External rotor permanent magnet synchronous motor noise prediction method, system, equipment and medium

Abstract

The invention discloses an external rotor permanent magnet synchronous motor noise prediction method, system and device and a medium, and relates to the technical field of motor noise, and the method comprises the steps: building a two-dimensional electromagnetic model of an external rotor permanent magnet synchronous motor ERPMSM; analyzing the spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet through the two-dimensional electromagnetic model to obtain the electromagnetic force; establishing a three-dimensional structure model of the ERPMSM, and loading the electromagnetic force into the three-dimensional structure model in a nodal force form to obtain a nodal force acting on the surface of the permanent magnet in a three-dimensional space; based on the three-dimensional structure model and the nodal force, the radial vibration acceleration of the ERPMSM outer rotor is calculated by adopting a mode superposition method; and taking the radial vibration of the outer rotor as an acoustic boundary condition, and predicting the noise of the ERPMSM by using an acoustic boundary element method.

Description

Technical Field

[0001] The present invention relates to the technical field of motor noise prediction, and in particular to a method, system, equipment and medium for predicting noise of an outer rotor permanent magnet synchronous motor. Background Art

[0002] External rotor permanent magnet synchronous motor (ERPMSM) has the advantages of short axial size, high power / torque density, wide speed regulation range, good reliability, strong overload capacity and high efficiency. In recent years, ERPMSM has been increasingly used in new energy vehicles, but the electromagnetic noise it generates is one of the main noise sources of the vehicle. Therefore, how to reduce the electromagnetic noise of this type of motor is a key issue in the design process of automotive motors, which is of great significance to the overall ride comfort and safety of electric vehicles.

[0003] For PMSMs with different pole-slot combinations, fractional slots have advantages over integer slots, but they also have disadvantages such as low electromagnetic force spatial order, prominent electromagnetic vibration and noise problems. At present, in order to solve these problems. ZQ Zhu et al. studied the influence of different pole-slot combinations and winding layers on the spatial distribution and frequency characteristics of electromagnetic force on the surface of the motor stator. In addition, Shang Hai et al. studied the influence of different pole-slot combinations on the electromagnetic vibration of the motor, and pointed out that fractional slots will lead to greater radial electromagnetic force and vibration, resulting in more obvious vibration. The stator structure of the PMSM will produce tooth torque, resulting in increased torque pulsation, and produce certain mechanical vibration and noise, thereby affecting the operating performance of the motor.

[0004] At present, in order to weaken the vibration and noise of PMSM, the commonly used method is to reduce the amplitude of the electromagnetic force close to the structural modal frequency. When the motor runs at a low speed, the electromagnetic force wave frequency is close to the modal frequency, which may cause large vibration. Compared with the lower-order vibration mode order, the natural frequency of the motor's high-order mode is higher and is not easy to cause vibration. In order to avoid the motor resonance problem, Zheng Yaoda et al. reduced the vibration noise of PMSM by optimizing the motor stator structure and adjusting the modal frequency that causes the main vibration noise of the motor. R.Islam et al. minimized vibration and noise by forming an optimal gap on the rotor pole surface. In this method, by eliminating the third-order harmonic of the air gap field, the electromagnetic excitation with a harmonic order of 12n, which contributes the most significant component to the overall noise, is reduced. F.Lin et al. found that the vibration noise can be effectively reduced by reducing the low spatial harmonic radial force, and the noise can be suppressed by the optimal magnet tilt angle.

[0005] However, the currently used predictive noise reduction methods have difficulties in accurately modeling and predicting ERPMSM vibration and noise reduction due to the complexity and diversity of noise sources caused by the uneven geometry of permanent magnets, resulting in errors in the prediction results. Summary of the invention

[0006] In view of the deficiency that the existing technology has errors in the prediction results of ERPMSM vibration reduction and noise reduction due to the complexity and diversity of noise sources caused by the uneven geometric shape of permanent magnets when reducing vibration and noise of ERPMSM, the present invention proposes a noise prediction method, system, equipment and medium for an outer rotor permanent magnet synchronous motor. Based on a three-dimensional structural model and nodal forces, the modal superposition method is used to calculate the radial vibration acceleration of the ERPMSM outer rotor; the radial vibration of the outer rotor is used as an acoustic boundary condition, and the acoustic boundary element method is used to predict the noise of the ERPMSM, thereby solving the problems existing in the existing technology.

[0007] A method for predicting noise of an outer rotor permanent magnet synchronous motor comprises the following steps:

[0008] Obtain the structural parameters of the outer rotor permanent magnet synchronous motor ERPMSM;

[0009] A two-dimensional electromagnetic model of ERPMSM is established according to the structural parameters. The spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet are analyzed through the two-dimensional electromagnetic model to obtain the electromagnetic force acting on the surface of the permanent magnet.

[0010] The electromagnetic force is loaded into the three-dimensional structural model of ERPMSM in the form of nodal force to obtain the nodal force acting on the surface of the permanent magnet in three-dimensional space;

[0011] Based on the three-dimensional structural model and nodal force, the modal superposition method is used to calculate the radial vibration acceleration of the outer rotor of the ERPMSM. The radial vibration acceleration of the outer rotor is used as the acoustic boundary condition, and the acoustic boundary element method is used to predict the noise of the ERPMSM.

[0012] Furthermore, the structural parameters of the ERPMSM include the fundamental magnetic field Bmn of the permanent magnet, the fundamental magnetic field Bmn of the armature reaction, av , speed n, pole pair number p, angular velocity ω r 、s v Rotation direction of the vth harmonic of the armature reaction field, rotor position and relative air gap permeance λ a .

