Outer rotor permanent magnet synchronous motor full rotation speed noise prediction method

Abstract

The invention discloses a full-rotation-speed noise prediction method for an outer rotor permanent magnet synchronous motor, and relates to the technical field of motor noise prediction. According to the method, through the full-chain coupling design of air gap magnetic flux density correction, electromagnetic force calculation, rotor vibration modeling, motion sound source parameter extraction, dynamic acoustic modeling with Doppler effect, signal verification and full-rotation-speed efficient optimization, the problems that a traditional model is insufficient in multi-physics field coupling and low in full-rotation-speed calculation efficiency are solved. The core of the method is to establish a correlation between magnetic flux density and electromagnetic force through Maxwell stress tensor, obtain rotor vibration displacement based on an oscillatory differential equation, combine a dynamic Green function correction boundary element method (BEM) to adapt to the Doppler effect of a motion sound source, and verify the accuracy of the model through UCMA + TVSFR signal processing. And finally, accurate and efficient prediction of the motor noise under the full-rotation-speed working condition is realized. The motor noise prediction precision is improved, the application range of the motor noise prediction is widened, and reliable theoretical support is provided for motor noise reduction design.

Description

Technical Field

[0001] This invention belongs to the field of motor noise prediction technology, and more specifically, relates to a method for predicting noise at full speed of an external rotor permanent magnet synchronous motor. Background Technology

[0002] External rotor permanent magnet synchronous motors (ERPMSMs) are widely used in new energy vehicles, wind turbines, and home appliances due to their compact structure and high torque density. The noise during motor operation is mainly caused by rotor vibration induced by electromagnetic excitation, which is then radiated through the air. In addition, the high-speed rotation of the rotor will produce the Doppler effect, resulting in distortion of the frequency and amplitude of the sound waves.

[0003] Existing methods for predicting motor noise have the following drawbacks: 1) Traditional magnetic flux density calculations do not consider the distortion effect of stator slot openings, resulting in large errors in electromagnetic force calculations; 2) The coupling between the vibration model and electromagnetic excitation is insufficient, and the transmission path of electromagnetic force to vibration displacement is not clearly defined; 3) Acoustic modeling often uses the static boundary element method (BEM), ignoring the Doppler effect caused by rotor motion, resulting in low noise prediction accuracy under high-speed conditions; 4) The signal separation effect is poor under multi-source interference, and there is a lack of effective means for model verification; 5) Repeated modeling is required under full-speed conditions, resulting in low computational efficiency and difficulty in meeting the needs of engineering applications.

[0004] Therefore, there is an urgent need for a noise prediction method that can take into account the Doppler effect, cover the entire speed range, and be efficient and accurate, so as to provide a reliable basis for motor structure optimization and noise reduction design. Summary of the Invention

[0005] 1. The problem to be solved To address the problems of low prediction accuracy, narrow applicable speed range, and insufficient computational efficiency of traditional motor noise prediction methods in the prior art, this invention provides a full-speed noise prediction method for external rotor permanent magnet synchronous motors.

[0006] 2. Technical Solution To solve the above problems, the technical solution adopted by the present invention is as follows: The present invention provides a method for predicting noise at full speed of an external rotor permanent magnet synchronous motor, comprising the following steps: Step 1, Air gap magnetic flux density correction: Obtain the ideal air gap radial magnetic flux density B r0 Tangential magnetic flux density B t0 and the real part of the complex permeability corresponding to the stator slot opening λ r virtual part λ i The actual air gap radial magnetic flux density is calculated using the magnetic flux density correction formula. Br_mag and tangential magnetic flux density Bt_mag ; Step 2, Electromagnetic force calculation: Based on Br_mag and Bt_mag The radial electromagnetic force was calculated using Maxwell's stress tensor method. F r and tangential electromagnetic force F t ; Step 3, Rotor vibration response modeling: [The text abruptly ends here, likely due to an incomplete sentence or a F r and F t Substituting this as the excitation into the rotor vibration differential equation, the tangential vibration displacement is obtained by solving. and radial vibration displacement ; Step 4, Modeling of moving sound source parameters: Based on and and rotor fundamental frequency speed Calculate the instantaneous position of the moving sound source Instantaneous velocity and Mach number M ; Step 5, Dynamic Green's Function Construction: Based on and M Construct a time-domain dynamic Green's function and a frequency-domain dynamic Green's function, wherein the frequency-domain dynamic Green's function is: ; Step 6: Acoustic modeling with Doppler effect: Substitute the frequency domain dynamic Green's function into the modified boundary element method integral equation to calculate the sound pressure at the observation point. The modified boundary element method integral equation is: Γ is the rotor surface boundary. Equivalent sound source intensity; Step 7, Multi-source signal processing verification: The synchronously acquired vibration and acoustic signals are processed by a uniform circular microphone array and time-varying spatial filtering rearrangement to correct the model parameters; wherein, the uniform circular microphone array (abbreviated as UCMA) is used to acquire acoustic signals, and the time-varying spatial filtering rearrangement (abbreviated as TVSFR) is used to extract target noise signals; Step 8: Efficient calculation and optimization at full speed: Parameterized dynamic Green's function, combined with fast filtering, outputs the noise distribution at full speed.

