A method for suppressing torque ripple in a dual salient pole motor with optimal current vector

Abstract

This invention discloses a torque ripple suppression method for the optimal current vector of a doubly salient pole motor. Starting from the idea of ​​controlling instantaneous torque, it achieves this by controlling... i q This invention discloses an optimal current vector control strategy based on a proposed instantaneous torque observer and lookup table method. This strategy calculates the optimal current vector and combines it with... i d =0 control method, which achieves current tracking through hysteresis switch, so that the motor works in the desired state.

Description

Technical Field

[0001] This invention relates to the field of motor control, and in particular to a method for suppressing torque ripple in a doubly salient pole motor. Background Technology

[0002] Electrically excited doubly salient pole motors (DSPs) have broad application prospects in aviation starting / generating fields due to their simple structure, high reliability, and flexible and convenient control. A DSP is a typical variable reluctance motor, with both its stator and rotor having salient pole structures, maximizing the ratio of maximum to minimum reluctance and exhibiting excellent electromechanical energy conversion characteristics. DSPs have a simple structure and manufacturing process, with no windings on the rotor, resulting in stable and reliable motor operation. By adjusting the excitation current, the air gap magnetic field of an electrically excited DSP can be changed. When combined with an inverter topology, the resulting motor system can be used in brushless DC motor applications. Currently, a three-phase full-bridge inverter topology is commonly used, with the DSP phase windings connected in a star configuration and directly connected to the diode inverter topology.

[0003] In recent years, torque ripple and its suppression technology in DSEM have been a research hotspot. Torque ripple is the root cause of motor vibration, noise and speed fluctuation, which limits the application scope of DSEM. Summary of the Invention

[0004] This invention proposes a current vector injection method based on torque ripple in DSEM, using i d The control method with a value of 0 achieves maximum torque-to-current ratio control by changing i in real time. q The magnitude (with two dimensions) is used to control the instantaneous torque, injecting the optimal current vector into the motor to keep the motor in a constant torque operating state, thereby reducing the torque ripple of the motor.

[0005] The technical solution of this invention is as follows:

[0006] Step 1, define the ratio of phase torque to phase current as the phase torque-current coefficient t. k (θ,i k i f Meanwhile, the average phase torque current coefficient t is defined. kave (θ,i f The three-phase average phase torque current coefficient t under different excitation currents is obtained offline. a t b t c Establish a DSEM instantaneous torque calculation model;

[0007] Step 2: Transform the three-phase average torque current coefficient and the three-phase armature current to a two-phase stationary coordinate system. On the α-β plane of the two-phase coordinate system, the (two-phase stationary) torque current coefficient vector t α t β and current vector iα i β The (rotational) torque current coefficient vector t and the current vector i can be synthesized respectively;

[0008] Step 3: Obtain the real-time three-phase average phase torque current coefficients using the Clarke transform and Park transform methods to obtain the real-time torque current coefficient vector t, and obtain the magnitude |t| of vector t. Based on the instantaneous torque calculation model, given the torque T... ref Thus, the corresponding given three-phase armature current amplitude |i| is obtained;

[0009] Step 4, using direct-axis current i d The control strategy of 0 ensures that the motor operates at the maximum torque-to-current ratio. It calculates the current vector of the motor output constant torque and injects the optimal current vector into the DSEM so that the DSEM can operate in a state with low torque ripple.

[0010] Preferably, the phase torque current coefficient t in step 1 k It concerns the rotor position θ and the phase current i k and excitation current i f The function of the average phase torque current coefficient t kave It concerns the rotor position θ and the excitation current i f The function.

[0011] Preferably, the instantaneous torque model described in step 1 is:

[0012] T esti =t a ·i a +t b ·i b +t c ·i c +t f ·i f

[0013] Among them, T esti The instantaneous torque of the motor is t. a t b t c i represents the three-phase average phase torque current coefficient under different excitation currents. a i b i c For the three-phase armature current, i f For the excitation current, t f This is the excitation torque current coefficient vector.