[0013] Furthermore, a three-dimensional structural model of the ERPMSM is established according to the pole-slot matching of the fractional slots of the outer rotor permanent magnet synchronous motor, the number of double-layer winding layers, the tooth profile parameters of the stator slots, the air gap length, and the inner and outer radii of the rotor and stator.

[0014] Furthermore, the two-dimensional electromagnetic model of the ERPMSM is established using Ansys Maxwell software, which specifically includes the following steps:

[0015] In a permanent magnet synchronous motor PMSM, according to the relative air gap permeance λ a , the resultant radial flux density B in the air gap rs It is expressed as:

[0016] B rs =(B r_mag +B r_arm ) a

[0017] Among them, B r _mag and Br_arm represent the unslotted radial permanent magnet reaction field and the unslotted radial armature reaction field, respectively;

[0018] Using the Maxwell stress tensor method, ignoring the tangential air gap flux density B ts , then the radial electromagnetic force density f r It is expressed as:

[0019]

[0020] Where u0 is the magnetic permeability of vacuum;

[0021] For ERPMSM, when the outer rotor is fixed, the stator rotates at an angular velocity ω r Rotating in the opposite direction, in the z rotor coordinate, B r_mag , B r_arm and λ a It is expressed as:

[0022]

[0023] Among them, θ is the spatial mechanical angle, t is the time, p is the number of pole pairs, Q is the number of slots, and B mn is the fundamental magnetic field of the permanent magnet, B av is the fundamental armature magnetic field, λ0 is the average air gap permeability, λ a is the relative air gap permeability considering the stator slots, s v Indicates the rotation direction of the vth harmonic of the armature reaction field;

[0024] For PMSMs with different slot-pole combinations and double-layer windings, (s v +v)p=3kN ps ;

[0025] Among them, N ps =GCD(2p,Q) is the greatest common divisor of the number of poles and the number of slots, k is a positive integer;

[0026] Then the radial electromagnetic force density is expressed as:

[0027]

[0028] Among them, λ au1 With λ au2 Represents the real part of the relative permeability when u takes different values.

[0029] Furthermore, the method further includes analyzing the motor cogging torque by using an energy method to obtain influencing factors affecting the motor cogging torque, which specifically includes the following steps:

[0030] Based on the energy method, the cogging torque is defined as:

[0031]

[0032] Wherein, E is the magnetic field energy in the permanent magnet when no power is supplied, and α is the angle between the center line of the permanent magnet and the center line of a stator tooth;

[0033] Since the stator winding is not energized, the magnetic field energy of the motor is expressed as:

[0034]

[0035] Among them, E Gap is the magnetic field energy in the air gap, E PM is the magnetic field energy generated by the permanent magnet, u0 is the vacuum magnetic permeability; B represents the air gap flux density of the fundamental wave permanent magnetic field, and V represents the air gap volume;

[0036] The air gap magnetic flux distribution is expressed as:

[0037]

[0038] Where, hml(θ) is the magnetization length of the permanent magnet, σ(θ, α) is the effective length of the air gap, and Brd(θ) is the remanent magnetization distribution of the permanent magnet;

[0039] The magnetic field energy of the motor is further expressed as:

[0040]

[0041] When the permanent magnet and stator are evenly distributed, B is expanded by Fourier decomposition. rd (θ) and the square of the relative air gap permeance, B 2 rd B after Fourier decomposition of (θ) rn and The Fourier decomposition of G n It is expressed as:

[0042]

[0043] Then the cogging torque T is obtained Cog (α), expressed as:

[0044]

[0045] Among them, l a is the armature core length, z represents the number of stator teeth, r2 is the rotor outer radius, r1 is the stator inner radius, and n is an integer that makes nz / 2p an integer. It can be obtained that the slot width, pole arc coefficient and speed factors all have an impact on the motor's cogging torque.

[0046] Furthermore, the radial vibration acceleration of the ERPMS M outer rotor is calculated by using the modal superposition method based on the three-dimensional structural model and the node force, and the calculation process includes the following steps:

[0047] According to the mass matrix [M] and stiffness matrix [K] of the ERPMSM structure, we can get [K][Φ i ]=λ i [M][Φ i ], where λ i is the i-th eigenvalue, [Φ i ] is the corresponding eigenvector.

[0048] Calculate the modal mass of the ith mode [M i ] and modal stiffness [K i ]:

[0049] [M i ]=[Φ i ] T [M][Φ i ]

[0050] [K i ]=[Φ i ] T [K][Φ i ]=λ i [M i ]

[0051] Calculate the modal participation coefficient Γ of the i-th mode i :

[0052]

[0053] For each mode i, according to the input vibration acceleration Calculate the modal response x i :

[0054]

[0055] Among them, ξ i is the damping ratio of the ith mode;

[0056] The modal response x is obtained by numerical integration method. i , then the total acceleration response of the ERPMSM structure is is the superposition of each modal acceleration response, which is expressed as:

[0057]

[0058] Among them, x i is the node displacement in modal coordinates, {Φ i} is the mode shape of mode order i, {F} is the time-varying electromagnetic nodal force, N is the modulus to be summed, [M], [C] and [K] are the mass matrix, damping matrix and stiffness matrix respectively, and T is the matrix transpose.

