Magnetic suspension motor rotor position error self-correction method based on maximum efficiency
By shaping the bus current of the magnetic levitation motor and iteratively adjusting the rotor position error correction angle, the problems of rotor position detection accuracy and system reliability of existing magnetic levitation motors are solved, and efficient motor operation is achieved.
Patent Information
- Application Number
- CN202511126047.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-12
- Publication Date
- 2025-10-28
AI Technical Summary
Among the existing magnetic levitation motor rotor position detection methods, the open-loop correction method has poor accuracy, and the closed-loop correction method requires additional circuits, resulting in increased system power consumption and reduced reliability.
A self-correction method for rotor position error of magnetic levitation motor based on maximum efficiency is adopted. By acquiring the bus current for shaping, the rotor position error correction angle is iteratively adjusted using a composite control strategy and a disturbance observation method, so that the mean square value of the bus current converges to the minimum value, thereby realizing closed-loop feedback control.
It improves the accuracy of rotor position detection and the reliability of the system, reduces power consumption, and ensures that the motor operates efficiently under different operating conditions.
Smart Images
Figure CN120855971A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of motor rotor correction technology, specifically to a self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency. Background Technology
[0002] Magnetic levitation high-speed motors have been widely used in high-end fields such as aerospace, precision manufacturing, and energy systems due to their advantages of frictionless operation, high speed, and high efficiency. Accurate acquisition of real-time rotor position information is a crucial prerequisite for achieving high-performance control of magnetic levitation high-speed motors. In many applications, to reduce costs, decrease size, and improve system reliability, sensorless control schemes are typically employed, which indirectly estimate the rotor position by detecting electrical signals such as the motor's back electromotive force.
[0003] In rotor position detection methods for sensorless motors, low-pass filters are typically used to suppress switching noise and commutation freewheeling spikes. However, filtering and hardware / software delays introduce phase errors into the rotor position detection signal, leading to commutation inaccuracies and reduced motor efficiency. Traditional open-loop correction methods, such as lookup table compensation and transfer function calculation, have poor accuracy and cannot return to accurate commutation when motor parameters change or are disturbed. In contrast, closed-loop correction methods offer high accuracy and strong anti-interference capabilities, making them a research hotspot in recent years. While closed-loop correction methods can achieve high-precision, dynamic, and accurate compensation for commutation errors, they generally require the design of voltage inductance circuits and filtering circuits to build the error closed-loop correction system. These additional circuits increase system power consumption and reduce reliability. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a self-correction method for rotor position error of magnetic levitation motors based on maximum efficiency. This method solves the problems of poor accuracy in existing open-loop correction methods such as lookup table compensation and transfer function calculation, which cannot return to the accurate commutation state when motor parameters change or are disturbed. Closed-loop correction methods require the design of voltage inductance circuits, filter circuits, etc. to build an error closed-loop correction system. These additional circuits increase system power consumption and reduce reliability.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency, comprising the following steps:
[0006] Step a: Obtain the bus current of the magnetic levitation motor;
[0007] Step b: Shape the bus current to suppress periodic fluctuations in the bus current;
[0008] Step c: Using the mean square value of the bus current after the shaping process as the optimization target and the rotor position error correction angle as the control variable, the mean square value of the bus current is converged to the minimum value by iteratively adjusting the rotor position error correction angle, thereby completing the phase correction of the rotor position detection signal.
[0009] In one specific embodiment, the step b, which involves shaping the bus current, is implemented using a composite control strategy consisting of a proportional-integral controller and a repetitive controller operating in parallel. This composite control strategy performs closed-loop control on the bus current to ensure that the bus current tracks a stable ideal current reference value.
[0010] Preferably, the repetitive controller includes a periodic delay element and a robustness coefficient. The periodic delay element is used to memorize the error information of the previous electrical cycle; the robustness coefficient is used to suppress periodic fluctuations while ensuring system stability.
[0011] Furthermore, the proportional-integral controller is used in the composite control strategy to achieve rapid dynamic response adjustment and steady-state error elimination of the bus current.
[0012] In one specific embodiment, the step of iteratively adjusting the rotor position error correction angle in step c is implemented based on the perturbation observation method.
[0013] Preferably, the disturbance observation method includes: superimposing or subtracting a preset angle increment on the current rotor position error correction angle, thereby applying an angle disturbance to the system.