[0014] Preferably, the Clarke transform and Parker transform methods in step 2 are as follows: the three-phase average torque current coefficient and the three-phase armature current are transformed to the two-phase coordinate system α-β plane, and the torque current coefficient vector t in the two-phase stationary coordinate system is... α t β and current vector i α i β The torque current coefficient vector t and the current vector i in the rotating coordinate system are synthesized respectively.

[0015] Preferably, the one-dimensional data table is established based on the region of electrical angles corresponding to one revolution of the double salient pole motor. The back EMF of the DSEM is obtained through offline measurement, and then the torque current coefficient vector position corresponding to the rotor position is obtained through coordinate transformation and arctangent calculation. The torque characteristic diagram is obtained through electromagnetic simulation of the motor model, and the corresponding three-phase torque current coefficient vector and excitation torque current coefficient vector can be obtained from the torque position.

[0016] Preferably, the given torque T is obtained in step 3. ref The method is as follows: Differentiate the torque current coefficient vector position obtained from the table to obtain the torque current coefficient vector rotational speed. Compare this speed with the reference speed and input it into the speed loop for dynamic adjustment, then output the given torque T. ref .

[0017] Preferably, the given current vector magnitude is expressed as: In the formula These represent the angular positions of the torque current coefficient vector and the current vector, respectively.

[0018] Preferably, the input current vector and the torque current coefficient vector need to be collinear to obtain the optimal current vector that enables the motor to output constant torque, satisfying the following conditions. In the formula These represent the angular positions of the torque current coefficient vector and the current vector, respectively.

[0019] As a preferred option, i d The control method with a value of 0 suppresses the odd and even harmonics of the torque current coefficient. Specifically, it sets the direct-axis current i to 0. d =0, calculate the optimal current vector i q The three-phase reference current is obtained by combining the vector position relationship of torque and current coefficients and inverse transformation calculation. The current tracking is achieved through the hysteresis switch, and the output hysteresis control quantity controls the inverter switching state through 2 / 3 transformation. Attached Figure Description

[0020] Figure 1 It is a DSEM drive circuit;

[0021] Figure 2These are the DSEM torque-current coefficient and the ideal drive current waveform;

[0022] Figure 3 It is the trajectory of the rotation of the torque current coefficient vector and the current vector in a two-phase stationary coordinate system;

[0023] Figure 4 This is the block diagram of the DSEM optimal current vector control system;

[0024] Figure 5 These are the three-phase current and output torque waveforms of the motor obtained from simulation under optimal current vector control;

[0025] Figure 6 It is the current trajectory obtained from simulation under optimal current vector control.

[0026] Beneficial effects:

[0027] This invention solves the problem of large torque ripple in electrically excited doubly salient pole motors. Simulation results show that as the output torque increases, the peak-to-peak torque changes very little, and the torque ripple gradually decreases with increasing output torque. Existing methods for suppressing torque ripple in electrically excited doubly salient pole motors are limited; most common methods rely on angle control. This invention, however, effectively suppresses torque ripple by injecting an optimal spatial current vector. Detailed Implementation

[0028] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings:

[0029] This invention can be implemented in many different forms and should not be considered limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully express the scope of the invention to those skilled in the art. In the drawings, components are enlarged for clarity.

[0030] During operation, permanent magnet synchronous motors exhibit coupling between the direct and quadrature axis voltages; that is, changes in the d-axis parameters cause changes in the q-axis parameters, which is detrimental to motor control. Vector control... d =0 control is essentially about achieving static decoupling of the current along the dq axis, i d When the magnetic flux is 0, it is entirely supplied by the permanent magnet. The motor has no armature reaction along the direct shaft, and the current along the direct shaft is zero. At this time, all the current in the motor is used to generate electromagnetic torque; only i needs to be controlled. q The value of this value can control the motor torque, thus achieving static decoupling of the motor.

[0031] Practice has proven that to suppress motor torque pulsation, it is necessary to start by controlling instantaneous torque. This invention, based on the idea of ​​controlling instantaneous torque, employs i... d =0 control method, by controlling iq By controlling the torque, maximum torque-to-current ratio control can be achieved, thereby enabling the motor to operate under constant torque and suppressing torque pulsation.

