A method for controlling rotor vibration of a full-speed magnetically suspended high-speed motor

Abstract

The invention discloses a full-speed magnetic suspension high-speed motor rotor vibration control method. The system lag phase at full speed is determined by a frequency sweep method, and then the lag phase at full speed changes with the speed curve is obtained; secondly, filtering is performed by an LMS algorithm, and then the phase shift compensation is performed according to the lag phase curve, and the obtained signal phase is used as the unbalanced vibration control signal phase; then the rotor system control signal amplitude at different speeds is obtained by an influence coefficient method and fitted, and the full-speed control signal amplitude changes with the speed curve is obtained; then, unbalance compensation is performed in the process of rotor speed increase and decrease according to the phase curve and the amplitude curve, and the motor rotor vibration is actively controlled. It can be applied to vibration suppression before the critical speed of the rotor, and the output control voltage signal amplitude and phase fitting curve can control the vibration of the rotor system in the full speed range.

Description

Technical Field

[0001] The present invention relates to the field of active magnetic suspension control, and in particular to a method for active vibration control of a magnetic suspension high-speed motor rotor system, which realizes stable control of rotor radial vibration in active control of a full-speed magnetic bearing-rotor system before over-criticality. Background Art

[0002] Active magnetic bearings (AMBs for short) are a new type of non-contact support method with the advantages of no friction, high rotor speed, and active controllable rotor dynamic characteristics. They have been widely used in precision instruments and equipment such as inertial momentum wheels, control torque gyroscopes, and high power density motors. However, with the development trend of high power and high power density, the rotor speed of magnetic bearings continues to increase. Since the unbalanced mass of the rotor cannot be completely eliminated during machining, the unbalanced force caused by the unbalanced mass of the rotor surges, and the rotor vibration problem is particularly prominent. Strong vibration levels not only affect the operation of the unit under rated conditions and the working environment, but also cause major accidents and bring huge economic losses.

[0003] At present, the research on active vibration control methods of AMB rotors is mainly divided into the following two points: the first is the minimum rotor displacement control method, the idea is to output the control voltage signal through the controller and then pass through the power amplifier to become the control current acting on the electromagnetic bearing to generate electromagnetic force, which is equal to the unbalanced force of the rotor itself and opposite in direction, so that the unbalanced force on the rotor is offset, so that it rotates around the geometric center of the stator, and the vibration displacement is minimized. The second is the minimum electromagnetic force control method of the electromagnetic bearing, the idea is to filter out the control current signal with the same frequency as the rotor speed through the filter, at this time the rotor is only affected by its own unbalanced force and rotates around its mass center, because the control current with the same frequency of the magnetic bearing is filtered out, the electromagnetic force generated by the electromagnetic bearing is minimized.

[0004] The adaptive filter using the LMS algorithm is used to extract the phase of the rotor's co-frequency vibration displacement signal in the rotor system imbalance compensation. However, there is a lag phase between the rotor's co-frequency vibration displacement signal and the actual unbalanced force on the rotor, as well as between each control element, which changes with the speed. CN202110102615.5 Although the design of the variable phase adaptive LMS notch filter can achieve system stability within the full speed range, it only selects the same phase offset angle in certain speed sections by analyzing the root locus diagram, and does not obtain the phase offset angle characteristics; and does not obtain information related to the control amplitude. The frequency of the control voltage is inconsistent with the rotor speed frequency, so the temperature vibration control effect cannot be obtained. Summary of the invention

[0005] The technical problem to be solved by the present invention is to propose a full-speed high-power density magnetic suspension high-speed motor rotor vibration control method, which solves the problem of stable control of radial vibration of the full-speed magnetic bearing rotor system before over-criticality.

[0006] Furthermore, the present invention is used to solve the problem of active vibration control of the rotor system of the magnetic levitation motor at full speed and the problem of system stability.