[0059] The present invention also includes a noise prediction system for an outer rotor permanent magnet synchronous motor, comprising:

[0060] An acquisition module is used to acquire structural parameters of an outer rotor permanent magnet synchronous motor ERPMSM;

[0061] A two-dimensional electromagnetic model building module is used to establish a two-dimensional electromagnetic model of the ERPMSM according to the structural parameters, and to analyze the spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet through the two-dimensional electromagnetic model to obtain the electromagnetic force acting on the surface of the permanent magnet;

[0062] A loading module is used to load the electromagnetic force into the three-dimensional structural model of the ERPMSM in the form of a nodal force to obtain the nodal force acting on the surface of the permanent magnet in three-dimensional space;

[0063] The prediction module is used to calculate the radial vibration acceleration of the ERPMS M outer rotor based on the three-dimensional structural model and the node force using the modal superposition method; the radial vibration acceleration of the outer rotor is used as the acoustic boundary condition, and the acoustic boundary element method is used to predict the noise of the ERPMSM.

[0064] The present invention also includes a computer device for predicting noise of an outer rotor permanent magnet synchronous motor, comprising: a memory, a processor, and a computer program stored in the memory, wherein the processor implements the steps of the outer rotor permanent magnet synchronous motor noise prediction method when executing the computer program.

[0065] The present invention also includes a readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, they are used to execute the steps of the outer rotor permanent magnet synchronous motor noise prediction method.

[0066] The present invention provides a method, system, device and medium for predicting noise of an outer rotor permanent magnet synchronous motor, which has the following beneficial effects:

[0067] The present invention establishes a two-dimensional electromagnetic model of ERPMSM, conducts in-depth research on the electromagnetic force characteristics of a permanent magnet under different working conditions, obtains the spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet, optimizes the motor, and improves its performance and efficiency; secondly, considering the uneven distribution characteristics of the electromagnetic force on the surface of the permanent magnet, the electromagnetic node force is loaded into the structural model in the form of concentrated force, and the accurate simulation of the structural response is realized, so that the mechanical behavior of the permanent magnet in practical applications can be more effectively evaluated, and potential failure modes can be avoided. The radial vibration acceleration of the outer rotor surface is calculated by the modal superposition method, and the vibration characteristics of the rotor are comprehensively analyzed to provide a basis for subsequent noise prediction. At the same time, the acoustic boundary element method is used to predict the sound radiation of the outer cover, and the accurate evaluation of the sound radiation characteristics is realized; by analyzing the uneven distribution of the electromagnetic force on the surface of the permanent magnet, the prediction accuracy of the vibration reduction and noise reduction of the ERPMSM is improved, and the problem of errors in the prediction results of the vibration reduction and noise reduction of the ERPMSM due to the complexity and diversity of the noise source is avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 Schematic diagram of the stator and rotor structure of the ERPMSM in an embodiment of the present invention;

[0069] Figure 2 The magnetic field line distribution diagram of the permanent magnet surface under no-load and load conditions in the embodiment of the present invention;

[0070] Figure 3 A schematic diagram of an excitation current curve in an embodiment of the present invention;

[0071] Figure 4 This is a spectrum diagram of radial electromagnetic force density when T=0Nm and T=60Nm in an embodiment of the present invention;

[0072] Figure 5 This is a flow chart of the ERPMSM noise prediction method in an embodiment of the present invention;

[0073] Figure 6 is a three-dimensional structural diagram of a motor in an embodiment of the present invention;

[0074] Figure 7 This is a schematic diagram of the modeling of the outer rotor test in an embodiment of the present invention;

[0075] Figure 8 Schematic diagram of the outer rotor modal test in an embodiment of the present invention;

[0076] Fig. 9 Schematic diagram of the modal orders and corresponding modal vibration shapes of the outer rotor in an embodiment of the present invention;

[0077] Fig.10 Schematic diagram of electromagnetic force density distribution and equivalent concentrated force and torque in an embodiment of the present invention;

[0078] Fig.11 Schematic diagram of the permanent magnet node force in an embodiment of the present invention;

[0079] Fig.12 This is a spatial distribution diagram of radial electromagnetic force in an embodiment of the present invention;

[0080] Fig.13 It is a schematic diagram of harmonic analysis of no-load radial electromagnetic force density in an embodiment of the present invention;

[0081] Fig.14 This is the acceleration spectrum of the center of the outer cover in the embodiment of the present invention;

[0082] Fig.15 Schematic diagram of the modal shape and frequency of the local mode of the outer cover in the embodiment of the present invention;

[0083] Fig.16 This is a schematic diagram of an acoustic grid structure in an embodiment of the present invention;

[0084] Fig.17 Schematic diagram of acoustic calculation field point grid in an embodiment of the present invention;

[0085] Fig.18 Schematic diagram of noise spectrum under no-load condition in an embodiment of the present invention;

[0086] Fig.19 Schematic diagram of target optimization comparison under different slot width conditions in an embodiment of the present invention;

[0087] Fig. 20 Schematic diagram of the comparison of the cogging torque between a slot width of 8 mm and a slot width of 4 mm in an embodiment of the present invention. DETAILED DESCRIPTION

[0088] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0089] The present invention proposes a noise prediction method for an outer rotor permanent magnet synchronous motor. First, the spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet are derived, taking into account the influence of the slot-pole combination and the slotting effect. Then, the cogging torque that affects the operating performance of the motor is analyzed, and several types of factors affecting the cogging torque are obtained. The electromagnetic force, structural vibration and sound radiation models of the ERPMSM are established, and modal tests are performed on the outer rotor to verify the accuracy of the finite element model. The electromagnetic force under no-load and load conditions is analyzed under the premise of considering the uneven distribution of the electromagnetic force when the electromagnetic force is loaded on the surface of the permanent magnet. The modal superposition method is used to calculate the radial vibration acceleration of the outer rotor, and the boundary element method (Boundary Element Method, BEM) is used to predict the noise of the ERPMSM. Specifically, the following steps are included:

[0090] S1. Electromagnetic force analysis: The electromagnetic force generated by the permanent magnetic field and armature reaction is calculated using the magnetic potential permeability method, and the relative penetration function λ is introduced a To analyze the influence of slotting effect on electromagnetic field.

