KR20150004030A

KR20150004030A - Parameter Identification Method of Induction Motor at Standstill

Info

Publication number: KR20150004030A
Application number: KR20130076949A
Authority: KR
Inventors: 이은우
Original assignee: 엘에스산전 주식회사
Priority date: 2013-07-02
Filing date: 2013-07-02
Publication date: 2015-01-12

Abstract

The present invention provides a method for controlling an induction motor which can estimate a motor parameter in a stationary state without rotating a motor. To this end, the method for controlling an induction motor according to the present invention comprises the steps of: applying an input value of a specific pattern to a motor using an inverter and measuring an output value; calculating a system matrix using the measured information by a subspace identification method; converting the system matrix into a continuous domain matrix; converting the continuous domain matrix into an observable canonical form; and calculating an induction motor parameter from the observable canonical form.

Description

[0001] The present invention relates to a method of identifying a static induction motor,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an induction motor, and more particularly, to a method for identifying an induction motor (or parameter) of a static induction motor capable of identifying a parameter of an induction motor.

Industrial three-phase induction motors are widely used throughout the industry due to their mechanical strength. With the development of power electronics technology, vector control of induction motors is possible and high-performance torque control can be performed.

In order to perform the vector control, it is necessary to know the parameters of the induction motor accurately. These parameters can be obtained by using methods such as classical no-load rotation test and constraint test, but they require complicated test equipment. Therefore, a method of estimating a parameter using an inverter, which is an apparatus for driving an electric motor without additional test equipments, has been studied. The estimation method is divided into a rotary estimation method and a fixed estimation method according to the rotation of the motor.

The fixed estimation method has the advantage that parameters can be obtained without rotating the motor in a system in which the motor can not be rotated arbitrarily like an elevator, so that the parameters can be easily and quickly detected without mechanical fluctuation of the system.

The present invention provides a method for identification of a stationary induction device capable of estimating a motor parameter in a stationary state without rotating the motor.

According to the present invention, there is provided a method of driving a motor, comprising: applying an input value of a specific pattern to an electric motor using an inverter; Obtaining a system matrix by a subspace identification method using the measured information; Transforming the system matrix into a continuous domain matrix; Transforming the continuous domain matrix into an observable canonical form; And calculating an induction motor parameter from the observable standard form.

Also, the input value of the specific pattern of the motor includes a voltage value having various frequencies, and the output value includes a current value.

The step of applying an input voltage of a specific pattern to the motor and measuring the output current is characterized by using equations (8) and (9) in the specification.

Wherein the step of obtaining the system matrix uses Equations (10) and (11) of the specification.

The step of transforming the system matrix into a continuous domain matrix is characterized by using equation (12) of the specification.

Also, the step of converting the continuous domain matrix into an observable canonical form is characterized by using Expression 13 of the specification.

Further, the step of calculating the induction motor parameters from the observable standard form is characterized by using equation (14) in the specification text.

Since the present invention provides a method of estimating the motor parameters in a stationary state without rotating the motor, it is expected that the efficiency of the motor control can be increased. In addition, since the parameter can be obtained by using the inverter provided for the motor control without driving the motor without additional equipment, the maintenance cost of the motor can be reduced.

Fig. 1 and Fig. 2 are provided for explaining the present invention, and are equivalent circuit diagrams showing an induction motor.
3 is a graph showing a correlation between the rotor impedance of the induction motor and the frequency.
4 is a block diagram showing a block necessary for controlling an induction motor according to the present embodiment;
Fig. 5 is a flowchart showing a method of identifying an established number of induction devices according to the present embodiment; Fig.
FIG. 6 is a waveform diagram used when applying the established water identification method of the induction machine shown in FIG. 5; FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, in order to facilitate a person skilled in the art to easily carry out the technical idea of the present invention. do.

The present invention proposes a method of estimating the parameters of an induction motor using a fixed estimation method using a subspace identification algorithm. It is a program that identifies the parameters of the induction motor mounted on the industrial three-phase inverter controlling the industrial three-phase induction motor.

Fig. 1 and Fig. 2 are diagrams for explaining the present invention, which are equivalent circuit diagrams showing an induction motor, and Fig. 3 is a graph showing a correlation between the rotor impedance and the frequency of the induction motor.

In US 5,883,344, an inverse-gamma model of the induction motor shown in FIG. 1 is converted into an equivalent circuit of FIG. 2 through circuit conversion, and as shown in FIG. 3, an impedance (Rotor Impedance) A method of obtaining the parameters of an induction motor using a property that varies with frequency is proposed. This is discussed below.