[0007] The present invention adopts the following technical solutions to solve the above technical problems:

[0008] A full-speed magnetic suspension high-speed motor rotor vibration control method, characterized by comprising the following steps:

[0009] (1) When the rotor is statically suspended, the controller outputs a full-speed variable-frequency sinusoidal control signal to act on the electromagnetic bearing to obtain the rotor's co-frequency vibration displacement signal and solve the phase difference between the sinusoidal control signal and the rotor's co-frequency vibration displacement signal. Required phase compensation angle According to different frequency values, the curve of phase compensation angle changing with speed in each speed range is obtained, that is, the full speed compensation phase curve;

[0010] (2) applying the phase compensation angle obtained in step (1) to the LMS algorithm, filtering through the LMS algorithm, obtaining the rotor same-frequency harmonic signal component and outputting the rotor same-frequency harmonic signal component as a control voltage phase signal after leading compensation by an angle of θ; the control voltage phase signal is negatively fed back to the rotor system, thereby realizing full-speed control signal phase extraction at different speed frequencies;

[0011] (3) Multiple fixed speeds ω in the full speed range 1 ~ω n The full speed control signal phase obtained in step (2) is combined with the influence coefficient method to measure the control voltage signal amplitude u 1 ~u n , the control voltage signal amplitude and the speed are fitted to obtain the full speed control signal amplitude curve;

[0012] (4) According to the full-speed compensation phase curve and amplitude curve of the control signal in (1) and (3), imbalance compensation is performed during the rotor speed increase and decrease process to actively control the motor rotor vibration.

[0013] In the above technical solution, the phase compensation angle θ in step (1) is determined by the phase lag of the power amplifier, the sensor, and the rotor unbalance force. The frequency domain phase angles of the power amplifier and the magnetic bearing transfer function are respectively and The frequency domain phase angle of the controller transfer function is: The frequency domain phase angle of the magnetic suspension rotor transfer function is: The frequency domain phase angle of the sensor transfer function is The phase difference between the electromagnetic force of the magnetic bearing and the unbalanced force of the rotor is 180°, so The phase difference between the sinusoidal control signal and the rotor's same-frequency vibration displacement signal is

[0014] In the above technical solution, step (1) outputs a variable frequency sinusoidal control signal through the controller Simulate the unbalanced force on the rotor and measure the rotor vibration displacement Where f is the linearly changing speed frequency value, is the phase of the control signal, A 1 is the amplitude of the control signal, is the phase of the displacement signal, A 2 is the amplitude of the displacement signal, then the voltage compensation phase angle

[0015] In the above technical solution, step (2) determines the compensation phase at multiple speeds based on the full-speed compensation phase curve of step (1), and obtains the control signal phase at multiple speeds in combination with the LMS algorithm.

[0016] In the above technical solution, the LMS algorithm in step (2) is expressed as follows:

[0017] y(t)=w 1 (t)sin(ωt)+w 2 (t)cos(ωt);

[0018] e(t)=d(t)-y(t);

[0019] w 1 (t+1)=w 1 (t)+ue(t)sin(ωt);

[0020] w 2 (t+1)=w 2 (t)+ue(t)cos(ωt);

[0021] Where y(t) represents the output signal of the notch extraction, d(t) represents the input signal required to realize the notch function, and w 1 (t) and w 2 (t) represents the weight coefficient of the sine and cosine components of the notch frequency, u represents the iteration step of the algorithm, and represents the notch frequency of the multi-frequency signal;

[0022] In the full speed range, when the rotor speed ω rWhen the speed changes, the corresponding ω of the LMS algorithm also changes. First, the rotor co-frequency vibration displacement phase is extracted, and then the phase angle θ is compensated based on the co-frequency vibration displacement phase as the speed changes. The sine and cosine values ​​of the phase compensation angle θ and the notch frequency weight coefficient w are respectively 1 (t) and w 2 (t) The weight coefficient after phase compensation is calculated as: w' 1 (t) = w 2 (t)sinθ-w 1 (t)cosθ, w' 2 (t) = w 2 (t)cosθ+w 1 (t) sinθ, the output signal of the phase-shifted LMS notch method is y′(t)=w′ 1 (t)sin(ωt)+w' 2 (t)cos(ωt).

[0023] In the above technical solution, step (3) obtains the amplitude of the control signal at multiple speeds by using the influence coefficient method according to the output signal y′(t) at multiple speeds obtained in step (2).

[0024] In the above technical solution, the influence coefficient method of step (3) is carried out as follows: a number of correction planes and a number of measuring planes are selected on the rotor for multiple operation corrections, and the vibration of a certain measuring plane caused by a unit correction voltage signal on a certain correction plane at a certain speed is taken as an influence coefficient; these influence coefficients are obtained by measurement or calculation, and according to the vibration caused by the unbalance, the size and direction of the control signal that should be added to each correction plane in order to limit the vibration of each measuring plane to below a certain amplitude are determined.

