Initial positioning control method for switched reluctance motor without position sensor

文档序号:1204083 发布日期:2020-09-01 浏览:22次 中文

阅读说明:本技术 一种开关磁阻电机无位置传感器初始定位控制方法 (Initial positioning control method for switched reluctance motor without position sensor ) 是由 陈昊 尹正凯 桑永豪 渠苏东 周大林 张珂 巩士磊 阎明 张战 袁利 李祥阳 于 2020-06-04 设计创作,主要内容包括:本发明公开了一种开关磁阻电机无位置传感器初始定位控制方法,属于开关磁阻电机控制的技术领域。本发明对开关磁阻电机的三相绕组同时注入高频脉冲,基于此方法可获得整个周期的脉冲电流信息,根据脉冲电流和绕组电感之间的关系,利用脉冲电流对电感实施平均分区,根据各分区脉冲电流的大小确定初始起动相,该方法适用于静止或具有一定初速度的开关磁阻电机的初始位置估计。该方法无需复杂的控制策略,算法简便,对位置的估计可靠且通用性较强。(The invention discloses a switched reluctance motor position sensorless initial positioning control method, and belongs to the technical field of switched reluctance motor control. The invention injects high-frequency pulses into the three-phase winding of the switched reluctance motor simultaneously, can obtain the pulse current information of the whole period based on the method, implements average partition on the inductor by using the pulse current according to the relation between the pulse current and the winding inductance, and determines the initial starting phase according to the size of the pulse current of each partition. The method does not need a complex control strategy, has simple and convenient algorithm, and has reliable estimation on the position and stronger universality.)

1. A switch reluctance motor initial positioning control method without a position sensor is characterized by comprising the following steps:

A. selecting proper pulse injection excitation time and frequency;

B. high-frequency pulses are injected into three-phase windings of the switched reluctance motor at the same time;

C. carrying out average partition on the inductor according to the pulse current of the pulse voltage response;

D. and determining the interval of the initial position of the rotor according to the peak value of each subarea pulse current.

2. The initial positioning control method of the switch reluctance motor without the position sensor according to claim 1, wherein the step a selects the proper pulse injection excitation time and frequency, the proper pulse injection excitation time can avoid the motor from generating reverse rotation, and the proper pulse injection frequency can avoid the adverse effect of skin effect on the equivalent resistance of the winding and the interference of larger negative torque, thereby ensuring the normal operation of the system.

3. The method as claimed in claim 2, wherein the step B injects high frequency pulses to the three phase windings of the switched reluctance motor simultaneously, so that the windings of the motor can respond to the pulse current.

4. The method as claimed in claim 3, wherein the step C is performed according to a function relationship between inductance and pulse current

Figure FDA0002523828320000011

5. The method as claimed in claim 4, wherein the step D is performed by comparing the magnitudes of the pulse currents in response to the respective zones to determine the initial position of the rotor.

Technical Field

The invention discloses a switched reluctance motor position sensorless initial positioning control method, and belongs to the technical field of switched reluctance motor control.

Background

The structure of the switched reluctance motor is fundamentally different from that of the traditional alternating current motor and direct current motor, and a rotor formed by laminating silicon steel sheets is not provided with a winding and a permanent magnet made of rare earth materials; the stator is wound with simple and concentrated coils, and the whole mechanical structure is relatively simple. The switched reluctance motor has a large starting torque and a high torque/current ratio, and is therefore often used in frequent starting. However, the position sensor is easily damaged in a severe environment, so that the motor cannot normally operate; meanwhile, the cost and the complexity of the system can be increased due to the position sensor, so that the reliability of the system is reduced, the popularization and the application of the switched reluctance motor are limited, and the position sensor-free control is adopted to help get rid of the dilemma.

In recent years, many studies have been made by domestic and foreign scholars in the field of sensorless control of switched reluctance motors, wherein estimation of the initial position of the motor is a precondition for realizing sensorless control of the switched reluctance motor, especially in the occasions such as electric vehicles, military equipment control and the like requiring the motor to run irreversibly, and the academic world has proposed many estimation techniques for the initial position of the switched reluctance motor, but most of the research methods only analyze the initial position of the motor in a static state in detail, and have less research on position estimation methods in an inertial state, wherein the research methods mainly include a pulse injection method, a non-conducting phase detection method, a detection method based on intelligent control, an additional element detection method and other position estimation algorithms.

The position sensorless control techniques proposed above all have their respective applicability and limitations, and these methods have their respective advantages and disadvantages. The bootstrap circuit method is one of additional element detection methods, and is characterized in that a capacitor in the bootstrap circuit is charged, then diagnostic pulse current is injected into each phase winding, and the interval where the rotor is located is judged according to the relation between the peak value of each phase diagnostic current and the position of the rotor. Although the method does not need an accurate system model and only needs enough training data to fit to obtain the relation of the rotor position-current-flux linkage to complete position estimation, the method needs a large amount of time to train data and has complex algorithm.

