CN114389486B - Commutation error compensation method for hybrid excitation doubly salient motor - Google Patents

Abstract

The invention discloses a commutation error compensation method for a hybrid excitation doubly salient motor, and belongs to the technical field of variable reluctance motor drive control. The method for compensating the commutation error of the hybrid excitation doubly salient motor provided by the invention is realized according to the following principle: when the current commutation point arrives, the sampling point of the controller is coincident with the commutation point, and thus accurate commutation can be realized to eliminate the commutation error of the hybrid excitation doubly salient motor. According to the invention, the phase-change error of the hybrid excitation doubly salient motor can be effectively reduced by adjusting the switching frequency by a small amplitude, the accurate phase-change of the hybrid excitation doubly salient motor is realized, the current capacity of a switching tube can be effectively utilized, the torque pulsation is reduced, and the torque output capacity is improved.

Description

A commutation error compensation method for hybrid excitation doubly salient motor

Technical Field

The invention belongs to the technical field of variable reluctance motor drive control.

Background technique

As a variable reluctance motor, the hybrid excitation double salient pole motor retains the stator-rotor structure of the switched reluctance motor on the basis of the switched reluctance motor, and introduces an additional excitation winding on the stator. Since there are no windings and permanent magnets on the rotor, the rotor structure is simple and reliable, and is naturally suitable for high-speed operation. However, there is currently no mature high-speed control algorithm for hybrid excitation double salient pole motors. In engineering practice, advance angle control or three-phase nine-state control methods are usually used to achieve high-speed operation of hybrid excitation double salient pole motors. Both control methods require commutation of the three-phase current, and whether the current can be accurately commutated has a great impact on the operating performance of the motor.

The commutation error is caused by the delay in the digital control system. The error is mainly divided into two categories: one is the error caused by the PWM update delay, and the other is the error caused by the angle position of discrete sampling. The first type of error is inherent in the PWM digital control system. The triangular wave generated by the digital processor counts as the carrier, which intersects with the modulation wave to generate the PWM drive signal. When the digital processor enters the interrupt, it will calculate the collected signal. If the update is performed immediately after the calculation is completed, the PWM signal may be missing due to the long calculation time. Therefore, the PWM signal will be updated only when the processor enters the next interrupt, which causes a delay of one switching cycle. The second type of error is caused by discrete sampling. Ideally, the sampling frequency is infinite and the collected signal is a continuous quantity. However, in reality, the sampling frequency of the controller is limited, and the collected signal is a discrete signal. The set commutation angle is a certain point. Due to the limitation of the sampling frequency, the system cannot accurately sample at this point, and the actual sampling point will fluctuate around the set point. The sampling frequency is usually equal to the switching frequency, so a commutation error of 0 to 1 switching cycle will be generated. Combining the above two types of errors, the total commutation error of the hybrid excitation doubly salient motor system is 1 to 2 switching cycles.

When the hybrid excitation double-salient-pole motor is running at low speed, the switching frequency is much greater than the motor's operating frequency, so the electrical angle corresponding to one switching cycle is very small, and the commutation error at low speed can be ignored. When the motor's operating speed increases, the electrical angle corresponding to one switching cycle cannot be ignored. If a higher switching frequency is required to keep the carrier ratio unchanged, the device cost and design difficulty will increase significantly, which is unrealistic. The commutation error can only be eliminated by not increasing the switching frequency.

Due to the existence of commutation error under high-speed operation, the three-phase current of the hybrid excitation double-salient-pole motor has low-frequency oscillation, the conduction interval width of each cycle is different, and the current amplitude is also unequal. On the one hand, the low-frequency oscillation of the current amplitude will cause the motor torque to produce corresponding pulsations; on the other hand, the oscillation of the current amplitude will cause the current amplitude to be small in some cycles and large in some cycles. Since the current stress of the switch tube should be selected according to the maximum current stress, the current capacity of the switch tube cannot be fully utilized. If the commutation error under high-speed operation can be eliminated, the current amplitude oscillation can be eliminated, the switch tube capacity can be fully utilized, and it will help to further increase the speed.

Summary of the invention

The hybrid excitation double-salient pole commutation error compensation method proposed in the present invention achieves the effect of eliminating the commutation error by slightly adjusting the switching frequency without changing the circuit topology.

In order to achieve the above object, the specific implementation steps of a commutation error compensation method for a hybrid excitation double-salient pole motor described in the present invention are as follows:

Step 1: Sample the current rotor position angle θ through the rotary transformer installed coaxially with the motor to determine whether it is in the following range, where the commutation angle is θ ₁ , the sampled position signal is θ, the switching frequency is f _s , the maximum switching frequency is nf _s , the electrical frequency is _fe , and the electrical angle corresponding to one switching cycle is θ _d = _fe *360/f _s :

(1)θ∈(θ ₁ -θ _d *2/n,θ ₁ -θ _d /n)

(2)θ∈(θ ₁ -(1+1/n)*θ _d ,θ ₁ -θ _d *2/n)

Step 2: If the current rotor position is not in the above range, it means that the current rotor position is far from the set commutation point, so no operation is required at this time, and the original switching frequency _fs can be kept unchanged.