[0091] The electromagnetic force is obtained through the fundamental magnetic field B of the permanent magnet of the motor mn , the fundamental magnetic field B of the armature reaction av , motor speed n, motor pole pair number p, motor angular velocity ω r ,s v Rotation direction of the vth harmonic of the armature reaction field, θ rotor position, t time, λ a Parameters such as relative permeability function.

[0092] Using the above parameters, we first calculate the radial electromagnetic field of the permanent magnet and the radial electromagnetic field of the armature reaction under the slotless condition, and then use the magnetic potential permeance method to calculate the radial electromagnetic field of the magnet and the radial electromagnetic field of the armature reaction under the slotted condition.

[0093] Consider the influence of slots: Under slotted conditions, the presence of slots will change the distribution of the magnetic field. Periodic conditions and symmetry can be used to simplify the calculation. By introducing the geometric parameters of the slots, a mathematical model of the slotted motor is established.

[0094] Modification of magnetic field equation: Under slotted conditions, the magnetic field equation needs to consider the boundary conditions of the slot, which can be modified based on the solution under no slot conditions. The perturbation method or finite difference method can be used to analyze the influence of the slot.

[0095] In PMSM, the air gap field is composed of the permanent magnetic field and the armature reaction field, which are expressed by the relative penetration function λ. a , the resultant radial flux density B in the air gap rs It is expressed as:

[0096] Brs =(B r_mag +B r_arm ) a (1)

[0097] In the formula, B r_mag and B r_arm They are respectively represented as the unslotted radial permanent magnet reaction field and the unslotted radial armature reaction field, λ a is the relative air gap permeability considering the stator slots. Using the Maxwell stress tensor method and ignoring the tangential air gap flux density, it can be expressed as:

[0098]

[0099] In the formula, f r is the radial electromagnetic force density, and u0 is the vacuum permeability.

[0100] For ERPMSM, the outer rotor is the main heat sink of electromagnetic noise. Therefore, the present invention focuses on the electromagnetic force acting on the inner surface of the permanent magnet. Therefore, it can be assumed that the outer rotor is fixed and the stator rotates at an angular velocity w r Rotate in the opposite direction. In the z rotor coordinate, B r_mag , B r_arm and λ a can be written as:

[0101]

[0102] Where θ is the spatial mechanical angle, t is the time, p is the number of pole pairs, and Q is the number of slots. mn is the fundamental permanent magnet magnetic field, B av is the fundamental armature magnetic field, λ0 is the average air gap permeability, λ a is the relative air gap permeability considering the stator slots, s v Indicates the rotation direction of the vth harmonic of the armature reaction field, s v When it is 1, it rotates in the positive direction, s v When it is -1, it rotates in the reverse direction (rearward). For PMSMs with different slot-pole combinations and double-layer windings, (s v The value of +v)p is:

[0103] (s v +v)p=3kN ps , k = 1, 2, 3, … (6)

[0104] Where N ps =GCD(2p, Q) is the greatest common divisor of the number of poles and the number of slots. The radial force density can be obtained by substituting (3) and (5) into (7), which can be further expressed as:

[0105]

[0106] According to the polynomial multiplication rule, the result of formula (8) is 9 terms, which can be divided into 6 categories according to the source. The amplitude, spatial order and frequency of these six force sources are obtained by the product of trigonometric functions, as shown in Table 1. According to Table 1, the frequency of the electromagnetic force under no-load is equal to uQf r ; The frequency of the electromagnetic force under load is 3kN ps f r , which comes from the interaction between the permanent magnetic field and the stator slots. When the machine is excited by a sinusoidal current, the frequency of the electromagnetic force acting on the stator surface is equal to 2kpf r Therefore, the frequency characteristics of the radiated noise of the ERPMSM are different from those of the external stator PMSM. However, the spatial characteristics of the two PMSMs are the same, and for the PMSM with double-layer windings, the spatial order is equal to N ps An integer multiple of .

[0107] The present invention adopts ERPMSM (such as Figure 1 The electromagnetic force acting on the surface of the permanent magnet is analyzed by using the ERPMSM (as shown in Table 2). The main parameters of the ERPMSM are shown in Table 2. In order to compare the simulated noise with the experimental noise under a load torque of 500r / min, the electromagnetic force under the same load torque and speed is also calculated. The rated speed of this motor is 300r / min and is used at low speed. When it is overloaded and works at high speed, the amplitude of the current harmonics increases. For an outer rotor with 21 poles and 18 slots, N ps = 3. Therefore, a 2D electromagnetic model for calculating electromagnetic force was established for electromagnetic force analysis.

[0108] Figure 2 The electromagnetic force distribution on the surface of the permanent magnet under no-load and load conditions is shown in Figure 2. Due to the influence of the armature reaction field, the uneven distribution along the surface of the permanent magnet under load is more serious, causing the air gap field to fluctuate greatly. Figure 3 is the frequency spectrum of the radial force density at the middle of the permanent magnet surface under no-load and load. The frequency of the electromagnetic force is equal to uQf r =21×500 / 60×u=175u (u=1, 2, 3,...).

[0109] Table 1 Source, amplitude, spatial order and corresponding frequency of radial electromagnetic force density

[0110]

[0111] Table 2 Main parameters of the motor

[0112]

[0113]

[0114] Under no-load conditions according to Table 1, when the motor is Figure 3 When the current shown is excited, the electromagnetic force harmonics become richer, especially at low frequencies. This part of the force harmonics comes from the interaction between the fundamental permanent magnet magnetic field and the armature reaction field generated by the current harmonics. Due to the low-frequency current harmonics of the fundamental permanent magnetic field, this part of the force is concentrated in the low-frequency band. Therefore, as Figure 4 As shown, in the high frequency band, this part of the force comes from the interaction between the magnetic field and the stator slots, so this part of the force is almost the same under no-load and load conditions, while the rich force harmonics in the low frequency band may increase the possibility of resonance and aggravate vibration and noise.

