A bearingless permanent magnet synchronous motor speed displacement double sensorless control method

Abstract

This invention discloses a sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor. The method involves acquiring the current and voltage components of the torque winding in a stationary coordinate system, processing them through a nonlinear flux linkage observer to obtain a rotor angle error signal, and then processing this error signal through an extended state observer phase-locked loop to obtain the rotor speed and angle. Based on the deviation between the rotor speed and the speed setpoint, a closed-loop speed adjustment is performed to generate a torque winding control signal. The method also involves acquiring the motor-side current component and the current and voltage components of the levitation winding in a rotating coordinate system, processing them through an effective flux linkage observer to obtain the rotor radial displacement component in the rotating coordinate system, and then transforming this displacement component based on the rotor angle to obtain the displacement in the stationary coordinate system. Based on the deviation between this displacement and the displacement setpoint, a closed-loop displacement adjustment is performed to generate a levitation winding control signal.

Description

Technical Field

[0001] This invention relates to a sensorless control method for speed and displacement of a bearingless permanent magnet synchronous motor. Background Technology

[0002] Bearingless permanent magnet synchronous motors (BPMSMs) are a new type of special motor that integrates magnetic levitation technology and permanent magnet synchronous motor technology. By integrating rotor rotation drive with radial suspension function, they achieve operating characteristics of no mechanical contact, no friction and wear, high efficiency, and high power density. These motors are widely used in special fields with stringent requirements for operational stability and environmental cleanliness, such as high-speed precision machining, turbomolecular pumps, centrifuges, aerospace, biomedicine, and vacuum cleanrooms.

[0003] Unlike traditional permanent magnet synchronous motors (PMSMs) with mechanical bearings, bearingless PMSMs do not rely on mechanical bearings to support the rotor. Instead, they achieve stable, contactless rotor levitation by controlling the radial levitation force generated by the stator windings, while simultaneously driving the rotor to rotate at high speed via torque windings. To ensure stable levitation and precise rotation of the rotor across the entire operating range, existing control systems typically require both speed and displacement sensors to detect rotor speed and radial displacement signals, respectively. However, as electromechanical components, sensors are prone to failure in harsh operating environments such as high speed, high temperature, high vacuum, or strong electromagnetic interference. Failure of these signal detection components can lead to rotor instability or even stator-rotor collisions, causing irreversible system damage. Therefore, researching dual sensorless speed and displacement control technology is crucial for achieving highly reliable operation of bearingless PMSMs.

[0004] Existing sensorless control technologies are mainly divided into two dimensions: speed sensorless control and displacement sensorless control, but both have significant limitations. In speed sensorless control, under medium- and high-speed conditions, back-EMF-based observation methods (such as sliding mode observers, model reference adaptive systems, and extended Kalman filters) are typically used for speed estimation. However, under zero-speed and low-speed operating conditions, due to the weak back-EMF signal, these methods struggle to achieve accurate observation. Existing technologies often employ high-frequency injection methods to address the low-speed domain observation challenge, but the introduction of high-frequency signals can cause jitter in the control system, directly affecting the accuracy of speed estimation and rotor suspension stability, thus limiting its application in high-precision scenarios. Furthermore, traditional observer-based methods often rely on accurate motor models and electrical angular velocity information, resulting in poor parameter robustness.

[0005] In sensorless displacement control, existing technologies mostly employ high-frequency signal injection or inverse system structures for displacement estimation. While high-frequency signal injection can acquire displacement information, it can cause adverse disturbances to high-performance bearingless permanent magnet synchronous motors, degrading control performance. Inverse system structures, on the other hand, place extremely high demands on controller performance, resulting in high implementation complexity and hindering product commercialization. More critically, the stable operation of bearingless permanent magnet synchronous motors depends on the coordinated operation of sensorless speed control of the rotating part of the motor and sensorless displacement control of the levitated rotor. These two mechanisms must seamlessly switch and coordinate across the entire speed range. However, existing technologies are mostly optimized for a single control objective (speed or displacement only), lacking system design for a coordinated strategy of dual sensorless control, making it difficult to achieve stable levitation and rotation control across the entire speed range from zero to high speed.