[0032] Torque is mainly related to i q Related, if i can be changed in real time q The magnitude of i can control the instantaneous torque in real time, which requires determining a suitable i. q To achieve constant torque control, the present invention provides an optimal current vector torque ripple suppression method for a doubly salient pole motor, comprising a full-bridge inverter, an electrically excited doubly salient pole motor, and a position sensor connected in sequence. The full-bridge inverter is used to drive the motor, and the position sensor is used to obtain the rotor position of the motor.

[0033] First, establish the instantaneous torque calculation model:

[0034] The ratio of phase torque to phase current is defined as the phase torque-current coefficient t. k (θ,i k i f This coefficient is related to the rotor position θ and the phase current i. k and excitation current i f The phase torque current coefficient is a function of the magnetic circuit saturation. Due to the influence of magnetic circuit saturation, the phase torque current coefficient under different phase currents exhibits slight differences due to varying degrees of magnetic circuit saturation. The average phase torque current coefficient t can be used. kave (θ,i f This coefficient is related to the rotor position θ and the excitation current i. f The function is used. The average phase torque current coefficient reduces the dimensionality of the independent variables; this coefficient is only a function of rotor position and excitation current.

[0035] Due to the asymmetry of the three-phase electromagnetic characteristics of the electrically excited doubly salient pole motor, it is necessary to obtain the three-phase average phase torque current coefficient t under different excitation currents offline. a t b t c Instantaneous torque T of an electrically excited doubly salient pole motor esti It can be estimated by the following formula:

[0036] T esti =t a ·i a +t b ·i b +t c ·i c +t f ·i f (1)

[0037] The DSEM armature winding uses a star connection, therefore it is often driven by a three-phase full-bridge inverter circuit. The equivalent circuit is as follows: Figure 1As shown. Where, U dc L is the DC side voltage. x For the sake of self-awareness, e x Let r be the phase induced electromotive force, r be the phase winding resistance, and N represent the motor neutral point. Unlike permanent magnet motors, this invention neglects motor magnetic circuit saturation and eddy current hysteresis losses. The voltage equation satisfied by each phase is:

[0038]

[0039] The back EMF waveform of an electrically excited doubly salient pole motor is an irregular trapezoidal wave. It is usually controlled by a six-state square wave current chopper similar to that of a permanent magnet brushless DC motor, with the only difference being the division of the conduction region. Figure 2 The DSEM back EMF waveform and ideal drive current waveform are presented. Due to the asymmetry of the back EMF, the torque current coefficient vector and the quadrature axis are not collinear. The position of maximum flux linkage in phase A winding (back EMF is zero) is defined as the initial zero position. To represent the relationship between the electrical angle and the torque current vector, based on the characteristics of the DSEM back EMF waveform, the electrical angle of motor operation is divided into 6 sectors, labeled with Roman numerals I to VI. Due to the asymmetry of electromagnetic characteristics, the width of each sector is different.

[0040] Based on the motor operating conditions, the rotor position and the three-phase average phase torque current coefficient t are established. a t b t c A one-dimensional data table showing the relationship between rotor position and torque current coefficient position is provided. This table is based on the electrical angle region corresponding to one revolution of a doubly salient pole motor, and the tool used is the MATLAB 2-D module. The DSEM back EMF is obtained through offline measurement, and then the torque current coefficient vector positions corresponding to the rotor positions are obtained through coordinate transformation and arctangent calculation. The three-phase average torque current coefficient vector can be obtained by obtaining the torque characteristic diagram through electromagnetic simulation of the motor model, from which the corresponding three-phase torque current coefficient vector and excitation torque current coefficient vector can be obtained from the torque position.

[0041] The real-time rotor position of the motor is obtained based on the position sensor using a lookup table method. Based on this rotor position, the real-time three-phase torque current coefficient t is then obtained using the lookup table method. a ,t b ,t c and three-phase armature current i a i b i c According to the theory of instantaneous reactive power in three-phase circuits, the three-phase average torque current coefficient (t) is... a ,t b ,t c ), three-phase armature current (i a ib i c Transformation from a three-phase coordinate system to a two-phase stationary coordinate system:

[0042]

[0043]

[0044] Where C 32 For the transformation matrix, in the α-β plane of the two-phase coordinate system, the torque current coefficient vector t α t β and current vector i α i β The (rotational) torque current coefficient vector t and the current vector i can be synthesized respectively.