Step 2: If the current rotor position is in the interval described in (1), calculate the angle difference between the current position angle θ and the commutation angle _θ1 , and set the appropriate switching frequency so that the position of the next sampling coincides with the commutation angle to ensure that the commutation point is accurately collected. Therefore, the set switching frequency is _fe *360/( _θ1 -θ).

Step 3: If the current rotor position is in the interval described in (2), directly set the switching frequency to the maximum switching frequency _nfs . After one switching cycle, at the next sampling point, the rotor position angle can be guaranteed to be in the interval described in (1).

Step 4: Accurately sample the commutation point position, perform calculations in the digital processor, and update the PWM signal to achieve the effect of eliminating the current commutation error.

The hybrid excitation double salient pole motor commutation error compensation method proposed in the present invention achieves the effect of eliminating the commutation error by adding a compensation algorithm to the control process. The compensation algorithm does not need to change the original hybrid excitation double salient pole motor system structure, and has the advantage of being easy to implement. The hybrid excitation double salient pole motor commutation error compensation method proposed in the present invention can effectively compensate for the commutation error of the hybrid excitation double salient pole motor system, eliminate the phase current oscillation and spikes caused by the commutation error, and reduce the current stress of the switch tube in the drive circuit, thereby effectively utilizing the current capacity of the switch tube. Furthermore, the hybrid excitation double salient pole motor commutation error compensation method proposed in the present invention is conducive to reducing the motor torque pulsation and improving the torque output capacity, and has a significant effect on improving the operating performance of the hybrid excitation double salient pole motor system.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a cross-sectional view of the structure of a three-phase 12/8-pole hybrid excitation doubly salient motor;

FIG2 is a three-phase full-bridge power converter topology for a hybrid excitation doubly salient motor;

Figure 3 shows the simplified inductance model of the hybrid excitation doubly salient motor and the standard angle control conduction mode;

FIG4 is a schematic diagram showing the principle of generation of the first type of commutation error;

FIG5 is a schematic diagram showing the principle of generation of the second type of commutation error;

FIG6 is a flow chart of the commutation error compensation algorithm.

Detailed ways

The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

As shown in Figure 1, it is a cross-sectional view of the structure of a three-phase 12/8 hybrid excitation double-pole motor, in which the excitation winding, armature winding and permanent magnet are all located on the stator, and there are no windings and permanent magnets on the rotor. Figure 2 shows the main power converter topology of the hybrid excitation double-pole motor, which is a three-phase full-bridge inverter. AX, BY and CZ are the armature windings of phase A, phase B and phase C, respectively, and are connected in star shape. The currents i _a , i _b and i _c are the corresponding phase currents, and the arrow direction is defined as the positive direction. E _a , E _b and E _c are the back electromotive force of the armature winding, respectively, and the "+" end is the end with the high back electromotive force in the winding. It can be seen that at this time, the current flows into the armature winding from the positive end of the back electromotive force, that is, the armature winding absorbs electrical energy, and the motor is in the electric working mode. V1~V6 are six switch tubes, of which V1, V3 and V5 are upper tubes, and V4, V6 and V2 are lower tubes. In the figure, D1~D6 are the body diodes of the switch tubes, which play the role of freewheeling in the phase current chopping control. In addition, in the figure, U _dc is the DC bus voltage, i _dc is the DC bus current, and C _f is the filter capacitor on the bus.

When current flows into one phase winding of a hybrid excitation doubly salient motor, the torque generated is:

Where _Tp is the total torque output of a single phase, _Tpr represents the single-phase reluctance torque, _Tpe represents the single-phase excitation torque, _Lp represents the self-inductance of the phase winding, _if represents the excitation current, _Lpf represents the mutual inductance of the excitation winding and the phase winding, and represents the rotor position angle. The self-inductance of the armature winding and the mutual inductance of the armature winding and the excitation winding are both related to the rotor position angle of the motor. The simplified inductance model and standard angle control conduction mode of the hybrid excitation double-pole motor are shown in Figure 3. Since the self-inductance of the armature winding is generally one order of magnitude smaller than the mutual inductance of the armature winding and the excitation winding, only the mutual inductance of the armature winding and the excitation winding is shown in Figure 3. When the hybrid excitation double-pole motor is running, a positive current is passed through the corresponding winding in the inductance rising area, and a negative current is passed through the corresponding winding in the inductance falling area, so that a continuous driving torque can be generated.

The hybrid excitation double-pole motor drive adopts PWM modulation, and a drive signal is generated once in each switching cycle. Figure 4 is a schematic diagram of the first type of commutation error. The controller enters an interrupt at time k, samples the position signal through the rotary transformer, samples the three-phase current signal through the current Hall sensor, and calculates the corresponding PWM signal based on the position signal and the current signal. Ideally, the controller should immediately output the calculated PWM signal, which acts between k and k+1. In actual situations, since the controller calculation requires a certain amount of time, it is impossible to output the calculated PWM signal immediately. It is necessary to wait until the next interrupt to output the PWM signal calculated in the previous interrupt, which acts between k+1 and k+2, resulting in a commutation error of a switching cycle.