[0014] Furthermore, the disturbance observation method also includes: comparing the current mean square value of the bus current after the angle disturbance is applied with the mean square value of the previous bus current before the angle disturbance is applied; if the current mean square value of the bus current decreases, then the next adjustment direction is determined to be consistent with the current disturbance direction; if the current mean square value of the bus current increases, then the next adjustment direction is determined to be opposite to the current disturbance direction.
[0015] In one embodiment, the transfer function G of the proportional-integral controller f1 The formula for expressing it is:
[0016]
[0017] In the formula, k p For proportional gain, k i Let s be the integral gain, and s be the Laplace operator.
[0018] In one embodiment, the composite control strategy, consisting of a proportional-integral controller and a repetitive controller operating in parallel, is implemented through a composite controller whose transfer function G... f2The formula for expressing it is:
[0019]
[0020] In the formula, T s This is the sampling period for the control system.
[0021] In one specific embodiment, the next adjustment direction is achieved by updating the rotor position error correction angle, specifically including: if the current bus current mean square value is less than the previous bus current mean square value, then the disturbance direction is maintained, specifically including:
[0022] When the previous adjustment is increased, the formula for calculating the rotor position error correction angle at the next moment is:
[0023]
[0024] When the previous adjustment is a decrease, the calculation formula is:
[0025]
[0026] If the current mean square value of the bus current is greater than the previous mean square value of the bus current, then the disturbance direction is reversed;
[0027] When the previous adjustment is increased, the formula for calculating the rotor position error correction angle at the next moment is:
[0028]
[0029] When the previous adjustment is a decrease, the calculation formula is:
[0030]
[0031] In the formula, The rotor position error correction angle for the next moment k+1. The rotor position error correction angle at the current time k. This is the preset angle increment.
[0032] This invention provides a self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency. It has the following beneficial effects:
[0033] 1. This invention uses the perturbation observation method to iteratively adjust the rotor position error correction angle, thus forming a closed-loop feedback control system. The actual bus current can be used as a feedback signal and compared with the applied rotor position error compensation effect in real time, thereby dynamically adjusting the correction amount. Therefore, it can adapt to parameter changes caused by motor temperature rise or wear, as well as fluctuations in external load. Therefore, compared with the open-loop correction method that relies on fixed parameters, it has higher correction accuracy.
[0034] 2. This invention actively applies minute angular disturbances and observes the system response, i.e., the increase or decrease of the mean square value of the bus current, to autonomously judge and determine the next adjustment direction. In this way, it can automatically search and converge to the optimal compensation angle, avoiding the complicated process of offline calibration or manual setting of compensation parameter tables in traditional methods.
[0035] 3. The present invention aims to stabilize the mean square value of the bus current at its minimum value as the ultimate control objective. Since this directly corresponds to the physical principle of maximizing motor operating efficiency, and under the premise of constant motor load, the power loss of the motor is proportional to the square of the drive current. Therefore, by driving the bus current to approach and stabilize at its minimum value, the power loss of the drive motor is minimized, thereby enabling the motor to operate at its highest efficiency point under the current operating conditions. Attached Figure Description
[0036] Figure 1 This is a schematic diagram of the method flow of the present invention;
[0037] Figure 2 This is a flowchart of the minimum bus current tracking algorithm of the present invention;
[0038] Figure 3 This is a schematic diagram of the parallel control scheme of repetitive control and proportional-integral control of the present invention;
[0039] Figure 4 This is a block diagram of the commutation error self-correction method based on maximum efficiency tracking control of the present invention; Detailed Implementation
[0040] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0041] To better understand the present invention, the above content will be described in detail below with reference to specific embodiments.
[0042] Please see the appendix Figure 1 - Appendix Figure 3 This invention provides a self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency. The method includes:
[0043] Step a: Obtain the bus current of the magnetic levitation motor. Specifically, the bus current signal is acquired in real time by a current sensor installed on the DC bus side of the motor drive inverter and transmitted to the controller. In addition to the DC component related to the rotor position error, the raw bus current signal also includes a periodic fluctuation component caused by inverter switching operations and motor electromagnetic torque pulsation.