[0007] Furthermore, the magnetic flux density correction formula mentioned in step 1 is: Among them, the results are obtained through analytical methods combined with finite element simulation.

[0008] Furthermore, the formula for calculating the electromagnetic force in step 2 is as follows: ,in .

[0009] Furthermore, the differential equation for rotor vibration mentioned in step 3 is: The solution is obtained by using the finite element method or the modal superposition method.

[0010] Furthermore, the formula for the instantaneous position of the moving sound source in step 4 is: ,in R r This is the nominal radius of the rotor.

[0011] Furthermore, the time-domain dynamic Green's function mentioned in step 5 is: The frequency domain dynamic Green's function By analyzing the time-domain dynamic Green's function The Fourier transform is performed to obtain the result.

[0012] Furthermore, in step 7, the sampling rate for signal acquisition is 20kHz, vibration signals are acquired through an accelerometer, and acoustic signals are acquired through a microphone array.

[0013] Furthermore, the uniform circular microphone array described in step 7 has at least 6 array elements and an array radius of 0.4-0.6m. When acquiring signals, an acoustic calibrator is used to ensure the sensitivity is within ±0.2dB, and a photoelectric encoder is used to achieve synchronous acquisition of rotational speed (accuracy ±1rpm).

[0014] Furthermore, the signal processing step size for the time-varying spatial filtering rearrangement described in step 7 is 4-6 ms / step. This ensures that the sound pressure level prediction error remains <3.5 dB under transient acceleration conditions, and the zero-angle spatial filtering enhances the peak energy at the time center of the target sound source, improving the identification capability of weak fault sound sources (such as bearing wear or local demagnetization of permanent magnets).

[0015] Furthermore, in step 8, the frequency domain dynamic Green's function is parameterized to the rotor's fundamental frequency and rotational speed. function This avoids repeated modeling at different speeds.

[0016] 3. Beneficial effects Compared with the prior art, the beneficial effects of the present invention are as follows: (1) This invention provides a method for predicting noise at full speed in an external rotor permanent magnet synchronous motor. This method solves the problems of insufficient multi-physics coupling and low calculation efficiency at full speed in traditional models by using a full-chain coupling design of "air gap magnetic flux density correction → electromagnetic force calculation → rotor vibration modeling → extraction of moving sound source parameters → dynamic acoustic modeling with Doppler effect → signal verification → high-efficiency optimization at full speed". Its core is to establish the coupling relationship between magnetic flux density and electromagnetic force through Maxwell stress tensor, obtain rotor vibration displacement based on vibration differential equation, combine dynamic Green's function to correct boundary element method (BEM) to adapt to the Doppler effect of moving sound source, and then verify the accuracy of the model through UCMA+TVSFR signal processing, so as to achieve accurate and efficient prediction of motor noise at full speed. This invention improves the accuracy of vibration displacement calculation, the precision and applicability of motor noise prediction, and provides reliable theoretical support for motor noise reduction design.

[0017] (2) The present invention uses dynamic Green's function to correct BEM and considers the Doppler effect of rotor motion, which solves the problem of low noise prediction accuracy under high-speed conditions; (3) The present invention combines UCMA+TVSFR signal processing to realize model verification, which further optimizes the prediction accuracy.