As shown in Figure 5, it is a schematic diagram of the second type of commutation error. Ideally, the sampling frequency is infinite, and the position signal sampled by the controller is a continuous signal. The controller determines the commutation position based on the continuous position signal and the set commutation angle parameter. At this time, the first type of commutation error is not considered, and the commutation position should be consistent with the set commutation angle. However, in actual situations, the sampling frequency of the controller is limited, and the position signal sampled by the controller is a discrete signal, and the situation shown in Figure 5 may exist. The position signal sampled by the controller at time k+1 is just before the set commutation angle parameter. At this time, the controller will compare the sampled position signal with the set commutation angle parameter, so that the PWM signal without commutation is applied between k+1 and k+2 (without considering the first type of commutation error). Until k+2, the controller applies the commutation PWM signal between k+2 and k+3 based on the sampled position signal and the set commutation angle parameter. This leads to the commutation error of angle α in the figure.

FIG6 is a flow chart of a commutation error compensation algorithm for a hybrid excitation double-salient-pole motor studied in the present invention. According to the algorithm flow chart, a specific implementation scheme is described in words as follows:

Step 1: Sample the current rotor position angle θ through the rotary transformer installed coaxially with the motor to determine whether it is in the following range, where the commutation angle is θ ₁ , the sampled position signal is θ, the switching frequency is f _s , the maximum switching frequency is nf _s , the electrical frequency is _fe , and the electrical angle corresponding to one switching cycle is θ _d = _fe *360/f _s :

(1)θ∈(θ ₁ -θ _d *2/n,θ ₁ -θ _d /n)

(2)θ∈(θ ₁ -(1+1/n)*θ _d ,θ ₁ -θ _d *2/n)

Step 2: If the current rotor position is not in the above range, it means that the current rotor position is far from the set commutation point, so no operation is required at this time, and the original switching frequency _fs can be kept unchanged.

Step 2: If the current rotor position is in the interval described in (1), calculate the angle difference between the current position angle θ and the commutation angle _θ1 , and set the appropriate switching frequency so that the position of the next sampling coincides with the commutation angle to ensure that the commutation point is accurately collected. Therefore, the set switching frequency is _fe *360/( _θ1 -θ).

Step 3: If the current rotor position is in the interval described in (2), directly set the switching frequency to the maximum switching frequency _nfs . After one switching cycle, at the next sampling point, the rotor position angle can be guaranteed to be in the interval described in (1).

Step 4: Accurately sample the commutation point position, perform calculations in the digital processor, and update the PWM signal to eliminate the current commutation error.

Finally, the compensation algorithm proposed in the present invention reduces the current fluctuation amplitude, and the phase current waveform no longer produces unstable peaks, thereby greatly reducing the torque pulsation and enabling the current capacity of the switch tube to be fully utilized.

The above implementation modes are only for illustrating the technical idea of the present invention and are not used to limit the protection scope of the present invention. Any modification, equivalent substitution, improvement, etc. made on the basis of the technical scheme of the present invention in accordance with the technical idea proposed by the present invention shall be included in the protection scope of the present invention.

Claims (1)

1. The phase-change error compensation method for the mixed excitation doubly salient motor is characterized in that the mixed excitation doubly salient motor is of a stator-rotor structure with double salient poles, a rotor is free of any winding and permanent magnets, an excitation winding, an armature winding and permanent magnets are arranged on the stator, the high-speed mixed excitation doubly salient motor adopts advance angle control or three-phase nine-state control, and the advance angle control and the three-phase nine-state control perform three-phase armature current phase-change control according to a rotor position angle;

the phase-change error compensation method comprises the following specific implementation steps:

(1) When theta is E (theta) ₁ -θ _d *2/n，θ ₁ -θ _d In the process of/n), the current rotor position angle theta and the commutation point theta are calculated ₁ The distance between them is calculated according to the formula f _e *360/(θ ₁ θ), setting the time corresponding to the distance as the next switching period, thereby ensuring that the next switching period is accurately sampled to the commutation point;

(2) When theta is E (theta) ₁ -(1+1/n)*θ _d ，θ ₁ -θ _d * 2/n), the switching frequency is set to the maximum switching frequency nf _s The method is applied to the next switching period, so that the sampling point of the next switching period is ensured to be positioned in the interval in the step (1);

(3) When θ is not within the above range, the switching frequency is kept at f _s Unchanged;

the phase change angle is theta ₁ The sampled position signal is theta and the switching frequency is f _s Maximum switching frequency nf _s The minimum switching frequency is (1/n) f _s The electrical frequency is f _e One switching cycle corresponds to an electrical angle of theta _d ；

θ _d The calculation formula of (2) is theta _d ＝f _e *360/f _s Different switching frequencies f _s And electric frequency f _e The switching frequency is at the minimum frequency (1/n) f corresponding to different electrical angles _s And a maximum frequency nf _s According to the principle, when the rotor angle is close to the current commutation point, the switching frequency is changed, so that the commutation point is just positioned at the sampling point, accurate commutation is realized, and the commutation error caused by discrete sampling is eliminated.