A five-axis magnetic suspension bearing winding inductance online identification method based on variable bias sine small signal injection

Abstract

The application discloses a five-axis magnetic suspension bearing winding inductance online identification method based on variable bias sine small signal injection, and belongs to the field of magnetic suspension bearing control. The application realizes real-time and high-precision identification of inductance parameters by injecting a discrete small-amplitude sine current signal into a control system and combining control timing, provides accurate winding model support for stable control of the magnetic suspension bearing, and finally repeats the above process, changes a winding current distribution strategy, identifies real-time updates of winding inductance and resistance parameters of each degree of freedom with a magnetic saturation degree, and changes different suspension positions, and identifies real-time updates of winding inductance and resistance parameters of each degree of freedom with a rotor position.

Description

Technical Field

[0001] This invention belongs to the field of magnetic levitation bearing control, and more specifically, relates to an online identification method for the winding inductance of a five-axis magnetic levitation bearing based on variable bias sinusoidal small signal injection. Background Technology

[0002] Magnetic levitation bearings achieve contactless rotor levitation through electromagnetic force, offering significant advantages such as frictionless operation, low loss, high speed, and long lifespan. They are widely used in high-end equipment fields such as aerospace, high-speed motors, and precision machine tools. Five-axis magnetic levitation bearings, in particular, require coordinated control of five degrees of freedom to achieve stable rotor levitation. The control accuracy of this system directly depends on the accuracy of electrical parameters such as winding inductance and resistance, forming the core foundation for constructing a high-precision magnetic levitation control model.

[0003] However, the winding inductance and resistance parameters of a five-axis magnetic levitation bearing are not constant values. They are affected by various factors such as rotor levitation displacement and winding current, exhibiting significant time-varying characteristics. For example, different winding current magnitudes lead to different degrees of magnetic saturation, causing changes in inductance parameters. Maintaining a constant equilibrium position and creating a rich excitation environment using variable bias current is crucial for identifying the real-time updates of winding inductance and resistance parameters for each degree of freedom with varying magnetic saturation. The rotor's different operating positions cause changes in air gap reluctance, directly inducing nonlinear fluctuations in inductance parameters. Therefore, identifying the real-time updates of winding inductance and resistance parameters for each degree of freedom with rotor position is also of great significance.

[0004] Existing inductance identification technology mainly relies on traditional offline measurement with LCR meters. Offline identification methods need to be completed when the control system is stopped, which cannot adapt to the time-varying characteristics of inductance during operation and makes it difficult to meet the control requirements under dynamic operating conditions.

[0005] Therefore, there is an urgent need for a technical solution that adapts to the resource limitations of microcontrollers, does not affect the stable operation of the system, and can achieve accurate online identification of time-varying inductance parameters. This solution would address the core pain points in existing technologies, such as the difficulty in identifying time-varying inductance of five-axis magnetic levitation bearing windings, poor real-time performance, and stability conflicts, and provide key technical support for the high-precision and high-stability control of five-axis magnetic levitation bearings. Summary of the Invention

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides an online identification method for the inductance of five-axis magnetic levitation bearing windings based on variable bias sinusoidal small-signal injection, aiming to solve the problem of difficulty in accurately identifying the inductance parameters of existing magnetic levitation bearing windings online.

[0007] To achieve the above objectives, this invention provides an online identification method for the winding inductance of a five-axis magnetic levitation bearing based on variable bias sinusoidal small-signal injection. The five-axis magnetic levitation bearing includes 10 windings, with each pair of windings forming a group to create opposite differential forces that control the rotor's levitation in one direction. The identification method includes the following steps: (1) The inductance and resistance of the windings in each direction (each group) are identified. Taking one direction as an example, the rotor suspension position is fixed, and the current of the two windings (winding A and winding C) is regulated by the control system to maintain the magnetic levitation bearing dynamically suspended in the specified position. (2) On the basis of the steady-state control current that keeps the magnetically levitated rotor balanced by winding A and winding C, a discrete small-amplitude sinusoidal current injection signal is added to obtain a comprehensive current signal containing the control signal. The current loop obtains a comprehensive control voltage, which is then amplified by power to obtain a winding voltage containing discrete small-amplitude sinusoidal components. This voltage is applied to the winding and then to the winding output current containing discrete small-amplitude sinusoidal components. (3) After obtaining the data, the discrete small-amplitude sinusoidal voltage component and the discrete small-amplitude sinusoidal current component with the same frequency as the injected signal are extracted from the winding voltage and the winding output current, respectively, using the Direct Discrete Fourier Transform (DFT) algorithm. (4) Calculate the winding inductance and resistance in real time based on the extracted sinusoidal voltage component and sinusoidal current component of the same frequency, so as to realize the online real-time update of parameters; (5) While keeping the suspension position unchanged, change the winding steady-state control current distribution strategy. The steady-state control current trend of the two windings in each group is the same increase and decrease. Create a new working condition, repeat the above steps, and identify the real-time update of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation. (6) Change the magnitude of the steady-state control current of one side winding to change different suspension positions, create new working conditions, repeat the above steps, and identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position.