[0018] (4) This invention achieves efficient calculation under all speed conditions by using parameterized Green's function and fast filtering, without the need for repeated modeling, thus improving the efficiency of engineering applications. Attached Figure Description

[0019] Figure 1 A schematic diagram of an external rotor permanent magnet synchronous motor, which is the research object of the motor noise prediction method of this invention. Figure 2 A schematic diagram of the ring model of the rotor of an external rotor permanent magnet synchronous motor, which is the research object of the motor noise prediction method of this invention. Figure 3 The research object of the motor noise prediction method of this invention is a two-dimensional electromagnetic model of an external rotor permanent magnet synchronous motor. Figure 4 This is a comparison chart of the theoretical magnetic force value and the finite element calculation value of the present invention; Figure 5 The magnetic flux density energy distribution diagram of an external rotor permanent magnet synchronous motor at different speeds is the research object of the motor noise prediction method of this invention. Figure 6 The research object of the motor noise prediction method of this invention is the magnetic flux density of an external rotor permanent magnet synchronous motor at different speeds. B r 2Distribution charts; (a) 250 r / min, (b) 500 r / min, (c) 750 r / min, (d) 1000 r / min; Figure 7 This is a schematic diagram of the noise test platform for the external rotor permanent magnet synchronous motor of the present invention; Figure 8 The acoustic boundary element model for the motor noise prediction method of this invention considers the Doppler effect; the elements marked in the figure, such as microphones 1-4, field points, and ground, are a schematic diagram of the modified dynamic boundary element method (DBEM) model, which is used to embed the Doppler effect to achieve distortion correction of moving sound sources.

[0020] Figure 9 Waterfall plot of multi-rotation vibration calculated using traditional static BEM; Figure 10 This is a multi-rotation vibration waterfall plot calculated by the dynamic BEM of the present invention; Figure 11 This is a multi-rotation vibration waterfall diagram after UCMA+TVSFR signal processing according to the present invention. Detailed Implementation

[0021] The more detailed description of embodiments of the invention below is not intended to limit the scope of the claimed invention, but is merely illustrative and does not limit the description of the features and characteristics of the invention, in order to suggest the best mode for carrying out the invention and to enable those skilled in the art to practice the invention. However, it should be understood that various modifications and variations can be made without departing from the scope of the invention as defined by the appended claims. The detailed description should be considered illustrative only and not restrictive, and any such modifications and variations shall fall within the scope of the invention described herein. Furthermore, the background art is intended to illustrate the current state of research and development and significance of the technology, and is not intended to limit the invention or the scope of application of this application.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0023] Each implementation method (including examples and comparative examples) takes a 24-pole, 48-slot ERPMSM as the research object, such as... Figure 1 As shown in Table 1, the core parameters of the experimental subject, ERPMSM, are shown in Table 1.

[0024] Table 1. Core Parameters of the Motor

[0025] This application uses experimental platform equipment such as... Figure 7 As shown, the data acquisition and control methods are as follows: Acoustic acquisition: 6-element uniform circular microphone array (UCMA, radius 0.5m, PCB 130E20 type), acoustic calibrator (accuracy ±0.1dB); Vibration acquisition: PCB 352C22 accelerometer, range ±500m / s², sensitivity 10mV / (m / s²); Rotational speed acquisition: photoelectric encoder, resolution 1024 lines / revolution, accuracy ±0.1 rpm; Data acquisition: NI cDAQ-9178 acquisition card, sampling rate 20kHz, synchronous trigger accuracy < 1μs; Load control: Magnetic powder brake (0-100Nm adjustable); Environmental control: semi-anechoic chamber (background noise <20dB (A), cutoff frequency 100Hz).

[0026] The motor noise prediction method of this application operates roughly as follows: (1) First, based on Maxwell's stress tensor, the expressions for radial and tangential electromagnetic forces are derived. The influence of the slot opening on the air gap magnetic flux density is corrected by the complex permeability (combined with analytical and finite element methods) to obtain the radial magnetic flux density under the slot: In the formula, the real part of the complex proportional permeability λ r virtual part λ i radial magnetic flux density Br_mag and tangential magnetic flux density Bt_mag ; (2) Secondly, establish a rotor ring equivalent model ( Figure 2 Define polar coordinate system (o-θ-z) , θ Tangential coordinates z (using radial coordinates), derive the equation for free vibration in the plane: In the formula, ρ The density of the toroid is... D=E h 3 / 12 For bending stiffness, K=E h For tensile stiffness, E For the equivalent elastic modulus, v , w The displacements are tangential and radial, respectively. The modal frequencies are solved using the separation of variables method and verified through a single-point excitation-multi-point vibration pickup experiment (force hammer excitation and accelerometer acquisition) to ensure that the modal error is ±5%.