[0044] Next, step b is performed: shaping the bus current. The purpose of this step is to suppress the aforementioned periodic fluctuations to avoid interfering with the subsequent minimum value optimization process. In this embodiment, this shaping process is implemented through a composite control strategy, which consists of a proportional-integral (PI) controller and a repetitive controller (RC) operating in parallel. The PI controller ensures a fast dynamic response to current commands and eliminates steady-state errors; the repetitive controller utilizes its internal periodic delay element to precisely suppress periodic harmonics of specific frequencies. Together, they ensure that the actual bus current close-loop tracks a stable ideal current reference value, thereby outputting a shaped bus current signal with filtered periodic fluctuations.
[0045] Finally, step c is executed: The mean square value of the shaped bus current is used as the optimization target for iterative adjustment. The physical principle behind this step is that under constant load, the larger the phase error of the rotor position detection, the less accurate the motor commutation, resulting in more ineffective power and thus a larger bus current. Therefore, the minimum point of the bus current corresponds to the point where the phase error is zero or minimal.
[0046] Furthermore, this step specifically employs the perturbation-observation method to achieve this optimization process. First, at the current rotor position error correction angle... Based on this, apply a preset small angle increment. As a disturbance, the mean square value of the bus current after shaping in step b is calculated over one or more electrical cycles after the disturbance is applied. This mean square value is then compared to the previous mean square value before the disturbance was applied: if the mean square value decreases, the disturbance direction is correct, and the adjustment in the next time step will maintain this direction; if the mean square value increases, the disturbance direction is incorrect, and the adjustment in the next time step will reverse the direction. This iterative process continues until the mean square value of the bus current converges and stabilizes at its minimum value. The corresponding rotor position error correction angle at this point is the optimal compensation value, thus completing the phase correction of the rotor position detection signal.
[0047] Please see the appendix Figure 2 , Figure 2 In the diagram, r represents the ideal steady current, y represents the bus current output by the motor, e represents the error signal, and z represents the error signal. -NFor a periodic delay element, N is the number of samples within one repetitive control cycle, Q(z) is the filter or a constant less than 1, and k p and k i These are the proportional and derivative coefficients, respectively, and M represents the controlled motor system. This is used to control the current waveform, making the bus current stable and eliminating interference by controlling it to its minimum value.
[0048] In a specific embodiment, the acquired bus current first needs to be shaped to suppress periodic fluctuations. Since these periodic fluctuations mainly originate from the inverter's switching actions and the motor's torque pulsation, if left untreated, they will be superimposed on the current changes caused by rotor position errors, thus interfering with the subsequent minimum-based optimization process and affecting the accuracy of the correction results.
[0049] To suppress this periodic fluctuation, this embodiment employs a composite control strategy consisting of a proportional-integral controller and a repetitive controller operating in parallel. This strategy uses the shaped bus current as a feedback signal and a stable ideal current reference value as an input command, forming a closed-loop control system. Through this closed-loop control, the actual bus current tracks the ideal current reference value, thereby outputting a shaped bus current signal with filtered periodic fluctuations.
[0050] In this composite control strategy, a proportional-integral controller is used to achieve rapid dynamic response adjustment of the bus current and elimination of steady-state errors. Its transfer function G... f1 The formula for expressing it is:
[0051]
[0052] In the formula, k p For proportional gain, k i Let s be the integral gain, and s be the Laplace operator.
[0053] In this composite control strategy, a repetitive controller is used to specifically suppress periodic harmonic components in the bus current. The repetitive controller includes a periodic delay element and a robustness coefficient. The periodic delay element memorizes error information from the previous electrical cycle, enabling the controller to accurately compensate for disturbances with a fixed period. The robustness coefficient adjusts the suppression strength of periodic fluctuations while ensuring system stability.
[0054] This composite control strategy, consisting of a proportional-integral controller and a repetitive controller operating in parallel, is implemented through a composite controller, whose overall transfer function G... f2 The formula for expressing it is:
[0055]
[0056] In the formula, T s The sampling period of the control system is given by the formula, where the coefficient 0.97 is a specific value of the robustness coefficient.
[0057] The composite controller takes the error between the ideal current reference value and the actual bus current as input and outputs a control signal to the inverter to complete the closed-loop control of the bus current shaping.
[0058] In one embodiment of the present invention, after obtaining the shaped bus current through a composite control strategy, the subsequent rotor position error self-correction step is based on the inherent correlation between the shaped bus current and the motor operating efficiency. Under constant motor load conditions, the total input power of the motor is proportional to the current absorbed from the DC bus. The aforementioned shaping process filters out high-frequency periodic fluctuations, allowing the mean or mean square value of the bus current to more accurately reflect the current changes caused by rotor position errors.