[0025] In the above technical solution, step (3) includes step (3.1): the influence coefficient method for a four-degree-of-freedom rigid rotor:

[0026] The front radial magnetic bearing and the rear radial magnetic bearing correspond to the two correction planes P on the rotor respectively. b1 and P b2 The front radial magnetic bearing sensor and the rear radial magnetic bearing sensor correspond to the rotor measurement surface P s1 and P s2 ; Measured P s1 and P s2 The initial vibration displacement vectors of the two measuring surfaces are: and

[0027] Assume that the plane P where the left and right magnetic bearings are located b1 and P b2 The plane P where the sensor is located is the rotor imbalance correction surface. s1and P s2 It is the unbalance detection surface;

[0028] At any equilibrium speed ω, on the correction surface P b1 Test and add control voltage signal There is no hysteresis angle in its voltage signal, and the vibration displacement vector at each detection surface is obtained as follows: and Find the correction surface P b1 Test the control voltage Rear detection surface P s1 and P s2 Influence coefficient of displacement:

[0029]

[0030]

[0031] Remove the try signal On the correction surface P b2 Test and add control voltage signal The vibration displacement vector at each detection surface is obtained as and Find the correction surface P b2 Test the control voltage Rear detection surface P s1 ,P s2 Influence coefficient of displacement:

[0032]

[0033]

[0034] Get the correction surface P b1 and P b2 The voltage signals to be added are:

[0035]

[0036] In the above technical solution, step (3) also includes step (3.2): according to the control voltage amplitude U measured at multiple speeds ω in step (3.1), 1 ,U 2 The least square method is used to fit the control voltage and the speed change curve; the least square method is one of the fourth-order polynomial fitting method, exponential function fitting, logarithmic function fitting, and power function fitting.

[0037] In the above technical solution, a quartic polynomial fitting method is used in step (3.2).

[0038] In the above technical solution, the specific process of applying step (4) to the vibration control of the magnetic suspension high-speed motor rotor is as follows: establish a control model of the electromagnetic bearing-flexible rotor system, introduce the signal phase compensation extracted by the LMS algorithm into the control model, and use the sensor G s The measured displacement signal X(s) is negatively fed back to the PID controller G c LMS notch filter with phase compensation G i The output amplitude of the phase compensation LMS notch filter is the control signal amplitude, which is then output to the power amplifier G a The output current of the power amplifier acts on the electromagnetic bearing to achieve active vibration control of the rotor.

[0039] In summary, the present invention determines the system lag phase at full speed through a frequency sweeping method; obtains the curve of the lag phase changing with the speed at full speed; secondly, filters through the LMS algorithm, and then compensates for the phase shift according to the lag phase curve, and uses the obtained signal phase as the unbalanced vibration control signal phase; then obtains the amplitude of the rotor system control signal at different speeds through the influence coefficient method and fits it, and obtains the curve of the amplitude of the full speed control signal changing with the speed; then analyzes the stability and effectiveness of the vibration control method at full speed, performs imbalance compensation during the rotor speed increase and decrease process according to the phase curve and the amplitude curve, and actively controls the vibration of the motor rotor.

[0040] Compared with the prior art, the full-speed vibration control method of the present invention can be applied to vibration suppression before the critical speed of the rotor, and the output control voltage signal amplitude and phase fitting curve can control the vibration of the rotor system in the full speed range.

[0041] The present invention outputs a control voltage signal corresponding to the real-time speed. Therefore, compared with the traditional rotor unbalance vibration suppression under constant speed, the rotor vibration of the magnetic levitation high-speed motor can be controlled online to a smaller level through fitting of the control signal phase curve and the control signal amplitude. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0043] Figure 1 The figure is a system flow chart of the method of the present invention.

[0044] Figure 2 The control principle diagram of the electromagnetic bearing-rotor system of the present invention;

[0045] Figure 3 Schematic diagram of compensation phase determination for frequency sweeping method;

[0046] Figure 4 This is a schematic diagram of rotor online unbalance compensation;

[0047] Figure 5 This is the principle block diagram of the influence coefficient method;

[0048] Figure 6 is a system compensation phase curve in the embodiment;

[0049] Figure 7 It is a fitting diagram of the amplitude of the system control signal in the embodiment;

[0050] Figure 8 It is a waterfall diagram of harmonic changes in the embodiment. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] The present invention proposes a rotor vibration control method for a full-speed high-power density magnetic suspension high-speed motor, such as Figure 1 As shown, the control principle diagram is as follows Figure 2 The specific steps include:

[0053] (1) Sweep frequency method Figure 3 As shown, when the rotor is statically suspended, the controller output The sinusoidal control voltage signal is a sine control voltage signal, where f is the frequency of the sweep signal, which increases linearly in the full speed range. The displacement sensor measures the rotor vibration displacement signal In order to ensure that the control electromagnetic force is in the opposite direction to the unbalanced force on the rotor, the control signal compensation phase corresponding to each frequency f is The full speed compensation phase changes with the speed frequency curve. The schematic diagram of rotor online unbalance compensation is as follows: Figure 4 shown.