[0027] (3) Furthermore, dynamic acoustic modeling for Doppler compensation is performed: 2.1) Modified Traditional Boundary Element Method (BEM): A dynamic Green's function is introduced to describe the propagation characteristics of the moving sound source. The position of the sound source on the rotor surface changes with time as follows: , e x 、e y Using coordinate basis vectors, derive the time-domain expression of the dynamic Green's function. , Instantaneous distance, τ For the time of dissemination, δ(t) (where Dirac function is used), and the frequency domain expression is obtained by combining the Lorentz transform and Fourier transform; 2.2) Constructing the modified BEM integral equation: Substituting the dynamic Green's function into the static BEM integral equation and adding the rotating sound source contribution term, we obtain: In the formula, For equivalent sound source intensity, This is a static distance used to achieve Doppler distortion correction.

[0028] (4) Further, through UCMA+TVSFR multi-source signal processing: 3.1) Construct a UCMA near-field acoustic observation system: Six array elements are evenly distributed around a circle with a radius of 0.5m. Combined with a PCB accelerometer and an NI synchronous acquisition system, acoustic and vibration signals are acquired synchronously (sampling rate ≥ 20kHz). 3.2) Time-varying spatial filter rearrangement (TVSFR): Based on the azimuth selectivity of zero-angle spatial filtering (the main lobe sound source gains gain and the side lobe sound source is suppressed), a spatiotemporal coupling characteristic space is constructed. By rearranging the time delay of multi-channel signals, the peak energy of the target sound source at the time center is enhanced, the strong source masking effect is suppressed, and the adaptive separation and Doppler correction of multi-source motion acoustic signals are realized.

[0029] (5) Finally, the full-speed high-efficiency calculation optimization is achieved by parameterized dynamic Green function correction, which covers the full speed range of 0-6000rpm (including steady state and transient state) at one time. Combined with TVSFR fast filtering algorithm, it avoids the repeated modeling of traditional static BEM, and the total time for multi-speed analysis is shortened to within 2 hours, with an efficiency improvement of 80%.

[0030] The present invention will be further described below with reference to specific embodiments.

[0031] Example 1 This embodiment provides a method for predicting noise at full speed of an external rotor permanent magnet synchronous motor, taking the aforementioned 24-pole 48-slot EPMSM as the research object. The method specifically includes the following steps: Step 1: Air gap flux density correction ① Establish a 2D transient electromagnetic finite element model using ANSYS Maxwell ( Figure 3 The rated speed is set to 600 rpm, and the base frequency current is 50 Hz.

[0032] ② Calculate the ideal magnetic flux density without slots: B r0 =0.92T, B t0 =0.35T (based on electromagnetic simulation results).

[0033] ③ Using analytical methods combined with finite element simulation, the complex permeability corresponding to the stator slot openings was obtained: λ r =0.93, λ i =0.07.

[0034] ④ Substitute into the magnetic flux density correction formula: Br_mag =0.92×0.93 + 0.35×0.07=0.88T Bt_mag =0.35×0.93 - 0.92×0.07=0.28T Step 2: Electromagnetic force calculation ① Substitute into Maxwell's stress tensor formula ( ): ②Verification: Compare the theoretical values ​​with the ANSYS Maxwell simulation values ​​( Figure 4 The radial electromagnetic force deviation was <3%, and the tangential electromagnetic force deviation was <5%, verifying the effectiveness of the electromagnetic force model.

[0035] Multi-speed verification: Set the speed to 250 rpm, 500 rpm, 750 rpm, and 1000 rpm respectively, and calculate the magnetic flux density energy distribution. Figure 5 )and distributed( Figure 6 The results showed that increasing the rotational speed caused the high-energy region to migrate towards the rotor edge. Maximum value increased by 12%-15%.