[0008] Furthermore, the magnetic levitation bearing must first be in a stable levitation state. The magnetic levitation bearing control system is activated, and the microcontroller drives the windings to generate a stable electromagnetic force, causing the rotor of the five-axis magnetic levitation bearing to gradually converge to the rated levitation position and maintain a stable levitation state.

[0009] Furthermore, a discrete small-amplitude sinusoidal current injection signal is added to windings A and C on top of the steady-state control current that maintains the balance of the magnetically levitated rotor. This includes superimposing a discrete small-amplitude sinusoidal current injection signal onto the steady-state control current that the microcontroller uses to maintain the balance of the magnetically levitated rotor. The expression for the discrete small-amplitude sinusoidal current injection signal is: ,in, The injected signal amplitude is much smaller than the rated operating current to avoid affecting the stable levitation of the magnetic bearing. To inject the signal angular frequency, For discrete time indexes, This refers to the control cycle of the microcontroller.

[0010] Furthermore, the amplitude of the discrete small-amplitude sinusoidal current injection signal The current should be much lower than the rated operating current to avoid affecting the stable levitation of the magnetic bearing. The value should be 5%-10% of the rated operating current, and the frequency range corresponding to the injected signal angular frequency should be greater than 100Hz. Care should be taken to avoid signal aliasing. Furthermore, within the control cycle, the microcontroller prioritizes completing the core logic of control error calculation and control quantity output, placing the discrete Fourier transform signal extraction and calculation in the idle time sequence of the control to ensure that the real-time performance of the control is not affected.

[0011] Furthermore, the control voltage, which obtains discrete small-amplitude sinusoidal components via the controller, is applied to the winding, resulting in an output current with discrete small-amplitude sinusoidal components after passing through the winding. This includes: the discrete small-amplitude sinusoidal current injection signal being adjusted by the microcontroller to form a control voltage containing discrete small-amplitude sinusoidal components; this voltage being amplified by a power amplifier module and then input to the winding, driving the winding to generate an output current signal containing discrete small-amplitude sinusoidal components.

[0012] Furthermore, the step of extracting discrete small-amplitude sinusoidal voltage components and discrete small-amplitude sinusoidal current components with the same frequency as the injected signal from the winding voltage and current signals respectively includes: using the direct discrete Fourier transform (DFT) algorithm to extract discrete small-amplitude sinusoidal voltage components and discrete small-amplitude sinusoidal current components with the same frequency as the injected signal from the winding voltage and current signals respectively.

[0013] Furthermore, the method of utilizing the Direct Discrete Fourier Transform (DFT) algorithm includes: First, preprocessing the acquired winding voltage and winding output current by filtering out DC bias components. Then, for the target frequency point corresponding to the injected signal frequency, calculating the sum of the real and imaginary parts of the DFT to avoid redundant calculations in the full-band transformation. Finally, calculating the amplitude and phase of the target frequency component based on the sum of the real and imaginary parts to ensure extraction accuracy.

[0014] Furthermore, the step of calculating the winding inductance and resistance in real time based on the extracted sinusoidal voltage and current components of the same frequency includes: calculating the amplitude ratio of the extracted sinusoidal voltage and current components of the same frequency. ( The amplitude of the sinusoidal voltage component. (Amplitude of sinusoidal current component) and phase difference ( The phase of the sinusoidal voltage component. (This refers to the phase of the sinusoidal current component). Based on the impedance characteristics of the inductor, the formula... , The inductance parameters of the winding are calculated. and series resistance parameters Complete the inductor model identification.

[0015] Furthermore, while maintaining the floating position unchanged, the winding current distribution strategy is changed to identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation. This includes: at the beginning of identification, the minimum value of winding A and winding C is 0; at the end of identification, the maximum value of winding A and winding C is the current limit value; the winding current is gradually increased in the middle; due to the differential control characteristics, the current of winding A and winding C increases and decreases together, keeping the floating position unchanged; by changing the current distribution strategy, about 10 working conditions are created to identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation.