[0006] In summary, existing sensorless control technology for bearingless permanent magnet synchronous motors still has shortcomings in low-speed observation accuracy, system jitter suppression, parameter robustness, and dual sensorless collaborative control. There is an urgent need for a method that can achieve high-precision dual sensorless control of speed and displacement across the entire operating range, from zero speed to low speed and medium-to-high speed, in order to eliminate potential sensor failures and improve system reliability. Summary of the Invention

[0007] The present invention provides a sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor to solve the problems existing in the prior art.

[0008] The technical solutions adopted in this invention are as follows:

[0009] A sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor includes the following steps:

[0010] The current and voltage components of the torque winding in the stationary coordinate system are collected. The rotor angle error signal is obtained by processing the current and voltage components through a nonlinear flux linkage observer. The rotor angle error signal is then processed by an extended state observer phase-locked loop to obtain the rotor speed and angle.

[0011] Based on the deviation between the rotor speed and the speed setpoint, a closed-loop speed regulation is performed to generate a torque winding control signal.

[0012] The motor-side current component and the current and voltage components of the suspension winding are collected in the rotating coordinate system. Based on the motor-side current component and the current and voltage components of the suspension winding, the rotor radial displacement component in the rotating coordinate system is obtained by processing the effective flux linkage observer. Based on the rotor angle, the rotor radial displacement component is transformed to obtain the displacement in the stationary coordinate system.

[0013] Displacement closed-loop adjustment is performed based on the deviation between the displacement in the stationary coordinate system and the given displacement value to generate a levitation winding control signal.

[0014] Furthermore, the nonlinear flux linkage observer calculates the stator flux linkage based on the current and voltage components, constructs a cost function using the gradient descent method, and iteratively updates it to obtain the rotor angle error signal, wherein:

[0015] The formula for calculating stator flux linkage is:

[0016] ,

[0017] The cost function is:

[0018] ,

[0019] The iterative update law is:

[0020] ,

[0021] in, The stator flux linkage vector in the stationary coordinate system. , The voltage component of the torque winding in the stationary coordinate system. For stator winding resistance, , This represents the current component of the torque winding in a stationary coordinate system. Let cost function be This refers to the flux linkage amplitude of the permanent magnet. This is the estimated value of the stator flux linkage. For vector mapping functions, For observer adaptive gain, This is the time derivative of the stator flux linkage vector.

[0022] Furthermore, the extended state observer phase-locked loop uses the rotor electrical angle, rotational speed, and angular acceleration as state variables to construct the following third-order extended state observer:

[0023] ,

[0024] in, This is the rotor angle error signal. , , These are the observed values ​​of rotor electrical angle, rotational speed, and angular acceleration, respectively. , , The observer gain coefficient, This represents the observation error.

[0025] Furthermore, the effective flux linkage observer obtains the rotor radial displacement component in the rotating coordinate system based on error corrections between the current model and the voltage model, wherein the current model satisfies:

[0026] ,

[0027] in, , For the direct-axis and quadrature-axis components of the flux linkage of the levitation winding, , These are the direct-axis and quadrature-axis components of the motor-side current. Mutual inductance coefficient, The equivalent excitation current, , The radial displacement component of the rotor in the rotating coordinate system is given.

[0028] Furthermore, the voltage model satisfies:

[0029] ,

[0030] in, , These are the voltage and current vectors of the levitation winding, respectively. For stator resistance, The rotor's electric angular velocity, Orthogonal rotating matrix commonly used in permanent magnet synchronous motor control

[0031] Furthermore, the coordinate transformation is an inverse Park transformation, used to convert the rotor radial displacement component in the rotating coordinate system into displacement in the stationary coordinate system.

[0032] Furthermore, the speed closed-loop regulation adopts a PI controller. The input of the PI controller is the deviation between the rotor speed and the speed setpoint, and the output is the torque winding quadrature-axis current setpoint, while the torque winding direct-axis current setpoint is set to zero.

[0033] Furthermore, the displacement closed-loop regulation adopts a PID controller, the input of which is the deviation between the displacement in the stationary coordinate system and the displacement setpoint, and the output is the levitation force setpoint.