[0115] S2. Cogging torque analysis: A mathematical model is established to analyze the motor cogging torque. Currently, the methods for defining the cogging torque include the Maxwell stress tensor method, the finite element analysis method, and the energy method. Since the energy method can more accurately calculate the energy loss during motor operation, the present invention uses the energy method to analyze the cogging torque.

[0116] Based on the energy method, the cogging torque is defined as:

[0117]

[0118] Where E is the magnetic field energy in the permanent magnet when no power is applied, and α is the angle between the center line of the permanent magnet and the center line of a stator tooth.

[0119] Since the stator winding is not energized, the magnetic field energy E in the permanent magnet is greatly weakened and can be ignored. Therefore, the magnetic field energy of the motor at this time is:

[0120]

[0121] In the formula, E Gap is the magnetic field energy in the air gap, E PM It is the magnetic field energy generated by the permanent magnet.

[0122] The air gap magnetic flux distribution is expressed as:

[0123]

[0124] Where, hml(θ) is the magnetization length of the permanent magnet, σ(θ, α) is the effective length of the air gap, and Brd(θ) is the remanent magnetization distribution of the permanent magnet.

[0125] Substituting formula (11) into (10), we can obtain the following formula (12):

[0126]

[0127] When the permanent magnet and stator are evenly distributed, B is expanded by Fourier decomposition. rd(θ) and the square of the relative permeability of the air gap. rn and G n They are expressed as follows: (13) and (17) respectively.

[0128]

[0129] The cogging torque is expressed as:

[0130]

[0131] In the formula, l a is the armature core length, z represents the number of stator teeth, r2 is the rotor outer radius, r1 is the stator inner radius, and n is an integer that makes nz / 2p an integer. From formula (18), it can be concluded that factors such as slot width, pole arc coefficient and speed have a certain influence on the cogging torque of the motor.

[0132] S3. Vibration and noise calculation: The vibration multi-physics model and noise prediction model of ERPMSM are proposed. The calculation process is as follows: Figure 5 As shown. First, a two-dimensional electromagnetic finite element model was established to calculate the nodal forces on the surface of the permanent magnet. Considering the periodicity of the ERPMSM, the electromagnetic force was assumed to be uniformly distributed in the axial direction, thereby obtaining the nodal forces in the three-dimensional full model. Subsequently, a three-dimensional structural model of the ERPMSM was established, and the nodal forces were transferred from the electromagnetic model to the structural model by the force interpolation method to consider the uneven distribution of the electromagnetic force. Finally, the modal superposition method was used to calculate the vibration of the ERPMSM, and the acoustic boundary element method was used to predict its noise characteristics. Figure 6 This is the three-dimensional structure diagram of the motor.

[0133] S3.1. Finite element and experimental analysis of outer rotor modal: The outer rotor modal is analyzed by finite element method and verified by modal test, such as Fig. 9 a is the 2nd order modal vibration shape of the outer rotor in finite element analysis, b is the 3rd order modal vibration shape of the outer rotor in finite element analysis, c is the 4th order modal vibration shape of the outer rotor in finite element analysis, d is the 5th order modal vibration shape of the outer rotor in finite element analysis, e is the 2nd order modal vibration shape of the outer rotor in modal experiment, f is the 3rd order modal vibration shape of the outer rotor in modal experiment, g is the 4th order modal vibration shape of the outer rotor in modal experiment, and h is the 5th order modal vibration shape of the outer rotor in modal experiment. When performing finite element analysis, a finite element model is established for the outer rotor with material properties (density is 7650kg / m3, elastic modulus is 200GPa, Poisson's ratio is 0.3). The modal vibration shapes and frequencies of each order of the outer rotor are calculated by finite element algorithm and modal test, as shown in Table 3 below.

[0134] Table 3 Modal order, modal vibration shape and frequency of outer rotor

[0135]

[0136] S3.1.1. Modal test verification: To verify the accuracy of the finite element model of the outer rotor structure, the present invention conducted a modal test on an outer rotor with 20 poles. The test used a hammer modal test, and the test hardware and analysis software were used to evaluate the prototype outer rotor. The modeling and layout of the test rotor are as follows: Figure 7 , Figure 8 These tests are intended to verify the reliability and accuracy of the finite element model.

[0137] According to the modal and test analysis results in Table 3, the relative error is within 7%, thus verifying the accuracy of the outer rotor finite element model. The modal vibration shape and corresponding frequency of the motor outer rotor can be combined with Figure 3 The electromagnetic force amplitude in is analyzed. Figure 4 It can be seen that the low-frequency electromagnetic force amplitude appears at 350Hz, which is quite different from the first five natural frequencies of the outer rotor structure. Therefore, it can be concluded that the electromagnetic force is unlikely to cause resonance in the motor structure.

[0138] S3.2. Nodal force loading: Fig.10 As shown in Figure 1, the electromagnetic force acting on the stator and rotor surfaces is a spatial rotation force wave. Since it is difficult to load this force into the structural model for vibration calculation, the electromagnetic force with uneven distribution along the tooth surface is simplified to be equivalent to the concentrated force and moment acting at the center of the surface.