[0059] When there is a phase error in the rotor position detection signal, the commutation timing of the motor inverter will deviate from the ideal position. This inaccurate commutation will cause the spatial angular relationship between the stator magnetic field and the rotor magnetic field to deviate from the optimal state, thereby generating additional reactive current components or harmonic current components.
[0060] These additional current components do not contribute to the generation of effective torque, but they generate additional power losses as they flow through the motor windings and inverter power devices, primarily manifested as increased copper losses. To overcome these additional losses while maintaining the same effective torque output (i.e., constant load), the motor must draw a larger current from the DC bus.
[0061] Therefore, the amplitude of the bus current after shaping and the phase error of the rotor position detection signal exhibit a functional relationship:
[0062] When the phase error is zero, the motor commutation is accurate, the ineffective loss is minimal, and the bus current reaches its theoretical minimum value. As the absolute value of the phase error increases, the ineffective loss increases, and the bus current also increases accordingly.
[0063] Based on this principle, by finding and stabilizing the minimum point of the shaped bus current, the optimal operating point for the rotor position detection signal without phase error can be determined in reverse. This embodiment of the invention utilizes this characteristic, using the mean square value of the shaped bus current as the optimization target, and iteratively adjusting the rotor position error correction angle to ultimately achieve precise correction of the phase error, thereby enabling the motor to operate at its highest efficiency.
[0064] Specifically include:
[0065] The minimum bus current tracking control method was adopted in the process of iteratively adjusting the rotor position error correction angle. This method uses the rotor position error correction angle as the reference value. To control the quantity, the goal is to minimize the mean square value of the shaped bus current i, which is achieved through the perturbation-observation method. The execution steps of this method are detailed in the appendix. Figure 3 .
[0066] In this method, the number of measurement data p used for each calculation of the mean square value of the bus current is first defined, as well as the angle increment used for the iterative calculation of the correction angle. In the k-th iteration cycle, the controller collects p data points of the shaped bus current i and calculates its mean square value, denoted as E[i]. 2 (k)]. This mean square value is compared with the mean square value E[i] calculated in the previous iteration cycle (the (k-1)th cycle). 2 The adjustment direction of the current cycle is determined by comparing (k-1) and combining it with the adjustment action of the previous cycle.
[0067] The specific iterative adjustment logic is as follows:
[0068] 1. When the previous adjustment action was to increase the rotor position error correction angle, and the current bus current mean square value is less than the previous bus current mean square value (E[i 2 (k)] <E[i 2 (k-1) indicates that the adjustment direction is correct, and the system continues to increase the correction angle. The rotor position error correction angle at the next moment. The calculation formula is:
[0069]
[0070] 2. When the previous adjustment action was to reduce the rotor position error correction angle, and the current bus current mean square value is less than the previous bus current mean square value (E[i 2 (k)] <E[i 2 (k-1) indicates that the adjustment direction is correct, and the system continues to reduce the correction angle. The rotor position error correction angle at the next moment. The calculation formula is:
[0071]
[0072] 3. When the previous adjustment action increased the rotor position error correction angle, and the current bus current mean square value is greater than or equal to the previous bus current mean square value (E[i 2 (k)]≥E[i 2 (k-1) indicates an incorrect adjustment direction. The system reverses the adjustment direction and begins to reduce the correction angle. The rotor position error correction angle at the next moment... The calculation formula is:
[0073]
[0074] 4. When the previous adjustment action was to reduce the rotor position error correction angle, and the current bus current mean square value is greater than or equal to the previous bus current mean square value (E[i 2 (k)]≥E[i 2 (k-1) indicates an incorrect adjustment direction. The system reverses the adjustment direction and begins to increase the correction angle. The rotor position error correction angle at the next moment... The calculation formula is:
[0075]
[0076] In the formula, The rotor position error correction angle for the next moment k+1. The rotor position error correction angle at the current time k. This is the preset angle increment.
[0077] The controller repeatedly executes the above iterative adjustment process until the mean square value of the bus current E[i] is reached. 2 [k] converges and stabilizes near its minimum value. At this point, the corresponding rotor position error correction angle... This is the optimal compensation value, thus completing the self-correction of rotor position detection error.