[0034] Further, the generation of the levitation winding control signal includes: obtaining a levitation winding current setpoint by converting the levitation force setpoint through a force-current conversion module; obtaining a levitation winding voltage setpoint by PI control based on the deviation between the levitation winding current setpoint and the actual current of the levitation winding; and generating the levitation winding control signal by coordinate transformation and SVPWM modulation based on the levitation winding voltage setpoint.

[0035] Furthermore, the rotor angle is fed back to the expansion state observer phase-locked loop as an input variable.

[0036] The present invention has the following beneficial effects:

[0037] (1) By adopting a dual sensorless control strategy for speed and displacement, closed-loop control of rotor speed and radial position can be achieved without installing speed and displacement sensors, eliminating the risk of system failure due to sensor hardware failure. It is particularly suitable for harsh working conditions such as high speed, high temperature and vacuum where sensors are prone to failure.

[0038] (2) The nonlinear flux observer directly extracts the flux signal instead of relying on the back EMF, which breaks through the limitation of the traditional back EMF observation method that cannot work in the zero speed and low speed domain due to the weak signal. Combined with the phase-locked loop of the extended state observer, it can obtain effective speed information during the motor start-up and low-speed operation stages.

[0039] (3) The displacement observation adopts the method based on the effective flux linkage model, which can estimate the radial displacement of the rotor without injecting high-frequency detection signals into the system. This avoids the control system jitter problem caused by high-frequency injection and helps to improve the suspension stability of the rotor.

[0040] (4) The extended state observer phase-locked loop uses angular acceleration as an extended state variable, which can actively observe and compensate for the total disturbance of the system, reduce the dependence on the precise mathematical model of the motor, and improve the robustness of the control system under parameter drift or load disturbance conditions.

[0041] (5) Through the collaborative design of the motor-side controller and the suspension force-side controller, sensorless observation of speed and displacement is realized simultaneously, covering the entire operating range from zero speed to high speed, and solving the problem that a single control strategy is difficult to take into account the stable operation of the entire speed range. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the invention.

[0043] Figure 2 yes Figure 1 Structural schematic diagram of the motor-side controller.

[0044] Figure 3 yes Figure 2 The structural principle diagram of the sensorless speed and angle calculation module.

[0045] Figure 4 yes Figure 1 Structural principle diagram of the mid-suspension force side controller.

[0046] Figure 5 yes Figure 4 The structural principle diagram of the sensorless displacement calculation module. Detailed Implementation

[0047] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. This embodiment is for the dual sensorless control of speed and displacement of a bearingless permanent magnet synchronous motor, without setting up physical speed sensors and displacement sensors throughout the entire process.

[0048] The entire control system includes a motor-side controller 1 and a levitation force-side controller 2, which together form a complete control closed loop with the bearingless permanent magnet synchronous motor 3. The rotor speed, angle and radial displacement are accurately estimated through the sensorless speed and angle calculation module 40 and the sensorless displacement calculation module 50. With the help of PI, PID closed-loop regulation and SVPWM modulation strategy, the rotor achieves stable rotation and contactless levitation, which is suitable for application scenarios with strict requirements for motor operation accuracy and stability, such as high-speed precision machining, turbomolecular pumps, aerospace, and biomedicine.

[0049] like Figure 1 As shown, the output terminal of the motor-side controller 1 is connected to the torque winding input terminal of the bearingless permanent magnet synchronous motor 3, and is used to generate the torque winding control voltage to drive the rotor to rotate. The output terminal of the levitation force side controller 2 is connected to the levitation winding input terminal of the bearingless permanent magnet synchronous motor 3, and is used to generate the levitation winding control voltage to realize the radial levitation of the rotor. The levitation force side controller 2 and the motor-side controller 1 share the output signal of the sensorless speed angle calculation module 40 to ensure the synchronization of speed control and displacement control. The operating status signal of the bearingless permanent magnet synchronous motor 3 will be fed back to the motor-side controller 1 and the levitation force side controller 2 respectively to complete the bidirectional signal interaction of the control system.