[0036] Step 3: Modeling the rotor vibration response ① Rotor structural parameters: average radius r =0.13m, bending stiffness D =1.8×10 5 N·m, tensile stiffnessK =2.5×10 6 N (calculated based on rotor size and material elastic modulus).

[0037] Will , Substituting this as an excitation term into the rotor vibration differential equation: ③ Modal verification: Single-point excitation-multi-point vibration pickup method is adopted ( Figure 7 A hammer (PCB 086C03) was used to excite 12 uniformly distributed points on the edge of the rotor. Eight accelerometers with a spacing of 45° were used to collect the response. The first to fifth modal frequencies were extracted by the frequency domain decomposition (FDD) method. The results were compared with the theoretical values ​​(as shown in Table 2). The error was less than 5%, which verified the accuracy of the vibration model.

[0038] ④ Solution results: Tangential vibration displacement radial vibration displacement .

[0039] Step 4: Modeling the parameters of the moving sound source ① Substitute into the formula for the instantaneous position of the moving sound source: ② High-speed operating condition (6000rpm, Below, to Differentiating gives ,Mach number M =86.5 / 343≈0.252.

[0040] Step 5: Construction of the dynamic Green's function ① Instantaneous distance between the observation point and the sound source (Observation point 1m from rotor center), speed of sound c =343m / s, wavenumber .

[0041] ② Time-domain dynamic Green's function: ③ Frequency domain dynamic Green's function (after Fourier transform): Step 6: Acoustic modeling with Doppler effect ① Experimental measurements show that the rotor surface boundary... Boundary normal sound pressure gradient equivalent sound source intensity .

[0042] ② Substitute into the modified BEM integral equation: ③ High-speed condition (6000rpm) results: With the engine speed set to 6000rpm (high-speed condition), the sound pressure level was calculated using dynamic BEM (after correction), as follows: Figure 8 As shown; the dynamic BEM high-frequency band (4kHz-6kHz) sound pressure level prediction error is 3.5dB, and the side-frequency component energy transfer trend is consistent with the experiment. Figure 10 The sound pressure level at the forward observation point is 8.2 dB higher than that of the static model, while it is 6.5 dB lower at the reverse observation point, which is consistent with the sound field law of the Doppler effect.

[0043] Step 7: UCMA+TVSFR signal processing verification ① Transient acceleration condition: 0-6000rpm, acceleration time 5s, multi-source acoustic signals (including rotor imbalance, electromagnetic force fluctuation, bearing vibration and other interference) are collected through UCMA.

[0044] ②TVSFR algorithm processing: High-frequency harmonics (>4kHz) of the original signal are masked by noise, and the time-frequency distribution is blurred (e.g.) Figure 9 As shown); after processing, the 4.2kHz and 5.8kHz harmonics are clearly distinguishable, the recognition accuracy is improved by 15dB, and the sound pressure level error is reduced to 3.2dB (as shown). Figure 11 ).

[0045] ③ Spatial directivity test: 8 observation points (spaced 45° apart) at a distance of 1m from the motor, the directivity deviation of this invention is < 5%.

[0046] Step 8: Efficient Calculation Optimization at Full Rotational Speed ① Dynamic Green's function in the frequency domain Parameterization function It covers a speed range of 0-6000rpm (in 300rpm increments, for a total of 21 operating conditions).

[0047] ② Computational efficiency: The modeling and solution of 21 working conditions were completed in one go, with a total time of 1.8 hours; the transient acceleration working condition (5ms / step, 1000 steps in total) took 1.2 hours, and the error was still < 3.5dB.

[0048] Table 2. Comparison of frequencies at different orders and error analysis Modal order Theoretical frequency (Hz) Experimental frequency (Hz) error(%) 1st order 128 132 3.1 2nd order 256 262 2.3 3rd order 384 395 2.9 4th order 512 525 2.5 5th order 640 658 2.8 Comparative Example 1

[0049] This comparative example also provides a click noise prediction method, which adopts traditional static BEM multi-speed analysis. The specific motor parameters and experimental platform are the same as those in Example 1, with a speed range of 0-6000 rpm (step size 300 rpm, a total of 21 operating conditions).