[0139] Fig.12 is the spatial and temporal distribution diagram of radial electromagnetic force under no-load and load conditions. Fig.12 It can be seen that the amplitude of the force on the permanent magnet surface can vary by about 50%. In order to consider the uneven distribution characteristics of the electromagnetic force on the permanent magnet surface, the electromagnetic force is loaded into the structural model in the form of nodal force. First, a 2D electromagnetic model is established to calculate the nodal force acting on the permanent magnet surface. In addition, according to the periodicity of ERPMSM and assuming that the electromagnetic force distribution is uniform along the axial direction, the nodal force in three-dimensional space is obtained. Fig.11 is the nodal force of the mesh acting on the surface of the permanent magnet, Fig.13 Represents the harmonic analysis of radial electromagnetic force under no-load condition.

[0140] S3.3, Vibration calculation: Based on the structural model and electromagnetic force load, the modal superposition method is used to calculate the radial vibration of the outer cover, which specifically includes the following steps:

[0141] S3.3.1, Modal analysis: solve the eigenvalue problem [K] [Φ] for the mass matrix [M] and stiffness matrix [K] of the structure. i ]=λ i [M][Φ i ], where λ iis the i-th eigenvalue, [Φ i ] is the corresponding eigenvector (modal shape). S3.3.2. Calculate the modal mass [M i ] and modal stiffness [K i ]:

[0142] [M i ]=[Φ i ] T [M][Φ i ]

[0143] [K i ]=[Φ i ] T [K][Φ i ]=λ i [M i ]

[0144] S3.3.3. Calculate the modal participation coefficient Γ of the i-th mode i :

[0145]

[0146] S3.3.4 For each mode i, according to the input vibration acceleration Calculate the modal response x i :

[0147]

[0148] Where ξ i is the damping ratio of the ith mode.

[0149] Solving the above equations, we get the modal response x i , usually numerical integration methods are used, such as the Newmark-beta method or the Runge-Kutta method.

[0150] S3.3.5 The total acceleration response of the structure is the superposition of the acceleration responses of each modal, which is expressed as:

[0151]

[0152] In the formula, x i is the node displacement in modal coordinates, {Φ i} is the mode shape of mode order i, {F} is the time-varying electromagnetic nodal force, N is the modulus to be summed, [M], [C], and [K] are the mass matrix, damping matrix, and stiffness matrix, respectively. i} was numerically calculated and obtained by modal test.

[0153] Fig.14is the acceleration spectrum of the center of the side cover under no-load and load conditions. Figure 4 The rich force harmonics generated by the medium current harmonics make the vibration under load more severe at low frequencies, such as between 0-400Hz. Fig.14 It can be seen that there are two maximum vibration peaks at frequencies of 350Hz and 1450Hz. Since the axial length of the ERPMSM is short and the outer rotor is connected to the stator through a sliding bearing, there is less restriction on the outer cover. In addition, in order to reduce the weight of the motor, a side cover model is established with a thickness of 3.5mm and the material added is aluminum alloy. Although the electromagnetic force acts on the inner surface of the permanent magnet connected to the rotor, the side cover exhibits more severe vibration due to its low stiffness.

[0154] like Fig.15 As shown in , the modal analysis of the side cover is performed at the same time, and the modal vibration shape close to the frequency corresponding to the two acceleration peaks is obtained. Since the stiffness of the side cover is low, and the modal frequency of the side cover appears at a lower frequency, at low frequencies, such as Figure 4 As shown in the figure, the amplitude of the electromagnetic force is high and the harmonics of the force are rich. Therefore, when the frequency of this part of the electromagnetic force amplitude is close to the modal frequency of the side cover, resonance will occur.

[0155] S3.4. ERPMSM radiation noise prediction: Based on the vibration calculation results, the acoustic BEM method is used to predict the noise of ERPMSM. This noise prediction method is different from the traditional finite element prediction method. It requires the surface vibration of the outer rotor as the acoustic boundary condition and does not separate the space between the sound source and the field point as the finite element method. Therefore, BEM is more effective than the traditional finite element method. The acoustic mesh is usually a trihedron or a tetrahedron, such as Fig.16 shown.

[0156] Under no-load conditions, it is observed that the noise caused by electromagnetic force has obvious frequency characteristics, and its noise peak mainly appears at the frequency of uQfr, which is caused by Fig.13 Harmonic analysis of the electromagnetic force density can obtain higher electromagnetic force densities corresponding to spatial orders 2 and 8. The present invention focuses on the influence of low-frequency harmonic currents at u equal to 2 and 8 on electromagnetic noise when the motor speed is 3000r / min. In this case, the affected electromagnetic force frequencies are 350Hz and 1400Hz, so the present invention focuses on electromagnetic noise below 1500Hz. In the low-frequency band, due to the presence of abundant electromagnetic force harmonics, the noise spectrum under load is richer than that under no-load. However, in the high-frequency band, due to the small difference between the force amplitude and harmonics under the two working conditions, the noise spectra under no-load and load conditions are almost the same.

[0157] A hemispherical shape with a radius of 1.2 meters was established at the center of the motor as a model of the motor radiation sound field, and a reflecting surface was set below it to simulate the actual noise test situation, such as Fig.17 shown.

[0158] S4. Noise and cogging torque optimization.

[0159] S4.1. Noise optimization: From Figure 4 , Fig.18 It can be concluded that the frequency generated by the interaction between the permanent magnetic field and the stator slots is uQf r The electromagnetic force of λ has the most significant effect on the electromagnetic noise of ERPMSM. In addition, the present invention can be found in Table 1 that the amplitude of this part of the force is related to λ au The amplitude is proportional to the opening width b of the slot. Fig.19 In the paper, the electromagnetic force, the vibration of the center of the side cover and the noise spectrum under no-load conditions are compared.