Disclosure of Invention

The invention aims to provide the initial positioning control method of the switched reluctance motor without the position sensor, which has good universality and strong transportability, can estimate the section of the rotor without predicting the electromagnetic characteristic data and an accurate mathematical model of the switched reluctance motor, and solves the technical problems of complicated calculation caused by the dependence on motor body parameters in the traditional initial positioning methods of the switched reluctance motor without the position sensor, such as a non-conducting phase detection method, an additional element detection method, intelligent control and the like.

In order to achieve the technical purpose, the invention adopts the following technical scheme to realize:

a switch reluctance motor initial positioning control method without a position sensor comprises the following steps:

A. selecting proper pulse injection excitation time and frequency;

B. high-frequency pulses are injected into three-phase windings of the switched reluctance motor at the same time;

C. carrying out average partition on the inductor according to the pulse current of the pulse voltage response;

D. and determining the interval of the initial position of the rotor according to the peak value of each subarea pulse current.

As a further optimization scheme of the initial positioning control method of the switched reluctance motor without the position sensor, the step A is that proper pulse injection excitation time and frequency are selected, the proper pulse injection excitation time can prevent the motor from rotating reversely, and the proper pulse injection frequency can prevent adverse effects of skin effect on winding equivalent resistance and interference of large negative torque, so that normal operation of the system is guaranteed.

Further, according to the initial positioning control method of the switched reluctance motor without the position sensor, high-frequency pulses are injected into the three-phase winding of the switched reluctance motor at the same time in the step B, so that the winding of the motor can respond to pulse current.

Furthermore, the initial positioning control method of the switched reluctance motor without the position sensor comprises the step C of obtaining a functional relation between the inductance and the pulse current

Figure BDA0002523828330000021

And carrying out average partition on the inductor according to the responding pulse current, wherein L (theta) is phase inductance, U is the amplitude of the detection voltage pulse, Δ t is the width of the detection voltage pulse, and Δ i is the increment of the current of the phase winding to be detected in the time of pulse injection Δ t.

Still further, according to the initial positioning control method of the switched reluctance motor without the position sensor, in the step D, the position interval can be determined by comparing the magnitude of the pulse current according to the response of each partition.

Based on the switched reluctance motor, during initial positioning, proper pulse excitation time and frequency are selected for each phase winding in the switched reluctance motor structure and injected into the three-phase winding, and the initial positioning of the switched reluctance motor can be realized according to the magnitude of the peak value of the responding pulse current.

Has the advantages that:

by adopting the technical scheme, the invention has the following technical effects: the initial position of the switched reluctance motor can be estimated without a complex flux linkage model, and the initial position can be positioned by combining a pulse injection method and an inductance partition method.

Drawings

FIG. 1 is an inductance section schematic;

FIG. 2 is a schematic diagram of a pulse injection method;

FIG. 3 is a graph of inductance curve versus pulse current;

FIG. 4 is a flow chart of initial positioning;

fig. 5 is a waveform diagram of three-phase pulse current and its envelope in the inertial state.

Detailed Description

An embodiment of the invention is further described below with reference to the accompanying drawings:

the method estimates the position information of the rotor by detecting the pulse current peak value in each period, and has strong universality and portability.

Fig. 1 is an inductance section schematic. When the switched reluctance motor is static or has a certain initial speed, high-frequency pulses are injected into the three-phase winding simultaneously, and the initial position of the rotor is estimated by comparing the magnitude of the pulse current of each phase to implement average partition on the inductor. Fig. 1 shows a three-phase switched reluctance motor inductance section diagram and an impulse response current peak envelope section diagram, respectively, where the three-phase current peak envelope in fig. 1 may be obtained by a numerical calculation method.

The excitation time and frequency of pulse injection are determined on the premise that the experiment is normally carried out, and when the motor is in a static state, if the pulse injection excitation time is too long, the motor can rotate, so that the initial position can be changed, and the motor can also rotate reversely; if the injected pulse excitation time is too short, the pulse current amplitude can be reduced, so that the problems of precision reduction, interference resistance reduction and the like are caused; the frequency of pulse injection also influences the measurement precision, and the overlarge pulse frequency not only increases the loss of a power electronic device, but also increases the equivalent resistance of a winding by the generated skin effect under high frequency to influence the estimation of the position; too low a pulse frequency may make the motor self-starting even if the pulse current is not easily detectable. The maximum and minimum excitation time of pulse injection is shown in formulas (1) and (2):

Figure BDA0002523828330000031

in the formula: l ismaxRepresents the maximum inductance, LminMinimum inductance is indicated, β s stator pole arc, iminIs the minimum current that can be measured by the current sensor.