[0054] (2) The phase compensation angle θ obtained in step (1) is applied to the LMS algorithm. The standard LMS algorithm is expressed as follows:

[0055] y(t)=w 1 (t)sin(ωt)+w 2 (t)cos(ωt),

[0056] e(t)=d(t)-y(t),

[0057] w 1 (t+1)=w 1 (t)+ue(t)sin(ωt),

[0058] w 2 (t+1)=w2 (t)+ue(t)cos(ωt).

[0059] Where y(t) represents the output signal of the notch extraction, d(t) represents the input signal required to realize the notch function, and w 1 (t) and w 2 (t) represent the weight coefficients of the sine and cosine components of the notch frequency respectively, u represents the iterative step size of the algorithm, and represents the notch frequency of the multi-frequency signal.

[0060] In the full speed range, when the rotor speed ω r When the LMS algorithm changes, the corresponding ω also changes accordingly. First, the rotor vibration displacement phase is extracted, and then the phase angle θ is compensated based on the vibration displacement phase with the speed change. The notch frequency sine and cosine components are used for phase compensation, w' 1 (t) = w 2 (t)sinθ-w 1 (t)cosθ,w' 2 (t) = w 2 (t)cosθ+w 1 (t) sinθ phase shift LMS notch extraction output signal y′(t) = w′ 1 (t)sin(ωt)+w' 2 (t)cos(ωt).

[0061] (3.1) According to the output signal y′(t) at multiple speeds obtained in step (2), the control signal amplitude at multiple speeds is obtained by the influence coefficient method.

[0062] Since the magnetic suspension rotor system is a multi-input and multi-output system, it is difficult to determine the compensation voltage amplitude through the transfer function, but it can be determined through the influence coefficient method, that is, select several correction planes and several measurement planes on the rotor for multiple operation corrections. The vibration of a certain measurement surface caused by a unit correction voltage signal on a correction plane at a certain speed is an influence coefficient. By measuring or calculating these influence coefficients, the size and direction of the control signal that should be added to each correction plane to limit the vibration of each measurement surface to below a certain amplitude can be determined based on the vibration caused by the unbalance.

[0063] Influence coefficient method Figure 5 The front radial magnetic bearing and the rear radial magnetic bearing of the four-degree-of-freedom rigid rotor system correspond to the two correction planes P on the rotor respectively. b1 and P b2 The front radial magnetic bearing sensor and the rear radial magnetic bearing sensor correspond to the rotor measurement surface P s1 and P s2 . Measured P s1 and Ps2 The initial vibration displacement vectors of the two measuring surfaces are: and

[0064] Assume that the plane P where the left and right magnetic bearings are located b1 and P b2 The plane P where the sensor is located is the rotor imbalance correction surface. s1 and P s2 is the unbalance detection surface. At any balancing speed ω, on the correction surface P b1 Test and add control voltage signal There is no hysteresis angle in its voltage signal. The vibration displacement vector at each detection surface is obtained as follows: and Find the correction surface P b1 Test the control voltage Rear detection surface P s1 and P s2 Displacement influence coefficient:

[0065]

[0066]

[0067] Remove the try signal On the correction surface P b2 Test and add control voltage signal The vibration displacement vector at each detection surface is obtained as and Find the correction surface P b2 Test the control voltage Rear detection surface P s1 ,P s2 Displacement influence coefficient:

[0068]

[0069]

[0070] The correction surface P can be obtained b1 and P b2 The voltage signals to be applied are

[0071]

[0072] (3.2) According to the control voltage amplitude U measured at multiple speeds ω in (3.1), 1 ,U 2 The least squares fitting method is used to obtain the curve of control voltage changing with speed. Exponential function fitting, logarithmic function fitting, power function fitting can be obtained. The residual error of quartic function fitting is small, and the calculation amount meets the conditions.

[0073] (4) According to the curves of voltage phase and voltage amplitude controlled by speed in (3.1) and (3.2), the rotor speed variation process is compensated for imbalance and the motor rotor vibration is actively controlled.