[0016] Further optimized, the step of changing the magnitude of the current in one winding to change the rotor's levitation position and identifying the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position includes: at the beginning of identification, a small bias current is applied to windings A and C, so that the magnetic levitation bearing is statically levitated at the center position; the current in one of the windings is gradually increased, and the rotor's levitation position is changed due to the differential control characteristics, creating about 10 working conditions, and identifying the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position.

[0017] The present invention also provides an electronic device, comprising: a computer-readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read executable instructions stored in the computer-readable storage medium and execute the above-described method.

[0018] The present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to perform the above-described method.

[0019] The present invention also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the above-described method.

[0020] In summary, compared with existing technologies, the technical solution conceived in this invention provides an online identification method for the inductance of a five-axis magnetic levitation bearing winding based on sinusoidal small-signal injection. First, the position control system is activated to bring the rotor of the five-axis magnetic levitation bearing to its rated levitation position and maintain a stable state. Then, a discrete small-amplitude sinusoidal current injection signal is superimposed after the differential control stage. This injection signal, after adjustment, drives the winding to generate voltage and current signals containing sinusoidal components of the same frequency. Utilizing the idle time of the microcontroller cycle, the sinusoidal voltage and current components with the same injection frequency are extracted from the signal using a direct DFT algorithm, and their amplitude ratio and phase difference are calculated. Based on the inductor impedance characteristics, the winding inductance L and series resistance R are derived through formula derivation, and the above process is repeated to achieve real-time parameter updates. In this invention, the injection signal amplitude is 5%-10% of the rated current to avoid interfering with system stability. The direct DFT algorithm eliminates the need for full-band computation, adapting to microcontroller resource limitations and balancing real-time identification with accuracy. This method solves the problem that traditional LCR meters cannot adapt to time-varying parameters in offline measurements, and provides accurate parameter support for the high-precision and stable control of five-axis magnetic levitation bearings. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the magnetic levitation bearing control system provided by the present invention.

[0022] Figure 2 This is a schematic diagram of the entire process of an online identification method for the winding inductance of a five-axis magnetic levitation bearing based on sinusoidal small-signal injection provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0024] To achieve the above objectives, this invention provides an online identification method for the winding inductance of a five-axis magnetic levitation bearing based on variable bias sinusoidal small-signal injection. The five-axis magnetic levitation bearing includes 10 windings, with each pair of windings forming a group to create opposite differential forces that control the rotor's levitation in one direction. Figure 1 and 2 As shown, the identification method includes the following steps: (1) The inductance and resistance of the windings in each direction (each group) are identified. Taking one direction as an example, the rotor suspension position is fixed, and the current of the two windings (winding A and winding C) is regulated by the control system to maintain the magnetic levitation bearing dynamically suspended in the specified position. (2) On the basis of the steady-state control current that keeps the magnetically levitated rotor balanced by winding A and winding C, a discrete small-amplitude sinusoidal current injection signal is added to obtain a comprehensive current signal containing the control signal. The current loop obtains a comprehensive control voltage, which is then amplified by power to obtain a winding voltage containing discrete small-amplitude sinusoidal components. This voltage is applied to the winding and then to the winding output current containing discrete small-amplitude sinusoidal components. (3) After obtaining the data, the Direct Discrete Fourier Transform (DFT) algorithm is used to extract the frequency of the injected signal from the winding voltage and the winding output current, respectively. Discrete small-amplitude sinusoidal voltage components and discrete small-amplitude sinusoidal current components; (4) Calculate the winding inductance and resistance in real time based on the extracted sinusoidal voltage component and sinusoidal current component of the same frequency, so as to realize the online real-time update of parameters; (5) While keeping the suspension position unchanged, change the winding steady-state control current distribution strategy. The steady-state control current trend of the two windings in each group is the same increase and decrease. Create a new working condition, repeat the above steps, and identify the real-time update of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation. (6) Change the magnitude of the steady-state control current of one side winding to change different suspension positions, create new working conditions, repeat the above steps, and identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position.

[0025] Specifically, the magnetic levitation bearing must first be in a stable levitation state. First, the magnetic levitation bearing control system is activated, and the microcontroller drives the windings to generate a stable electromagnetic force. The differential force generated by the two sets of windings acts on the rotor, causing the rotor to gradually converge to the rated levitation position and maintain a dynamic levitation steady state.