[0045]

[0046]

[0047] In the formula, |t| and |i| are the magnitudes of vectors t and i; These are the arguments of the vectors.

[0048] Figure 3 The trajectories of the torque current coefficient vector t and the current vector i rotating in a two-phase stationary coordinate system are given. Due to the presence of a large number of odd and even harmonics in the torque current coefficient, the waveform is severely distorted, and its trajectory is an irregular "hexagonal star". The initial position of the torque current coefficient vector does not coincide with the rotor zero position, but is located on the positive half-axis of the β axis, with a 90° phase difference between the two.

[0049] The instantaneous torque of a three-phase system is equal to the vector product of the torque coefficient vector t and the current vector i.

[0050] Differentiating the position of the torque current coefficient vector obtained from the table yields the rotor speed of the torque current coefficient vector. This speed is then compared with a reference speed and input to the speed loop for dynamic adjustment, outputting a given torque T. ref .

[0051] A given torque can be expressed using current and torque coefficient.

[0052]

[0053] Given a given torque, the magnitude of the corresponding given current vector can be calculated.

[0054]

[0055] If the current vector and the torque current coefficient vector can be guaranteed to run collinearly, then the optimal current vector that enables the motor to output a constant torque can be obtained, satisfying the following condition:

[0056]

[0057] According to the injected current vector in equation (9), the DSEM can operate in a state without torque ripple. Based on this, a drive control system designed is as follows: Figure 4 As shown. The system uses i d =0 control to obtain the maximum torque-to-current ratio, the speed loop output is used as the torque command, and the quadrature shaft reference current i is calculated according to equation (9). q * The required torque current coefficient is obtained by looking up a table, and after CLARK transformation, the magnitude of the torque current coefficient vector is obtained by modulo calculation.

[0058] Quadrature axis reference current i q * The three-phase reference current is obtained by combining the positional relationship of the torque and current coefficient vectors and inverse transformation calculation. Current tracking is achieved through hysteresis switch to ensure that the motor operates in the state of maximum torque-to-current ratio. The current vector of the motor output constant torque is calculated, and the optimal current vector is injected into the DSEM so that the DSEM can operate in a state with low torque ripple.

[0059] Figure 5 The waveforms of the three-phase current and output torque of the motor under optimal current vector control are presented. Figure 5 (a) is the waveform with an average torque of 10 Nm, a peak-to-peak torque of 5.7 Nm, and a torque ripple of 57.6%, which is 78.6% lower than that of square wave current control. Figure 5 (b) shows the waveform with an average torque of 20 Nm, a peak-to-peak torque of 5.5 Nm, and a torque ripple of 28%, which is 79.5% lower than that of square wave current control. It can be seen that the optimal current vector control strategy significantly reduces the motor torque ripple.

[0060] Figure 6 The waveforms of the three-phase current trajectory of the motor under optimal current vector control are given. It can be seen that the current trajectory under optimal current vector control is different from that under square wave current control, and it presents an irregular "clover" shape. Moreover, the shape of the current trajectory will also change as the output torque increases. This is because the amplitude of the current vector is related to the rotor position and the given torque, and needs to be changed continuously to ensure that the output torque is approximately constant.

[0061] Table 1 presents the simulation results of the two control strategies under different operating conditions. It can be seen that the torque ripple gradually decreases as the output torque increases. Due to the large current spikes and severe current waveform distortion, the optimal current vector control strategy performs worse than the square wave current control in terms of torque-to-current ratio and current THD. This indicates that suppressing torque ripple requires sacrificing a low torque-to-current ratio and high current THD.