The former method uses the property that the impedance of the rotor circuit changes with respect to frequency. When the frequency becomes sufficiently high, the inductance due to the inductance (L?) Becomes larger than the resistance (R2) in the rotor circuit of FIG. 1, and the current flows mostly through the resistor, which approximates the impedance (? Lx) The same effect can be obtained as the size becomes smaller. Therefore, at a sufficiently high frequency (Fhigh), the value of the imaginary part of the total impedance has a transient inductance value as shown in equation (1).

(Equation 1)

Also, by using the real part of the impedance at the same frequency, a value obtained by adding the stator resistance and the rotor resistance can be obtained.

(Equation 2)

In order to obtain the rotor time constant, the frequency (Fpeak) at which the imaginary part of the rotor impedance becomes maximum should be found as shown in equation (3). To do this, we must generate a frequency at a constant interval to obtain the response of the current and calculate the real and imaginary parts of the total impedance from the Fourier coefficients.

(Equation 3)

If this frequency is found, the rotor time constant can be obtained as shown in equation (4).

(Equation 4)

Magnetization inductance (Lφ) can be obtained by doubling the value of the imaginary part of the rotor impedance at the frequency when the rotor time constant is obtained.

(Equation 5)

The rotor resistance can be calculated using Equation 4 and Equation 5 as Equation 6

(Equation 6)

Finally, the stator resistance is calculated as in Eq. (7).

(Equation 7)

The problem with the technique of calculating the stator resistance so far is to experiment with various input frequencies in order to find the frequency at which the rotor impedance becomes maximum. In order to obtain the imaginary part of the rotor impedance value for each input frequency, And the complexity of the calculation process.

Further, in order to estimate the motor parameters, various frequencies should be used to find the frequency at which the rotor impedance becomes maximum. In order to do this, it is necessary to generate a frequency at a predetermined interval to obtain a current response and to calculate the magnitude of the current. Therefore, it may take a long time to perform the entire algorithm. The voltage synthesized by the inverter is a PWM voltage, which can not produce a clean sinusoid due to the dead time or the voltage drop of the device, which may cause a parameter error.

In order to solve this problem, the present invention provides a method for estimating a motor parameter in a stationary state without rotating the motor. A voltage signal is input while the motor is stopped, an output current is detected, and a parameter of the induction motor is estimated by a fixed estimation method through a subspace identification algorithm and a series of calculation processes. A voltage is applied to the induction motor by an input signal. By applying various frequency components, a parameter identification error can be reduced. To this end, the present invention provides a method for controlling a motor, comprising the steps of: applying an input voltage of a specific pattern to an electric motor using an inverter and measuring a current; obtaining a system matrix by a subspace identification method using measured input / output data; Transforming a matrix into a continuous domain matrix, converting the continuous domain matrix into an observable canonical form, and deriving an induction motor parameter from an observable canonical form .

4 is a block diagram showing a block necessary for controlling the induction motor according to the present embodiment.

Referring to FIG. 4, a three-phase voltage type PWM inverter 400 for controlling the induction motor 402 is provided and a control unit 401 is provided for controlling the three-phase voltage type PWM inverter 400. FIG. A current sensor 403 for measuring the current of the induction motor 402 and providing it to the control unit 401 is also provided. The three-phase voltage type PWM inverter 400 receives a phase voltage command from the controller 401 and synthesizes the voltage by the PWM method. Converts the phase current flowing in the induction motor 402 into a voltage signal, and then transmits the voltage signal to the control unit 401. The controller 401 performs the algorithm proposed in the present invention using the voltage command and the measured current value.

The present invention is characterized by providing a method of estimating parameters of an induction motor in a stationary state using subspace identification. Parameters included in the estimation include transient inductance, stator resistance, rotor time constant, and stator inductance.

5 is a flowchart showing a method of identifying the enacted water of the induction machine according to the present embodiment.

5, a memory 500 for storing a D-axis voltage command and a D-axis current, and a system matrix (501) using a subspace identification based on input / output data stored in a memory (502), transforming a system matrix represented in a discrete domain into a continuous domain (503), transforming the result into an observable canonical form (504) And calculating (505) an induction motor parameter using the result.

FIG. 6 is a waveform diagram used when applying the established water identification method of the induction machine shown in FIG. 5; FIG.

6 shows waveforms of the voltage applied to the induction motor in the stopped state and the output current, the graph 600 in FIG. 6A shows the waveform of the D-axis voltage command, and the graph in FIG. 6B (601) shows a change in frequency with time. A graph 602 in (c) of FIG. 6 is a graph showing the current flowing in the induction motor when a voltage is applied.

Next, referring to Figs. 4 to 6, a description will be given of a method of identifying the established water of the induction device according to the present embodiment.

The present invention uses the stator formulas and the rotor formulas of the induction motor as shown in Equation (8).