[0050] Traditional static BEM modeling is used, ignoring rotor motion and Doppler effect, and each working condition is modeled and solved separately.

[0051] The experimental results are as follows: Calculation time: Modeling each working condition takes about 30 minutes, and the total time for 21 working conditions is 10.5 hours.

[0052] Prediction accuracy: The prediction error for sound pressure level in the high-frequency band (4kHz-6kHz) is 12.7dB, with significant spectral broadening (e.g. Figure 9 (As shown).

[0053] Spatial directivity: Directivity deviation is 37%-42%, which cannot accurately reflect the distribution pattern of the sound field.

[0054] Comparative Example 2

[0055] This comparative example also provides a click noise prediction method, which uses the traditional method for transient acceleration condition analysis. The specific motor parameters and experimental platform are the same as in Example 1. The transient acceleration condition is 0-6000rpm, acceleration time is 5s, 5ms / step, and a total of 1000 steps.

[0056] The traditional static BEM combined with piecewise interpolation method is used to calculate the sound pressure field separately for each step.

[0057] The experimental results are as follows: Calculation time: The total time was 6.2 hours, which is 5.2 times that of Embodiment 1 of the present invention.

[0058] Prediction accuracy: High-frequency error >10dB, severe time-domain signal distortion, unable to accurately capture the spectral shift caused by the Doppler effect.

[0059] Engineering applicability: It takes too long and cannot meet the needs of rapid optimization design in actual engineering projects.

[0060] Table 3. Performance Comparison Table of Embodiments and Comparative Examples of this Application Comparison indicators Example 1 (Invention) Comparative Example 1 (Traditional Static BEM) Comparative Example 2 (Traditional Transient Method) Total time consumed at multiple speeds (21 operating conditions) 1.8 hours 10.5 hours / Transient acceleration time (1000 steps) 1.2 hours / 6.2 hours High-frequency sound pressure level error (4-6kHz) <3.5dB 12.7dB >10dB Spatial orientation deviation <5% 37%-42% >35% Doppler effect adaptability Fully compatible Incompatible Partially compatible (severe distortion) In summary, to make the technical effects of the embodiments of the present invention clearer compared with the prior art, the above technical effect data are summarized in Table 3, as shown above. The performance comparison in Table 3 clearly shows the core advantages of the embodiments of the present invention: in terms of efficiency, the multi-speed calculation time is only 1 / 6 of that of the traditional static BEM, and the transient acceleration time is less than 1 / 5 of the traditional method, significantly reducing the calculation cost under all operating conditions; in terms of accuracy, the high-frequency sound pressure error is < 3.5dB and the spatial directivity deviation is < 5%, far superior to traditional solutions; at the same time, it fully adapts to the Doppler effect, solving the distortion problem of traditional models under high-speed conditions, and can accurately support the engineering requirements of full-speed noise prediction for external rotor permanent magnet synchronous motors. It is evident that the embodiments of the present invention have outstanding substantive features and significant progress.

[0061] The present invention has been described in detail above with reference to specific exemplary embodiments. However, it should be understood that various modifications and variations can be made without departing from the scope of the invention as defined by the appended claims. The detailed description should be considered illustrative only and not restrictive, and any such modifications and variations shall fall within the scope of the invention described herein. Furthermore, the background art is intended to illustrate the current state of development and significance of the technology and is not intended to limit the present invention or the scope of application of the present application.

[0062] More specifically, although exemplary embodiments of the invention have been described herein, the invention is not limited to these embodiments, but includes any and all embodiments modified, omitted, such as combinations between various embodiments, adaptive changes, and / or substitutions, as would be apparent to those skilled in the art from the foregoing detailed description. The limitations in the claims are to be interpreted broadly as used in the language of the claims and are not limited to the examples described in the foregoing detailed description or during the implementation of this application, which should be considered non-exclusive. Any step listed in any method or process claim may be performed in any order and is not limited to the order set forth in the claims. Therefore, the scope of the invention should be determined solely by the appended claims and their legal equivalents, and not by the description and examples given above.