[0160] like Fig.19 As shown in (c), in the initial design with b being 8 mm, the overall sound pressure level reaches 35.6 dBA. When b is reduced to 4 mm, the overall sound pressure level drops to 27.2 dBA, and the overall sound pressure level is weakened by 23.6%.

[0161] S4.2. Cogging torque optimization: Under no-load conditions, considering the influence of slotting effect on electromagnetic force, the torque of ERPMSM is optimized by changing the stator slot width. When the stator slot width is reduced from 8mm to 4mm, Fig.19 As shown in FIG. 1 , the cogging torque decreases as the stator slot width decreases, so the method of reducing the stator slot width adopted in the present invention is effective in weakening the motor cogging torque. Fig. 20 It can be seen that the peak value of the cogging torque is reduced from 32.4Nm to 18.6Nm, a decrease of 42.6%, indicating that reducing the slot width has a significant effect on reducing the cogging torque.

[0162] The present invention aims to explore the electromagnetic force spatial distribution and frequency characteristics of ERPMSM. Compared with the external stator ERPMSM, ERPMSM has similar spatial harmonic characteristics. Under no-load and load conditions, the electromagnetic force frequencies of ERPMSM are uQf r and 3kN ps f r The present invention proposes a method for predicting ERPMSM noise and establishes a model involving electromagnetic force, structural vibration and noise radiation. At the same time, the accuracy of the finite element model is verified by an outer rotor modal test to predict the noise level of the motor. The simulation and calculation results show that the electromagnetic force frequency caused by the interaction between the permanent magnet magnetic field and the stator slot is equal to uQf rThe present invention also proposes the relationship between electromagnetic force and slot width to reduce the noise level of ERPMSM. When the slot width is reduced from 8mm to 4mm, the sound pressure level drops by 5dBA, and the peak value of the motor's cogging torque is weakened by 42.6%. Therefore, it is an effective and practical method to reduce the noise of ERPMSM and improve the motor performance by optimizing the slot width of the stator.

[0163] Based on the same inventive concept, the present invention also proposes a noise prediction system for an outer rotor permanent magnet synchronous motor, comprising:

[0164] The acquisition module is used to obtain the structural parameters of the outer rotor permanent magnet synchronous motor ERPMSM.

[0165] The two-dimensional electromagnetic model building module is used to establish a two-dimensional electromagnetic model of the ERPMSM according to the structural parameters. The spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet are analyzed through the two-dimensional electromagnetic model to obtain the electromagnetic force acting on the surface of the permanent magnet.

[0166] The loading module is used to load the electromagnetic force into the three-dimensional structural model of the ERPMSM in the form of nodal force to obtain the nodal force acting on the surface of the permanent magnet in three-dimensional space.

[0167] The prediction module is used to calculate the radial vibration acceleration of the ERPMS M outer rotor based on the three-dimensional structural model and the node force by using the modal superposition method, and the radial vibration acceleration of the outer rotor is used as the acoustic boundary condition to predict the noise of the ERPMSM by using the acoustic boundary element method.

[0168] The present invention also proposes a computer device for predicting noise of an outer rotor permanent magnet synchronous motor, comprising: a memory, a processor and a computer program stored in the memory, and the processor implements the steps of the method for predicting noise of an outer rotor permanent magnet synchronous motor when executing the computer program.

[0169] The present invention also proposes a readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, they are used to execute the steps of the outer rotor permanent magnet synchronous motor noise prediction method.

[0170] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for predicting noise of an outer rotor permanent magnet synchronous motor, characterized in that: The following steps are involved: Obtain the structural parameters of the outer rotor permanent magnet synchronous motor ERPMSM; A two-dimensional electromagnetic model of ERPMSM is established according to the structural parameters. The spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet are analyzed through the two-dimensional electromagnetic model to obtain the electromagnetic force acting on the surface of the permanent magnet. The electromagnetic force is loaded into the three-dimensional structural model of ERPMSM in the form of nodal force to obtain the nodal force acting on the surface of the permanent magnet in three-dimensional space; Based on the three-dimensional structural model and nodal force, the modal superposition method is used to calculate the radial vibration acceleration of the outer rotor of the ERPMSM. The radial vibration acceleration of the outer rotor is used as the acoustic boundary condition, and the acoustic boundary element method is used to predict the noise of the ERPMSM.

2. The method for predicting noise of an outer rotor permanent magnet synchronous motor according to claim 1, characterized in that: The structural parameters of the ERPMSM include the fundamental magnetic field Bmn of the permanent magnet, the fundamental magnetic field Bmn of the armature reaction, av , speed n, pole pair number p, angular velocity ω r 、s v Rotation direction of the vth harmonic of the armature reaction field, rotor position and relative air gap permeance λ a .

3. The method for predicting noise of an outer rotor permanent magnet synchronous motor according to claim 1, characterized in that: According to the pole-slot matching of the fractional slots of the outer rotor permanent magnet synchronous motor, the number of double-layer winding layers, the tooth profile parameters of the stator slots, the air gap length, and the inner and outer radii of the rotor and stator, a three-dimensional structural model of the ERPMSM is established.