The dynamic equation of an induction motor is composed of a D-axis stationary equation, a Q-axis stationary equation, a D-axis rotator, and a Q-axis rotator. .

Vds is the D-axis stator voltage, ids is the D-axis stator current, idr is the D-axis rotor current, Rs is the stator resistance, Ls is the stator inductance, Lm is the magnetizing inductance, Lr is the rotor inductance and Rr is the rotor resistance , and p represents d / dt as a differential operator.

(Equation 8)

^{[Stator Equation]}

^{[Rotor Equation]}

Equation 8 can be used to derive the relationship between the stator current (ids) and the stator voltage (Vds) by erasing the rotor current (idr) in the stator equation using the rotor equation. From this equation, it can be seen that the transfer function of the stator current to the stator voltage is a secondary system having a second order denominator.

(Equation 9)

When a system that can be represented as a quadratic transfer function is transformed into an observable standard form, the coefficient of the transfer function and the coefficient of the system matrix have a constant relationship as shown in Equation 10. A system can be represented by a large number of system matrices, and once a system matrix is obtained, it can be converted into an observable standard.

(Equation 10)

To apply the D-axis stator voltage (Vds) to the motor, use the following equation to convert the DQ-axis voltage to the motor phase voltage (Vas, Vbs, Vcs)

(Equation 11)

Subsequently, step 503 of transforming the system matrix shown in Fig. 5 into a continuous domain will be described.

Since the system matrix obtained by the subspace model identification is a discrete domain matrix, it is necessary to convert it into a continuous domain. Tsamp is a period for sampling input / output data, and log is a logarithm calculation of a matrix. -1 means inverse matrix. Ad, Bd, and Cd are the values of the step (see 502 in FIG. 5) as a matrix expressed in the discrete domain, and Ac, Bc, and Cc are the results of the step (see 503 in FIG.

(Equation 12)

In step 504 of converting to the observable standard form of FIG. 5, a standard form that can be observed is calculated through a process as shown in Equation (13). tr () is a function to obtain a transpose matrix

Equation (13)

After obtaining the observable standard form, the parameters of the induction motor can be calculated using Equation 10 as shown in Equation 14.

(14)

The combined voltage is the sum of the direct current value and the alternating current value as shown in 600 of FIG. 6 so that the sign of the current flowing in each phase does not change in order to facilitate the compensation of the dead time. This can be expressed as Equation 15.

(Equation 15)

The voltage input that can be used for a static induction motor varies. A step input, a chirp signal, a sinusoidal signal, or a random signal. However, by using the simulation, it is possible to obtain induction motor parameters that are relatively noise-resistant by using a voltage obtained by adding sinusoidal signals of various frequencies to a certain offset.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, I will understand. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the scope of the appended claims, as well as the appended claims.

Claims

Applying an input value of a specific pattern to an electric motor using an inverter and measuring an output value;
Obtaining a system matrix by a subspace identification method using the measured information;
Transforming the system matrix into a continuous domain matrix;
Transforming the continuous domain matrix into an observable canonical form; And
Calculating an induction motor parameter from the observable standard form
Wherein the predetermined number of the induction devices is identified.

The method according to claim 1,
Wherein the input value of the specific pattern of the motor includes a voltage value having various frequencies, and the output value includes a current value.

3. The method of claim 2,
The step of applying an input voltage of a specific pattern to the motor and measuring the output current
The following formula (16) and (17) are used.
(Expression 16)

^{[Stator Equation]}

^{[Rotor Equation]}
(Equation 17)

(Vds is the D axis stator voltage, ids is the D axis stator current, idr is the D axis rotor current, Rs is the stator resistance, Ls is the stator inductance, Lm is the magnetization inductance, Lr is the rotor inductance, , And P denotes d / dt as a differential operator)

The method according to claim 1,
Wherein the step of obtaining the system matrix uses the following equations (18) and (19).
(Eq. 18)

(Expression 19)

(Vds is the D-axis stator voltage (Vds), Vas, Vbs and Vcs are the phase voltage of the motor)

The method according to claim 1,
The step of transforming the system matrix into a continuous domain matrix
Wherein the following formula (20) is used.
(Equation 20)

(Where Tsamp denotes a period of sampling input and output data, log denotes a logarithm calculation of a matrix, -1 denotes an inverse matrix, Ad denotes a matrix value expressed in a discrete domain, Bd and Cd denotes a matrix value expressed in a discrete domain, Value)

The method according to claim 1,
Wherein the step of transforming the continuous domain matrix into an observable canonical form uses Equation (21): " (21) "
Equation (21)

(tr () is a function for obtaining a transpose matrix)

The method according to claim 1,
Wherein the step of calculating the induction motor parameter from the observable standard form uses the following equation: < EMI ID = 22.0 >
(Equation 22)