[0050] like Figure 2 As shown, the motor-side controller 1 includes a speed closed-loop regulator 70, a first PI controller 71, a second PI controller 72, a third PI controller 73, a first coordinate transformation module 74, a second coordinate transformation module 75, a first SVPWM inverter 76, a current sensor 77, and a sensorless speed-angle calculation module 40. The sensorless speed-angle calculation module 40 consists of a nonlinear flux linkage observer 41 and an extended state observer phase-locked loop 42. During actual control, the current sensor 77 collects the two-phase current of the torque winding of the bearingless permanent magnet synchronous motor 3. and The acquired current signal is input into the second coordinate transformation module 75. After being processed by Clarke transformation and Park transformation in sequence, the torque winding current component in the stationary coordinate system is obtained. , and the actual values ​​of the quadrature and direct axis currents in the rotating coordinate system , Simultaneously, the torque winding voltage components in the stationary coordinate system are collected. , , to the current component , and voltage components , The values ​​are input together into the nonlinear flux linkage observer 41 of the sensorless speed-angle calculation module 40. A preliminary estimate of the rotor electrical angle is first obtained using the formula:

[0051] ,

[0052] The expression for the air gap flux linkage in the stationary coordinate system is:

[0053] ,

[0054] in, and The air gap flux linkages for the α and β axes are respectively composed of the stator flux linkage and the rotor flux linkage. The stator flux linkage is calculated using the following formula:

[0055] ,

[0056] Based on the above magnetic flux linkage calculation results, a vector function is defined. The function satisfies:

[0057] ,

[0058] Redefine the cost function:

[0059] ,

[0060] The gradient corresponding to this cost function is:

[0061] ,

[0062] The iterative update law for the observer is constructed using the gradient descent method, and the formula is as follows:

[0063] ,

[0064] After the above series of calculations, the nonlinear flux linkage observer 41 outputs the rotor angle error signal. .

[0065] like Figure 3 As shown, the rotor angle error signal The input is fed into the extended state observer phase-locked loop 42 of the sensorless velocity-angle calculation module 40. First, the angle error signal is processed using the following formula:

[0066] ,

[0067] To improve the dynamic performance of the control system, the derivative of the rotational speed is used as the extended state variable of the extended state observer. The physical meaning of this variable is angular acceleration. The electrical angle of the motor, the rotational speed, and the angular acceleration satisfy a derivative relationship, as shown in the formula:

[0068] ,

[0069] The above formula satisfies the cascaded canonical form of integrals and is observable. Therefore, a third-order extended state observer is constructed based on this relationship, with the following formula:

[0070] ,

[0071] After calculation by the extended state observer phase-locked loop 42, the actual rotor speed of the bearingless permanent magnet synchronous motor 3 is output. and rotor angle value and the rotor angle value Feedback is sent back to the extended state observer phase-locked loop 42 as an input variable to form a closed-loop correction for angle observation, thereby improving the estimation accuracy of rotational speed and angle.

[0072] The actual rotor speed With speed setpoint The speed error is calculated and input to the speed closed-loop regulator 70, which is composed of the first PI controller 71. After adjustment by the first PI controller 71, the given value of the quadrature axis current of the torque winding is output. and set the direct-axis current of the torque winding to a given value. Set to 0; set the quadrature-axis current setpoint. Actual value of quadrature axis current The difference is input to the second PI controller 72, and after adjustment, the quadrature-axis voltage setpoint is obtained. Set the direct-axis current value Compared with the actual value of direct-axis current The difference is input to the third PI controller 73, and after adjustment, the direct-axis voltage setpoint is obtained. Then set the quadrature-axis voltage value. and direct-axis voltage setpoint The input to the first coordinate transformation module 74, after inverse Park transformation, yields the torque winding voltage in the stationary coordinate system. and The voltage signal is input to the first SVPWM inverter 76, and after modulation by the first SVPWM inverter 76, it generates the three-phase input voltage of the torque winding of the bearingless permanent magnet synchronous motor 3. , , The three-phase voltage is input to the bearingless permanent magnet synchronous motor 3, which drives the rotor to achieve stable rotation.

[0073] like Figure 4 As shown, the levitation force side controller 2 consists of a displacement closed-loop regulator 80, a first PID controller 81, a second PID controller 82, a force-current conversion module 83, a fourth PI controller 84, a fifth PI controller 85, a coordinate transformation module 86, a second SVPWM inverter 90, a coordinate transformation module 91, a current sensor 92, a sensorless velocity-angle calculation module 40, and a sensorless displacement calculation module 50. The sensorless displacement calculation module 50 is composed of an effective magnetic flux observer 51 and a coordinate transformation module 52.