The pulse frequency should satisfy the following formula (3):

meanwhile, the range of the switching frequency of the power electronic device is also considered, so that the maximum frequency of the injection pulse satisfies the formula (4):

fmax≤fMOSFET max(4)

in summary, for different types of switched reluctance motors, the bus voltage, the maximum switching frequency of the power electronic device, the measurement accuracy of the current sensor, the starting torque, the minimum/maximum inductance of the motor, and the a/D conversion time of the controller are obtained, and the intervals of the excitation time and frequency of the injection pulse can be calculated according to the equations (1), (2), (3), and (4)

Fig. 2 is a schematic diagram of a pulse injection method, and a circuit equation of a certain phase winding m of the switched reluctance motor is shown in formula (5):

Um=imRm+dψm/dt (5)

flux linkage psi of m-phase winding of switched reluctance motormAnd current imThe specific expression is as shown in formula (6) in relation to the rotor position theta) Shown in the figure:

ψm(i,θ)=Lm(i,θ)im(6)

in the formula: u shapemIs m-phase terminal voltage, RmIs m-phase resistance, imFor m phase current, psimIs an m-phase magnetic chain.

Combining the equations (5) and (6), a circuit equation shown in equation (7) can be obtained:

the formula (7) is composed of three parts of resistance voltage drop, transformer electromotive force and motion electromotive force. As shown in fig. 2, during a short Δ t time, a high frequency low voltage pulse is injected into the phase winding under test, and a response current will appear in the phase winding. The voltage pulse with short time and low amplitude makes the pulse current of the detected phase response small, and the resistance value of the motor winding is small, so the resistance voltage drop can be ignored. In summary, the expression of any phase inductance can be obtained after the simplification process, as shown in formula (8):

in the formula: l (theta) is phase inductance, U is amplitude of the detection voltage pulse, delta t is width of the detection voltage pulse, and delta i is increment of current of the phase winding to be detected in the time of injecting the pulse into delta t.

Fig. 3 is a graph showing inductance curve-pulse current relationship, where the initial position determination is the basis for implementing the sensorless operation of the switched reluctance motor, and in order to make the motor rotate normally, a phase capable of generating positive torque must be determined, and this phase is taken as an initial starting phase, that is, the starting phase must be in an inductance rising region of the phase.

Fig. 4 is a flow chart of initial positioning, and position estimation with a stationary motor or a certain initial speed can be achieved through a certain calculation procedure. Under the static condition, the position of the rotor can be estimated by only comparing the magnitude of the pulse current in each partition, and then the initial starting phase is determined; under the condition of certain initial speed, in order to judge the partition where the initial starting phase is located more accurately, the pulse current is subjected to envelope extraction by means of a numerical calculation method based on Hilbert transform to obtain position information, and therefore the initial starting phase is determined. Setting a real-valued function x (t), Hilbert transforming it to obtain a new function denoted as x ^ (t) (or denoted as H [ x (t) ]), then

Figure BDA0002523828330000052

Inverse transformation to

Comparing the convolution concept, it can be found that the expression of hilbert transform in the above formula is actually the result of convolution operation between the original signal and another signal (assumed as h (t)). I.e. a new signal obtained after a hilbert transformIs the result of the operation of the original signal x (t) through a filter (or a system) h (t), the impulse response of which is h (t) ═ 1/π t. The Fourier transform operation is carried out on the data to obtain the following data:

H(jω)=-jsgn(ω) (11)

it can also be written as:

where sgn () is a sign function. Such a conclusion can be reached from equation (12): all positive frequency components are hilbert transformed with a back-shift term of-90 deg., and all negative frequency components are hilbert transformed with a back-shift term of +90 deg., which is equivalent to a quadrature filter, but the magnitude is not changed accordingly.

In the case of only real signals, if they are to be converted into complex signals, analytical signals need to be constructed. Taking original signal x (t) as real part of complex signal, and subjecting the signal after Hilbert transformAs an imaginary part of the complex signal, the following complex signal is then obtained:

a signal has both amplitude and phase information, so that:

x(t)=A(t)cos(ω0t+θ(t)) (14)

when this is substituted into the formula (13):

according to the formula, the compound has the advantages of,in order to be able to obtain phase information,

Figure BDA0002523828330000066

is amplitude information. The desired envelope information can be obtained by taking the absolute value of the above formula.

Fig. 5 is a waveform diagram of three-phase pulse current and its envelope in an inertial state, when the motor has a certain initial speed, a high-frequency pulse is applied to the three-phase winding to respond to the pulse current, and the envelope can be obtained by the hilbert transform, so that the partition where the rotor is located can be determined more intuitively, and thus the initial starting phase can be determined.

10页详细技术资料下载
上一篇:一种医用注射器针头装配设备
下一篇:故障确定设备、电动机驱动系统和故障确定方法

网友询问留言

已有0条留言

还没有人留言评论。精彩留言会获得点赞!

精彩留言,会给你点赞!