[0026] Specifically, windings A and C are supplemented with discrete small-amplitude sinusoidal current injection signals based on the steady-state control current used by the microcontroller to maintain the balance of the magnetically levitated rotor. This includes superimposing discrete small-amplitude sinusoidal current injection signals onto the steady-state control current used by the microcontroller to maintain the balance of the magnetically levitated rotor. The expression for the discrete small-amplitude sinusoidal current injection signal is: ,in, The injected signal amplitude is much smaller than the rated operating current to avoid affecting the stable levitation of the magnetic bearing. To inject the signal angular frequency, For discrete time indexes, This refers to the control cycle of the microcontroller.

[0027] Specifically, the amplitude of the discrete small-amplitude sinusoidal current injection signal The value should be 5%-10% of the rated operating current, and the frequency range corresponding to the injected signal angular frequency should be greater than 100Hz. Care should be taken to avoid signal aliasing. Furthermore, within the control cycle, the microcontroller prioritizes completing the core logic of control error calculation and control quantity output, placing the discrete Fourier transform signal extraction and calculation in the idle time sequence of the control to ensure that the real-time performance of the control is not affected.

[0028] Specifically, the voltage with discrete small-amplitude sinusoidal components obtained by the controller is applied to the winding, and after passing through the winding, an output current with discrete small-amplitude sinusoidal components is obtained. This includes: the discrete small-amplitude sinusoidal current injection signal is adjusted by the microcontroller to form a voltage containing discrete small-amplitude sinusoidal components, which is then amplified by the power amplifier module to drive the winding to generate an output current signal containing discrete small-amplitude sinusoidal components.

[0029] Specifically, the step of extracting discrete small-amplitude sinusoidal voltage components and discrete small-amplitude sinusoidal current components with the same frequency as the injected signal from the winding voltage and current signals includes: using the Direct Discrete Fourier Transform (DFT) algorithm to extract discrete small-amplitude sinusoidal voltage components and discrete small-amplitude sinusoidal current components with the same frequency as the injected signal from the winding voltage and current signals, respectively.

[0030] Specifically, the step of calculating the winding inductance and resistance in real time based on the extracted sinusoidal voltage and current components of the same frequency includes: calculating the amplitude ratio of the extracted sinusoidal voltage and current components of the same frequency. ( The amplitude of the sinusoidal voltage component. (Amplitude of sinusoidal current component) and phase difference ( The phase of the sinusoidal voltage component. (This refers to the phase of the sinusoidal current component). Based on the impedance characteristics of the inductor, the formula... , The inductance parameters of the winding are calculated. and series resistance parameters Complete the inductor model identification.

[0031] Specifically, the method of using the Direct Discrete Fourier Transform (DFT) algorithm includes: First, preprocessing the acquired voltage signal and output current signal by filtering out DC bias components. Then, for the target frequency point corresponding to the injected signal frequency, calculating the sum of the real and imaginary parts of the DFT to avoid redundant calculations in the full-band transformation. Finally, calculating the amplitude and phase of the target frequency component based on the sum of the real and imaginary parts to ensure extraction accuracy.

[0032] Specifically, while maintaining the floating position, the winding current distribution strategy is changed to identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation. This includes: at the beginning of identification, the minimum value of winding A and winding C is 0; at the end of identification, the maximum value of winding A and winding C is the current limit value; the winding current is gradually increased in the middle; due to the differential control characteristics, the current of winding A and winding C increases and decreases together, keeping the floating position unchanged; by changing the current distribution strategy, about 10 working conditions are created to identify the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation.

[0033] Specifically, the process of changing the magnitude of the current in one winding to change the rotor's levitation position and identifying the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position includes: at the beginning of identification, a small bias current is applied to windings A and C, causing the magnetic levitation bearing to be statically levitated at the center position; the current in one of the windings is gradually increased, and the rotor's levitation position changes due to the differential control characteristics; approximately 10 working conditions are created, and the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position are identified.