[0074] During actual control, current sensor 92 collects the two-phase current of the suspension winding of bearingless permanent magnet synchronous motor 3. and The acquired current signal is input into the coordinate transformation module 91, and after Clarke transformation and Park transformation, the actual values ​​of the quadrature and direct axis currents of the levitation winding in the rotating coordinate system are obtained. , Simultaneously, the voltage components of the levitation winding in the rotating coordinate system are collected. , The current component of the levitation winding , Voltage components , The motor-side current component in the rotating coordinate system output by motor-side controller 1 , The data are input together into the effective magnetic flux observer 51 of the sensorless displacement calculation module 50.

[0075] like Figure 5 As shown, the effective flux linkage observer 51 first calculates the flux linkage of the levitation winding based on the current model. The formula for the current model is:

[0076] ,

[0077] The above current model is defined as ,in , , Next, based on the voltage equation, the voltage model is written. The formula for the voltage equation is:

[0078] ,

[0079] Based on this voltage equation, the voltage model can be described as follows:

[0080] ,

[0081] Then, the flux linkage error between the current model and the voltage model is calculated. The error formula is as follows:

[0082] ,

[0083] An effective flux linkage observer is constructed based on this flux linkage error. The observer formula is as follows:

[0084] ,

[0085] in This is the observer gain.

[0086] After calculation using the above series of formulas, the effective flux linkage observer 51 outputs the rotor radial displacement component in the rotating coordinate system. and The displacement component is input into the coordinate transformation module 52, and the rotor angle is calculated based on the output of the sensorless velocity-angle calculation module 40. After processing with the inverse Park transform, the actual displacement of the rotor in the stationary coordinate system is obtained. and .

[0087] The actual displacement of the rotor , With displacement given value , The difference calculations are performed separately, and the resulting displacement errors are input to the displacement closed-loop controller 80, which consists of a first PID controller 81 and a second PID controller 82. After adjustment by the two PID controllers, the setpoint of the suspension force is output. and ; Set the levitation force to a given value and The input force-current conversion module 83 processes the input to obtain the quadrature-axis current setpoint of the suspension winding. and direct-axis current setpoint Set the quadrature-axis current value Actual value of quadrature axis current The difference is input to the fourth PI controller 84, and after adjustment, the quadrature-axis voltage setpoint is obtained. Set the direct-axis current value Compared with the actual value of direct-axis current The difference is input to the fifth PI controller 85, and after adjustment, the direct-axis voltage setpoint is obtained. Then set the quadrature-axis voltage value. and direct-axis voltage setpoint The input coordinate transformation module 86, after inverse Park transformation, yields the levitation winding voltage in the stationary coordinate system. and The voltage signal is input to the second SVPWM inverter 90, and after modulation by the second SVPWM inverter 90, it generates the three-phase input voltage of the suspension winding of the bearingless permanent magnet synchronous motor 3. , , The three-phase voltage is input to the bearingless permanent magnet synchronous motor 3 to achieve radial non-contact stable levitation of the rotor.

[0088] In this embodiment, the motor-side controller 1 and the levitation force-side controller 2 share rotor speed and angle signals through the sensorless speed and angle calculation module 40, which fundamentally ensures the coordination and synchronization of rotor rotation control and radial levitation control. Throughout the process, the nonlinear flux observer 41, the extended state observer phase-locked loop 42, and the effective flux observer 51 complete the sensorless estimation of speed, angle, and displacement. The entire process does not rely on weak back electromotive force signals, nor does it require the injection of high-frequency signals into the control system. It can achieve accurate signal observation in the full operating range of zero speed, low speed, and medium and high speed.

[0089] The extended state observer phase-locked loop 42 can actively observe and compensate for the total disturbance of the system without relying on a precise mathematical model of the motor. This effectively improves the parameter robustness of the control system and completely solves the rotor instability problem when detection elements such as speed sensors and displacement sensors fail. It realizes high-precision and high-stability dual sensorless control of the speed and displacement of the bearingless permanent magnet synchronous motor 3, while avoiding the control system jitter problem caused by high-frequency signal injection. This improves the rotor suspension stability and speed estimation accuracy, ensuring that the bearingless permanent magnet synchronous motor 3 can operate reliably under various harsh working conditions.

