimization model is established and optimized to obtain the optimized sector angle and thickness parameters.

3. The flux sinusoidal axial flux motor according to claim 2, characterized in that: The process of establishing an optimization model is: Perform Latin hypercube sampling on the optimization variables. Based on the Latin hypercube sampling results, perform finite element solution on the sample data to obtain the corresponding average torque, torque ripple, cogging torque and back electromotive force harmonic rate values. Based on the sampling results of the optimization variables and the corresponding optimization target values, the Kriging method is used to find the relationship between the optimization variables and the optimization targets, and a response surface model is established, which is the optimization model. The particle swarm optimization algorithm is used to optimize the response surface model to obtain the optimal parameters of the sector angle and thickness of the permanent magnet.

4. The flux sinusoidal axial flux motor according to claim 2, characterized in that: The response surface model established is as follows: in, x i To optimize the variables, f ( x i ) is the optimization variable x i The corresponding optimization objective function value when changes, i =1, 2, 3…; T av0 、 T r0 、 T c0 、 are the initial values ​​of average torque, torque ripple, cogging torque and back EMF harmonic rate, respectively, T av ( x i )、 T r ( x i )、 T c ( x i )、 E THD ( x i ) are the optimization variables x i The corresponding average torque, torque ripple, cogging torque and back EMF harmonic rate, λ 1. λ 2. λ 3. λ 4 are the weight coefficients of average torque, torque ripple, cogging torque and back electromotive force harmonic rate respectively.

5. The flux sinusoidal axial flux motor according to claim 2, characterized in that: The optimal permanent magnet sector angle and thickness parameters are brought into the finite element model for solution. The average torque, torque ripple, cogging torque and back-electromotive force harmonic rate obtained by the solution are compared with the average torque, torque ripple, cogging torque and back-electromotive force harmonic rate values ​​obtained by the response surface model to verify the performance of the response surface model.

6. A parameter optimization method for a flux sinusoidal axial flux motor, applied to the flux sinusoidal axial flux motor according to any one of claims 1 to 5, characterized in that: include: The sector angle and thickness of each permanent magnet in each permanent magnet unit are used as optimization variables, and the average torque, torque ripple, cogging torque and back electromotive force harmonic rate are used as optimization targets. Perform Latin hypercube sampling on the optimization variables. Based on the Latin hypercube sampling results, perform finite element solution on the sample data to obtain the corresponding average torque, torque ripple, cogging torque and back electromotive force harmonic rate values. Based on the sampling results of the optimization variables and the corresponding optimization target values, the Kriging method is used to find the relationship between the optimization variables and the optimization targets, and a response surface model is established; The particle swarm optimization algorithm is used to optimize the response surface model to obtain the optimal parameters of the sector angle and thickness of the permanent magnet.

7. The parameter optimization method of a flux sinusoidal axial flux motor according to claim 6, characterized in that: The response surface model established is as follows: in, x i To optimize the variables, f ( x i ) is the optimization variable x i The corresponding optimization objective function value when changes, i =1, 2, 3…; T av0 、 T r0 、 T c0 、 are the initial values ​​of average torque, torque ripple, cogging torque and back EMF harmonic rate, respectively, T av ( x i )、 T r ( x i )、 T c ( x i )、 E THD ( x i ) are the optimization variables x i The corresponding average torque, torque ripple, cogging torque and back EMF harmonic rate, λ 1. λ 2. λ 3. λ 4 are the weight coefficients of average torque, torque ripple, cogging torque and back electromotive force harmonic rate respectively.

8. The parameter optimization method of a flux sinusoidal axial flux motor according to claim 6, characterized in that: Also includes: The optimal permanent magnet sector angle and thickness parameters are brought into the finite element model for solution. The average torque, torque ripple, cogging torque and back-electromotive force harmonic rate obtained by the solution are compared with the average torque, torque ripple, cogging torque and back-electromotive force harmonic rate values ​​obtained by the response surface model to verify the performance of the response surface model.

Images