[0062] Table 1 Comparison of simulation results of the two control strategies under different operating conditions

[0063]

[0064] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0065] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for suppressing torque ripple in a doubly salient pole motor based on the optimal current vector, characterized in that, Includes the following steps: Step 1, define the ratio of phase torque to phase current as the phase torque-current coefficient t. k (θ,i k i f This coefficient is related to the rotor position θ and the phase current i. k and excitation current i f The function, and simultaneously defines the average phase torque current coefficient t. kave (θ,i f This coefficient is related to the rotor position θ and the excitation current i. f The function; offline acquisition of the three-phase average phase torque current coefficient t under different excitation currents. a t b t c Establish a DSEM instantaneous torque calculation model T esti ; Step 2, establish the motor's rotor position and the three-phase average phase torque current coefficient t. a t b t c A one-dimensional data table showing the relationship between the rotor position and the position of the torque current coefficient is used. The real-time rotor position of the motor is obtained through a position sensor, and the real-time three-phase average phase torque current coefficient and torque current coefficient vector position are obtained through a lookup table method. Step 3: Obtain the real-time three-phase average phase torque current coefficients using the Clarke transform and Park transform methods to obtain the real-time torque current coefficient vector t, and obtain the magnitude |t| of vector t. Based on the instantaneous torque calculation model, given the torque T... ref Specifically, the torque current coefficient vector position obtained from the table is differentiated to obtain the rotational speed of the torque current coefficient vector. After comparing it with the reference speed, the speed is input to the speed loop for dynamic adjustment, and the given torque T is output. ref Thus, the corresponding given three-phase armature current amplitude |i| is obtained; Step 4, based on |t| and |i|, use the direct-axis current i d The control strategy with a value of 0 calculates the quadrature axis reference current i. q * The value of the reference current is calculated, and the current tracking is achieved through the hysteresis current regulator to ensure that the motor operates in the maximum torque-to-current ratio state. The current vector of the motor output constant torque is calculated, and the optimal current vector is injected into the DSEM.

2. The optimal current vector torque ripple suppression method for a doubly salient pole motor according to claim 1, characterized in that, The instantaneous torque calculation model mentioned in step 1 is as follows: T esti =t a ·i a +t b ·i b +t c ·i c +t f ·i f Among them, T esti The instantaneous torque of the motor is t. a t b t c i represents the three-phase average phase torque current coefficient under different excitation currents. a i b i c For the three-phase armature current, i f For the excitation current, t f This is the excitation torque current coefficient vector.

3. The torque ripple suppression method for the optimal current vector of a doubly salient pole motor according to claim 2, characterized in that, The one-dimensional data table is established based on the region of electrical angles corresponding to one revolution of the double salient pole motor. The back EMF of the DSEM is obtained through offline measurement, and then the torque current coefficient vector position corresponding to the rotor position is obtained through coordinate transformation and arctangent calculation. The torque characteristic diagram is obtained through electromagnetic simulation of the motor model, and the corresponding three-phase torque current coefficient vector and excitation torque current coefficient vector can be obtained from the torque position.

4. The torque ripple suppression method for the optimal current vector of a doubly salient pole motor according to claim 3, characterized in that, Step 3, specifically the Clarke transform and Parker transform methods, involves transforming the three-phase average torque current coefficient and the three-phase armature current onto the two-phase coordinate system α-β plane. The torque current coefficient vector t in the two-phase stationary coordinate system... α t β and current vector i α i β The torque current coefficient vector t and the current vector i in the rotating coordinate system are synthesized respectively.

5. The method for suppressing torque ripple in the optimal current vector of a doubly salient pole motor according to claim 2, characterized in that, The magnitude of a given current vector is expressed as: In the formula These represent the angular positions of the torque current coefficient vector and the current vector, respectively.

6. The torque ripple suppression method for the optimal current vector of a doubly salient pole motor according to claim 5, characterized in that, The input current vector simultaneously satisfies: At this point, the input current vector and the torque current coefficient vector need to be collinear to obtain the optimal current vector that enables the motor to output a constant torque.

7. The torque ripple suppression method for the optimal current vector of a doubly salient pole motor according to claim 6, characterized in that, Using i d The control method with a value of 0 suppresses the odd and even harmonics of the torque current coefficient. Specifically, it sets the direct-axis current i to 0. d =0, calculate the optimal current vector, i.e., the quadrature-axis reference current i. q * The three-phase reference current is obtained by combining the vector position relationship of torque and current coefficients and inverse transformation calculation. The inverter switching state is controlled by the hysteresis current regulator and 2 / 3 transformation to make the motor work in the desired state.

Images