[0063] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. In case of conflict, the definitions in this specification shall prevail. When flow rate, power, refractive index, time, or other values ​​or parameters are expressed as ranges, preferred ranges, or a series of upper and lower preferred values, this should be understood as specifically disclosing all ranges formed by any pair of any upper or preferred value with any lower or preferred value, regardless of whether such range is disclosed individually. For example, the range 1-50 should be understood to include any number, combination of numbers, or subrange selected from 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50, as well as all decimal values ​​between the integers mentioned above, such as 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, and 1.9. Regarding subranges, specifically consider "nested subranges" extending from any endpoint of the range. For example, nested sub-ranges of the exemplary range 1-50 may include 1-10, 1-20, 1-30 and 1-40 in one direction, or 50-40, 50-30, 50-20 and 50-10 in another direction.

Claims

1. A method for predicting noise at full speed in an external rotor permanent magnet synchronous motor, characterized in that, Includes the following steps: Step 1, Air gap magnetic flux density correction: Obtain the ideal air gap radial magnetic flux density B r0 Tangential magnetic flux density B t0 and the real part of the complex permeability corresponding to the stator slot opening λ r virtual part λ i The actual air gap radial magnetic flux density is calculated using the magnetic flux density correction formula. Br_mag and tangential magnetic flux density Bt_mag ; Step 2, Electromagnetic Force Calculation: Based on Br_mag and Bt_mag The radial electromagnetic force was calculated using Maxwell's stress tensor method. F r and tangential electromagnetic force F t ; Step 3, Rotor vibration response modeling: [The text abruptly ends here, likely due to an incomplete sentence or a F r and F t Substituting this as the excitation into the rotor vibration differential equation, the tangential vibration displacement is obtained by solving. and radial vibration displacement ; Step 4, Modeling of moving sound source parameters: Based on and and rotor fundamental frequency speed Calculate the instantaneous position of the moving sound source Instantaneous velocity and Mach number M ; Step 5, Dynamic Green's Function Construction: Based on and M Construct a time-domain dynamic Green's function and a frequency-domain dynamic Green's function, wherein the frequency-domain dynamic Green's function is: ; Step 6: Acoustic modeling with Doppler effect: Substitute the frequency domain dynamic Green's function into the modified boundary element method integral equation to calculate the sound pressure at the observation point. The modified boundary element method integral equation is: Γ is the rotor surface boundary. Equivalent sound source intensity; Step 7, Multi-source signal processing verification: The synchronously acquired vibration and acoustic signals are processed by a uniform circular microphone array and time-varying spatial filtering rearrangement to correct the model parameters; wherein, the uniform circular microphone array is used to acquire acoustic signals, and the time-varying spatial filtering rearrangement is used to extract target noise signals; Step 8: Efficient calculation and optimization at full speed: Parameterized dynamic Green's function, combined with fast filtering, outputs the noise distribution at full speed.

2. The method according to claim 1, characterized in that, The magnetic flux density correction formula mentioned in step 1 is: Among them, the results are obtained through analytical methods combined with finite element simulation.

3. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, The formula for calculating electromagnetic force in step 2 is as follows: ,in .

4. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, The differential equation for rotor vibration mentioned in step 3 is: The solution is obtained by using the finite element method or the modal superposition method.

5. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, The formula for the instantaneous position of the moving sound source in step 4 is: ,in R r This is the nominal radius of the rotor.

6. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, The time-domain dynamic Green's function mentioned in step 5 is: The frequency domain dynamic Green's function By analyzing the time-domain dynamic Green's function The Fourier transform is performed to obtain the result.

7. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, In step 7, the sampling rate for signal acquisition is 20kHz. Vibration signals are acquired through an accelerometer, and acoustic signals are acquired through a microphone array.

8. A method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1 or 7, characterized in that, The uniform circular microphone array described in step 7 has at least 6 array elements and an array radius of 0.4-0.6m. When acquiring signals, an acoustic calibrator is used to ensure that the sensitivity is within ±0.2dB, and a photoelectric encoder is used to achieve synchronous acquisition of rotational speed.

9. A method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1 or 7, characterized in that, The signal processing step size for the time-varying spatial filtering rearrangement described in step 7 is 4-6 ms / step.

10. The method for predicting noise at full speed of an external rotor permanent magnet synchronous motor according to claim 1, characterized in that, In step 8, the frequency domain dynamic Green's function is parameterized to the rotor's fundamental frequency and rotational speed. function .

Images