[0074] According to steps (1) to (4), an active control method is designed for the magnetic suspension motor rotor system. The parameters of the magnetic bearing rotor system are shown in Table 1:

[0075] Table 1

[0076] Serial number parameter Numeric 1 Distance between the centers of radial magnetic bearings at both ends (m) 0.697 2 Distance between radial magnetic bearing at end a and center of mass (m) 0.340 3 Distance between radial magnetic bearing at end b and center of mass (m) 0.357 4 Rotor mass (kg) 18.14 5 <![CDATA[Rotor cross-sectional moment of inertia (kg·m 2 )]]> 0.527 6 <![CDATA[Rotor pole moment of inertia (kg·m 2 )]]> 0.016 7 Displacement stiffness coefficient of magnetic bearing (N / m) <![CDATA[-1.11×10 6 ]]> 8 Magnetic bearing current stiffness coefficient (N / A) 213.5 9 PID controller proportional coefficient 2 10 PID controller integral coefficient 36 11 PID controller differential coefficient 0.003 12 Sensor gain (v / m) <![CDATA[5×10 3 ]]>

[0077] Through the active control method, the system compensation phase curve can be obtained as follows: Figure 6 As shown, the control signal amplitude fitting curve is as follows Figure 7 shown. Figure 8 The harmonic change waterfall diagram can clearly show the difference before and after control. The three coordinate axes represent the rotor speed frequency, vibration harmonic frequency and vibration displacement amplitude. It can be seen that as the rotor speed increases, the displacement amplitudes of the first and second harmonics of the vibration are more obvious. For the vibration displacement amplitude of the first harmonic, the maximum same-frequency vibration displacement of the rotor before compensation is 27.63μm, and the maximum axis trajectory displacement of the rotor after compensation is 8.9μm, which is 67.78% less than the previous control. The minimum same-frequency vibration displacement of the rotor before compensation is 5.92μm, and the minimum axis trajectory displacement of the rotor after compensation is 1.9μm, which is 67.91% less than the previous control. The same-frequency vibration suppression effect is good.

[0078] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A full-speed magnetic suspension high-speed motor rotor vibration control method, characterized in that The following steps are involved: (1) When the rotor is statically suspended, the controller outputs a full-speed variable-frequency sinusoidal control signal to act on the electromagnetic bearing to obtain the rotor's co-frequency vibration displacement signal and solve the phase difference between the sinusoidal control signal and the rotor's co-frequency vibration displacement signal. Required phase compensation angle According to different frequency values, the curve of phase compensation angle changing with speed in each speed range is obtained, that is, the full speed compensation phase curve; (2) applying the phase compensation angle obtained in step (1) to the LMS algorithm, filtering through the LMS algorithm, obtaining the rotor same-frequency harmonic signal component and outputting the rotor same-frequency harmonic signal component as a control voltage phase signal after leading compensation by an angle of θ; the control voltage phase signal is negatively fed back to the rotor system, thereby realizing full-speed control signal phase extraction at different speed frequencies; (3) Multiple fixed speeds ω1~ω in the full speed range n The full speed control signal phase obtained in step (2) is combined with the influence coefficient method to measure the control voltage signal amplitude u1~u n , the control voltage signal amplitude and the speed are fitted to obtain the full speed control signal amplitude curve; (4) According to the full-speed compensation phase curve and amplitude curve of the control signal in (1) and (3), imbalance compensation is performed during the rotor speed increase and decrease process to actively control the motor rotor vibration.

2. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1, characterized in that Step (1) The phase compensation angle θ is determined by the phase lag of the power amplifier, sensor, and rotor unbalance force. The frequency domain phase angles of the power amplifier and magnetic bearing transfer function are: and The frequency domain phase angle of the controller transfer function is: The frequency domain phase angle of the magnetic suspension rotor transfer function is: The frequency domain phase angle of the sensor transfer function is The phase difference between the electromagnetic force of the magnetic bearing and the unbalanced force of the rotor is 180°, so The phase difference between the sinusoidal control signal and the rotor's same-frequency vibration displacement signal is 3. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1, characterized in that Step (1) Outputting a variable frequency sinusoidal control signal through the controller Simulate the unbalanced force on the rotor and measure the rotor vibration displacement Where f is the linearly changing speed frequency value, is the phase of the control signal, A1 is the amplitude of the control signal, is the phase of the displacement signal, A2 is the amplitude of the displacement signal, then the voltage compensation phase angle 4. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1, characterized in that Step (2) determines the compensation phase at multiple speeds based on the full speed compensation phase curve of step (1), and obtains the control signal phase at multiple speeds by combining the LMS algorithm.