4. The method for predicting noise of an outer rotor permanent magnet synchronous motor according to claim 2, characterized in that: The two-dimensional electromagnetic model of the ERPMSM is established using Ansys Maxwell software, which specifically includes the following steps: In a permanent magnet synchronous motor, according to the relative air gap permeance λ a , the resultant radial flux density B in the air gap rs It is expressed as: B rs =(B r_mag +B r_arm )l a Among them, B r _mag and Br_arm represent the unslotted radial permanent magnet reaction field and the unslotted radial armature reaction field, respectively; Using the Maxwell stress tensor method, ignoring the tangential air gap flux density B ts , then the radial electromagnetic force density f r It is expressed as: Where u0 is the magnetic permeability of vacuum; For ERPMSM, when the outer rotor is fixed, the stator rotates at an angular velocity ω r Rotating in the opposite direction, in the z rotor coordinate, B r_mag , B r_arm and λ a It is expressed as: Among them, θ is the spatial mechanical angle, t is the time, p is the number of pole pairs, Q is the number of slots, and B mn is the fundamental magnetic field of the permanent magnet, B av is the fundamental armature magnetic field, λ0 is the average air gap permeability, λ a is the relative air gap permeability considering the stator slots, s v Indicates the rotation direction of the vth harmonic of the armature reaction field; For PMSMs with different slot-pole combinations and double-layer windings, (s v +v)p=3kN ps ; Among them, N ps =GCD(2p,Q) is the greatest common divisor of the number of poles and the number of slots, k is a positive integer; Then the radial electromagnetic force density is expressed as: Among them, λ au1 With λ au2 Represents the real part of the relative permeability when u takes different values.

5. The method for predicting noise of an outer rotor permanent magnet synchronous motor according to claim 4, characterized in that: The method also includes analyzing the established three-dimensional structural model of the ERPMSM by using an energy method to obtain the influencing factors affecting the cogging torque of the ERPMSM, which specifically includes the following steps: Based on the energy method, the cogging torque is defined as: Wherein, E is the magnetic field energy in the permanent magnet when no power is supplied, and α is the angle between the center line of the permanent magnet and the center line of a stator tooth; Since the stator winding is not energized, the magnetic field energy of the motor is expressed as: Among them, E Gap is the magnetic field energy in the air gap, E PM is the magnetic field energy generated by the permanent magnet, u0 is the vacuum magnetic permeability; B represents the air gap flux density of the fundamental wave permanent magnetic field, and V represents the air gap volume; The air gap magnetic flux distribution is expressed as: Where, hml(θ) is the magnetization length of the permanent magnet, σ(θ, α) is the effective length of the air gap, and Brd(θ) is the remanent magnetization distribution of the permanent magnet; The magnetic field energy of the motor is further expressed as: When the permanent magnet and stator are evenly distributed, B is expanded by Fourier decomposition. rd (θ) and the square of the relative air gap permeance, B 2 rd B after Fourier decomposition of (θ) rn and The Fourier decomposition of G n It is expressed as: Then the cogging torque T is obtained Cog (α), expressed as: Among them, l a is the armature core length, z represents the number of stator teeth, r2 is the rotor outer radius, r1 is the stator inner radius, and n is an integer that makes nz / 2p an integer. It can be obtained that the slot width, pole arc coefficient and speed factors all have an impact on the cogging torque of ERPMSM.

6. The method for predicting noise of an outer rotor permanent magnet synchronous motor according to claim 1, characterized in that: The radial vibration acceleration of the ERPMSM outer rotor is calculated based on the three-dimensional structural model and the node force using the modal superposition method, and the calculation process includes the following steps: According to the mass matrix [M] and stiffness matrix [K] of the ERPMSM structure, we can get [K][Φ i ]=λ i [M][Φ i ], where λ i is the i-th eigenvalue, [Φ i ] is the corresponding eigenvector. Calculate the modal mass of the ith mode [M i ] and modal stiffness [K i ]: [M i ]=[Φ i ] T [M][F] i ] [K i ]=[Φ i ] T [K][F] i ]=λ i [M i ] Calculate the modal participation coefficient Γ of the i-th mode i : For each mode i, according to the input vibration acceleration Calculate the modal response x i : Among them, ξ i is the damping ratio of the ith mode; The modal response x is obtained by numerical integration method. i , then the radial vibration acceleration of the outer rotor of the ERPMSM structure is the superposition of the acceleration responses of each modal, which can be expressed as: Among them, x i is the node displacement in modal coordinates, {Φ i } is the mode shape of mode order i, {F} is the time-varying electromagnetic nodal force, N is the modulus to be summed, [M], [C] and [K] are the mass matrix, damping matrix and stiffness matrix respectively, and T is the matrix transpose.

7. A noise prediction system for an outer rotor permanent magnet synchronous motor, characterized in that: include: An acquisition module is used to acquire structural parameters of an outer rotor permanent magnet synchronous motor ERPMSM; A two-dimensional electromagnetic model building module is used to establish a two-dimensional electromagnetic model of the ERPMSM according to the structural parameters, and to analyze the spatial distribution and frequency characteristics of the electromagnetic force acting on the surface of the permanent magnet through the two-dimensional electromagnetic model to obtain the electromagnetic force acting on the surface of the permanent magnet; A loading module is used to load the electromagnetic force into the three-dimensional structural model of the ERPMSM in the form of a nodal force to obtain the nodal force acting on the surface of the permanent magnet in three-dimensional space; The prediction module is used to calculate the radial vibration acceleration of the ERPMS M outer rotor based on the three-dimensional structural model and the node force by using the modal superposition method; the radial vibration acceleration of the outer rotor is used as the acoustic boundary condition, and the acoustic boundary element method is used to predict the noise of the ERPMSM.

8. A computer device for predicting noise of an outer rotor permanent magnet synchronous motor, characterized in that: include: A memory, a processor and a computer program stored in the memory, wherein when the processor executes the computer program, the steps of the outer rotor permanent magnet synchronous motor noise prediction method according to any one of claims 1 to 6 are implemented.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor, they are used to execute the steps of the outer rotor permanent magnet synchronous motor noise prediction method according to any one of claims 1 to 6.

Images