[0090] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be considered within the scope of protection of the present invention.

Claims

1. A sensorless control method for speed and displacement of a bearingless permanent magnet synchronous motor, characterized in that: Includes the following steps: The current and voltage components of the torque winding in the stationary coordinate system are collected. The rotor angle error signal is obtained by processing the current and voltage components through a nonlinear flux linkage observer. The rotor angle error signal is then processed by an extended state observer phase-locked loop to obtain the rotor speed and angle. Based on the deviation between the rotor speed and the speed setpoint, a closed-loop speed regulation is performed to generate a torque winding control signal. The motor-side current component and the current and voltage components of the suspension winding are collected in the rotating coordinate system. Based on the motor-side current component and the current and voltage components of the suspension winding, the rotor radial displacement component in the rotating coordinate system is obtained by processing the effective flux linkage observer. Based on the rotor angle, the rotor radial displacement component is transformed to obtain the displacement in the stationary coordinate system. Displacement closed-loop adjustment is performed based on the deviation between the displacement in the stationary coordinate system and the given displacement value to generate a levitation winding control signal.

2. The sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The nonlinear flux linkage observer calculates the stator flux linkage based on the current and voltage components, constructs a cost function using the gradient descent method, and iteratively updates it to obtain the rotor angle error signal, wherein: The formula for calculating stator flux linkage is: ， The cost function is: ， The iterative update law is: ， in, The stator flux linkage vector in the stationary coordinate system. , The voltage component of the torque winding in the stationary coordinate system. For stator winding resistance, , This represents the current component of the torque winding in a stationary coordinate system. Let cost function be This refers to the flux linkage amplitude of the permanent magnet. This is the estimated value of the stator flux linkage. For vector mapping functions, For observer adaptive gain, This is the time derivative of the stator flux linkage vector.

3. The dual sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The extended state observer phase-locked loop uses the rotor electrical angle, rotational speed, and angular acceleration as state variables to construct the following third-order extended state observer: ， in, This is the rotor angle error signal. , , These are the observed values ​​of rotor electrical angle, rotational speed, and angular acceleration, respectively. , , The observer gain coefficient, This represents the observation error.

4. The speed and displacement dual sensorless control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The effective flux linkage observer obtains the rotor radial displacement components in the rotating coordinate system based on error corrections of the current model and voltage model. The current model satisfies: ， in, , For the direct-axis and quadrature-axis components of the flux linkage of the levitation winding, , These are the direct-axis and quadrature-axis components of the motor-side current. Mutual inductance coefficient, The equivalent excitation current, , The radial displacement component of the rotor in the rotating coordinate system is given.

5. The dual sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor according to claim 4, characterized in that: The voltage model satisfies: ， in, , These are the voltage and current vectors of the levitation winding, respectively. For stator resistance, The rotor's electric angular velocity, It is an orthogonal rotating matrix commonly used in the control of permanent magnet synchronous motors.

6. The dual sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The coordinate transformation is the inverse Park transformation, which is used to convert the rotor radial displacement components in the rotating coordinate system into displacements in the stationary coordinate system.

7. The dual sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The speed closed-loop regulation adopts a PI controller. The input of the PI controller is the deviation between the rotor speed and the speed setpoint, and the output is the torque winding quadrature axis current setpoint, while the torque winding direct axis current setpoint is set to zero.

8. The dual sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The displacement closed-loop regulation adopts a PID controller. The input of the PID controller is the deviation between the displacement in the stationary coordinate system and the displacement setpoint, and the output is the levitation force setpoint.

9. The sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 8, characterized in that: The process of generating the levitation winding control signal includes: obtaining a levitation winding current setpoint by converting the levitation force setpoint through a force-current conversion module; obtaining a levitation winding voltage setpoint by PI control based on the deviation between the levitation winding current setpoint and the actual levitation winding current; and generating the levitation winding control signal by coordinate transformation and SVPWM modulation based on the levitation winding voltage setpoint.

10. The sensorless speed and displacement control method for a bearingless permanent magnet synchronous motor as described in claim 1, characterized in that: The rotor angle is fed back to the phase-locked loop of the expansion state observer as an input variable.

Images