5. The full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1 is characterized in that The LMS algorithm in step (2) is expressed as follows: y(t)=w1(t)sin(ωt)+w2(t)cos(ωt); e(t)=d(t)-y(t); w1(t+1)=w1(t)+ue(t)sin(ωt); w2(t+1)=w2(t)+ue(t)cos(ωt); Where y(t) represents the output signal of the notch extraction, d(t) represents the input signal required to realize the notch function, w1(t) and w2(t) represent the weight coefficients of the sine and cosine components of the notch frequency, respectively, u represents the iterative step size of the algorithm, and represents the notch frequency of the multi-frequency signal; In the full speed range, when the rotor speed ω r When the speed changes, the corresponding ω of the LMS algorithm also changes accordingly. First, the rotor co-frequency vibration displacement phase is extracted, and then the phase angle θ is compensated on the basis of the co-frequency vibration displacement phase as the speed changes. The sine and cosine values ​​of the phase compensation angle θ are calculated with the notch frequency weight coefficients w1(t) and w2(t) respectively to obtain the weight coefficients after phase compensation: w'1(t)=w2(t)sinθ-w1(t)cosθ, w'2(t)=w2(t)cosθ+w1(t)sinθ. The output signal of the phase shift LMS notch method is y′(t)=w'1(t)sin(ωt)+w'2(t)cos(ωt).

6. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 5, characterized in that Step (3) is to obtain the control signal amplitude at multiple speeds by using the influence coefficient method according to the output signal y′(t) at multiple speeds obtained in step (2).

7. The full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1 is characterized in that Step (3) The influence coefficient method is carried out as follows: select a number of correction planes and a number of measuring planes on the rotor for multiple operation corrections, and take the vibration of a certain measuring plane caused by a unit correction voltage signal on a certain correction plane at a certain speed as an influence coefficient; obtain these influence coefficients by measurement or calculation, and determine the size and direction of the control signal that should be added to each correction plane in order to limit the vibration of each measuring plane to below a certain amplitude based on the vibration caused by the unbalance.

8. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 6, characterized in that Step (3) includes step (3.1): Influence coefficient method for a four-degree-of-freedom rigid rotor: The front radial magnetic bearing and the rear radial magnetic bearing correspond to the two correction planes P on the rotor respectively. b1 and P b2 The front radial magnetic bearing sensor and the rear radial magnetic bearing sensor correspond to the rotor measurement surface P s1 and P s2 ; Measured P s1 and P s2 The initial vibration displacement vectors of the two measuring surfaces are: and Assume that the plane P where the left and right magnetic bearings are located b1 and P b2 The plane P where the sensor is located is the rotor imbalance correction surface. s1 and P s2 It is the unbalance detection surface; At any equilibrium speed ω, on the correction surface P b1 Test and add control voltage signal There is no hysteresis angle in its voltage signal, and the vibration displacement vector at each detection surface is obtained as follows: and Find the correction surface P b1 Test the control voltage Rear detection surface P s1 and P s2 Influence coefficient of displacement: Remove the try signal On the correction surface P b2 Test and add control voltage signal The vibration displacement vector at each detection surface is obtained as and Find the correction surface P b2 Test the control voltage Rear detection surface P s1 ,P s2 Influence coefficient of displacement: Get the correction surface P b1 and P b2 The voltage signals to be added are:

9. A full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 8, characterized in that Step (3) also includes step (3.2): performing least squares fitting on the control voltage amplitudes U1 and U2 measured at multiple speeds ω in step (3.1) to obtain a curve of control voltage changing with speed; the least squares fitting is one of a quartic polynomial fitting method, an exponential function fitting, a logarithmic function fitting, and a power function fitting.

10. The full-speed magnetic suspension high-speed motor rotor vibration control method according to claim 1 is characterized in that The specific process of applying step (4) to the vibration control of the magnetic suspension high-speed motor rotor is as follows: establish a control model of the electromagnetic bearing-flexible rotor system, introduce the signal phase compensation extracted by the LMS algorithm into the control model, and use the sensor G s The measured displacement signal X(s) is negatively fed back to the PID controller G c LMS notch filter with phase compensation G i The output amplitude of the phase compensation LMS notch filter is the control signal amplitude, which is then output to the power amplifier G a The output current of the power amplifier acts on the electromagnetic bearing to achieve active vibration control of the rotor.

Images