[0034] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for online identification of winding inductance of a five-axis magnetic levitation bearing based on variable bias sinusoidal small-signal injection, wherein the magnetic levitation bearing comprises 10 windings, with each pair of windings forming a group to create opposite differential forces controlling the rotor's levitation in one direction; characterized in that, Online identification of winding inductance includes the following steps: (1) For any set of windings, the rotor suspension position is fixed, and the current flowing through the two windings is regulated by the control system to maintain the magnetic levitation bearing dynamically suspended at the specified position. (2) On the basis of the steady-state control current that keeps the magnetically levitated rotor balanced by the two windings, a discrete small-amplitude sinusoidal current injection signal is added to obtain a comprehensive current signal including the control signal. The comprehensive control voltage is obtained through the current loop. After power amplification, the winding voltage containing discrete small-amplitude sinusoidal components is obtained and applied to the winding. After passing through the winding, the winding output current containing discrete small-amplitude sinusoidal components is obtained. (3) Using the direct discrete Fourier transform algorithm, sinusoidal voltage components and sinusoidal current components with the same frequency as the injected signal are extracted from the winding voltage and winding output current, respectively. (4) Calculate the winding inductance and resistance in real time based on the extracted sinusoidal voltage component and sinusoidal current component of the same frequency, so as to realize the online real-time update of parameters; (5) While keeping the suspension position unchanged, change the winding steady-state control current distribution strategy. The steady-state control current trend of the two windings in each group is the same increase and decrease. Create a new working condition and repeat steps (2)-(4) to identify the real-time update of the inductance and resistance parameters of each degree of freedom winding with the degree of magnetic saturation. (6) Change the magnitude of the steady-state control current of one side winding to change different suspension positions, create new working conditions, repeat steps (2)-(4), and identify the real-time update of the inductance and resistance parameters of each degree of freedom winding with the rotor position.

2. The online identification method for the winding inductance of a five-axis magnetic levitation bearing according to claim 1, characterized in that, In step (1), the magnetic levitation bearing is controlled by the microcontroller of the control system, which drives the winding to generate a stable electromagnetic force, so that the rotor of the five-axis magnetic levitation bearing gradually converges to the rated levitation position and maintains a stable levitation state.

3. The online identification method for the winding inductance of a five-axis magnetic levitation bearing according to claim 2, characterized in that, Based on the steady-state control current that maintains the magnetic levitation rotor balance between the two windings, a discrete small-amplitude sinusoidal current injection signal is added, including: A discrete small-amplitude sinusoidal current injection signal is superimposed on the steady-state control current that maintains the balance of the magnetically levitated rotor by the microcontroller. ,in, The injected signal amplitude should be 5%-10% of the rated operating current. To inject the signal angular frequency, For discrete time indexes, The control cycle of the microcontroller is denoted as .

4. The method for online identification of the winding inductance of a five-axis magnetic levitation bearing according to claim 1, characterized in that, The direct discrete Fourier transform algorithm includes: The collected winding voltage and winding output current are preprocessed by filtering out DC bias components; for the target frequency point corresponding to the injected signal frequency, the sum of the real part and the sum of the imaginary part of the discrete Fourier transform are calculated; the amplitude and phase of the target frequency component are calculated based on the sum of the real and imaginary parts.

5. The online identification method for the winding inductance of a five-axis magnetic levitation bearing according to claim 3, characterized in that, The winding inductance and resistance are calculated in real time based on the extracted sinusoidal voltage and current components of the same frequency, including: Calculate the amplitude ratio of the extracted sinusoidal voltage and current components of the same frequency. and phase difference Based on the impedance characteristics of inductors, using the formula... , The inductance parameters of the winding are calculated. and series resistance parameters Complete the inductor model identification, among which The amplitude of the sinusoidal voltage component. The amplitude of the sinusoidal current component. The phase of the sinusoidal voltage component. The phase of the sinusoidal current component.

6. The method for online identification of the winding inductance of a five-axis magnetic levitation bearing according to claim 1, characterized in that, While keeping the levitation position unchanged, the steady-state control current distribution strategy of the windings is changed to identify the real-time updates of the inductance and resistance parameters of each degree of freedom of the windings as a function of magnetic saturation, including: At the start of identification, the minimum value of the two windings is 0. At the end of identification, the maximum value of the two windings is the current limit value. The winding current is gradually increased in the middle. Due to the differential control characteristics, the current of the two windings increases and decreases at the same time, keeping the floating position unchanged. By changing the current distribution strategy, multiple working conditions are created, and the inductance and resistance parameters of each degree of freedom winding are identified in real time as the magnetic saturation degree changes.

7. The online identification method for the winding inductance of a five-axis magnetic levitation bearing according to claim 1, characterized in that, By changing the magnitude of the steady-state control current of one winding, the real-time updates of the inductance and resistance parameters of each degree of freedom winding with the rotor position are identified, including: At the start of the identification process, a small bias current is applied to both windings, causing the magnetic levitation bearing to be statically suspended in the center position. The current in one of the windings is gradually increased, and the rotor's suspension position changes due to the differential control characteristics. Multiple operating conditions are created, and the inductance and resistance parameters of each degree of freedom winding are identified in real time as the rotor position changes.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1 to 7.

Images