[0078] Please refer to Figure 4 The present invention also provides another embodiment;
[0079] In another embodiment, the PI controller for speed regulation receives an externally set given speed and a feedback-acquired actual motor speed. The PI controller outputs a current command based on the difference between the two.
[0080] The composite controller receives the current command and the actual bus current obtained through feedback. Specifically, the difference between the current command and the actual bus current is simultaneously input to both the PI controller and the repetitive controller. The outputs of both are superimposed and used to adjust the system drive so that the actual bus current tracks the current command; this process achieves the shaping of the bus current.
[0081] Simultaneously, the actual bus current obtained from the feedback is also sent to the minimum bus current tracking control module. This module, based on the perturbation observation method, performs iterative calculations with the goal of finding the minimum bus current and outputs a correction angle.
[0082] The correction angle is superimposed on an original rotor position detection signal to generate a phase-corrected rotor position signal.
[0083] The commutation logic module receives the corrected rotor position signal and generates control signals for controlling the switching of the inverter power devices.
[0084] The inverter receives control signals from the commutation logic module and supplies power to the magnetic levitation motor (M) accordingly, thereby completing closed-loop control of the motor's operation. Through this system, the rotor position detection error is corrected in real time, enabling the motor to operate at high efficiency.
Claims
1. A self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency, characterized in that, Includes the following steps: Step a: Obtain the bus current of the magnetic levitation motor; Step b: Shape the bus current to suppress periodic fluctuations in the bus current; Step c: Using the mean square value of the bus current after the shaping process as the optimization target and the rotor position error correction angle as the control variable, the mean square value of the bus current is converged to the minimum value by iteratively adjusting the rotor position error correction angle, thereby completing the phase correction of the rotor position detection signal.
2. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 1, characterized in that, Step b, specifically the step of shaping the bus current, includes: A composite control strategy consisting of a proportional-integral controller and a repetitive controller in parallel is adopted to perform closed-loop control on the bus current, so that the bus current tracks the ideal current reference value smoothly.
3. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 2, characterized in that, The repetitive controller includes a periodic delay element and a robustness coefficient; The periodic delay element is used to memorize the error information of the previous electrical cycle, and the robustness coefficient is used to suppress periodic fluctuations.
4. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 2, characterized in that, The proportional-integral controller is used in the composite control strategy to rapidly adjust the dynamic response of the bus current and eliminate steady-state errors.
5. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 1, characterized in that, In step c, the rotor position error correction angle is adjusted iteratively based on the disturbance observation method.
6. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 5, characterized in that, The perturbation observation method includes: A preset angle increment is added to or subtracted from the current rotor position error correction angle to apply an angle disturbance to the system.
7. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 5, characterized in that, The perturbation observation method further includes: Compare the current mean square value of the bus current after applying the angle disturbance with the mean square value of the previous bus current before applying the angle disturbance; If the current mean square value of the bus current decreases, the next adjustment direction will be consistent with the direction of this disturbance. If the current mean square value of the bus current increases, the next adjustment direction will be opposite to the current disturbance direction.
8. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 2, characterized in that, The proportional-integral controller transfer function G f1 The formula for expressing it is: In the formula, k p For proportional gain, k i Let s be the integral gain, and s be the Laplace operator.
9. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 8, characterized in that, The composite control strategy, consisting of a proportional-integral controller and a repetitive controller operating in parallel, is implemented through the composite controller, whose transfer function G... f2 The formulas for expression include: In the formula, T s This is the sampling period for the control system.
10. The self-correction method for rotor position error of a magnetic levitation motor based on maximum efficiency according to claim 7, characterized in that, The next adjustment direction is achieved based on updating the rotor position error correction angle, specifically including: If the current mean square value of the bus current is less than the previous mean square value of the bus current, the disturbance direction is maintained. If the previous adjustment was to increase, the formula for calculating the rotor position error correction angle at the next moment is: If the previous adjustment is a decrease, then the calculation formula is: If the current mean square value of the bus current is greater than the previous mean square value of the bus current, then the disturbance direction is reversed. If the previous adjustment was to increase, then the formula for calculating the rotor position error correction angle at the next moment is: If the previous adjustment is a decrease, then the calculation formula is: In the formula, The rotor position error correction angle for the next moment k+1. The rotor position error correction angle at the current time k. This is the preset angle increment.