KR20030012363A - Torque control method for direct torque control driving induction motor having largeness capacity - Google Patents

Abstract

PURPOSE: A method for controlling torque of a mass inductive motor driven by DTC(Direct Torque Control) is provided to control the torque by utilizing a switching look-up table using an intermediate vector. CONSTITUTION: A speed control portion(10) determines torque of a control target by calculating a command value of calculated torque and estimated torque of a motor. A flux control portion(20) determines flux. A torque comparator(30) compares the torque of the speed control portion(10) with a reference torque pattern. A flux comparator(40) compares a variation of flux of the flux control portion(20) with a reference flux change pattern. A switching control portion(50) is formed with a switching lookup table(52) and a voltage vector selection portion(51) in order to output a gating signal of a 3-level inverter(120). A calculation portion(70) detects a voltage of an inverter. A current sensor portion(80) detects current from the 3-level inverter(120). An adaptive observer(90) estimates flux of a stator, flux of a rotator, a speed of the rotator, and a resistor of the stator. A torque estimation portion(100) estimates torque from the flux of the stator and the flux of the rotator. A flux estimation portion(110) estimates intensity of flux.

Description

Torque control method for direct torque control driving induction motor having largeness capacity}

The present invention relates to a torque control method for improving performance during low speed operation of a large-capacity induction motor driven by DTC (DTC).

Induction motors by direct torque control (DTC) are widely used in industrial applications.

Such a DTC method has advantages such as fast torque response, simple controller structure, robustness due to motor parameter variation, and can be implemented without attaching a speed sensor to an induction motor.

Therefore, DTC algorithm is spreading from small capacity to large motor drive.

Although many control algorithms have been studied and published about the direct torque control method of an induction motor driven by a two-level inverter, few studies have been conducted on the three-level system.

This is especially true for high-voltage, high-capacity, three-level inverters where switching frequencies are limited to less than 1 kHz due to cooling problems in switching devices in three-level high-capacity systems.

Accordingly, the present invention intends to propose a torque control method for a low speed operation region of an induction motor driven by a three-level inverter, and provides a simple hysteresis control method for a three-level inverter, and stator flux drop in the low speed operation region. I would like to propose a new way to solve the problem.

Here, in order to improve the performance of DTC, an adaptive omnidirectional observer that estimates stator flux, rotor speed, and stator resistance in a wide speed range is applied.

In the present invention, in the torque control of a DTC-driven induction motor, the torque can be controlled by using a new switchinglook-up table using an intermediate vector of voltage vectors during low speed operation control. This is to solve the stator flux drop in the operating area.

1 shows a typical spatial voltage vector of a three level inverter output voltage.

2 shows a three level inverter torque gradient pattern.

3 is a view illustrating a switching method during low speed operation in a torque control method of a large capacity induction motor driven by DTC of the present invention.

Figure 4 is a block diagram of a torque control device to which the torque control method of a large-capacity induction motor driven by the present invention DTC (DTC) is applied.

The DTC principle and the torque control method of the three-level inverter system will be described first.

Figure 1 shows a typical spatial voltage vector of a three level inverter output voltage.

here, The subscripts z, h, i, and f denote zero vector, half vector, intermediate vector, and full vector.

Three-level inverters have 27 switching voltage vectors to choose from, and more complex inverter switching voltage vectors are selected than two-level inverters.

Neglecting the voltage drop of the stator resistance from the induction motor equation, the relationship between the inverter output voltage vector and the stator flux change can be expressed as Equation 1 below.

here, : Inverter output voltage vector, t _sp : Sampling period.

Equation 1 shows that the supplied stator voltage vector generates a stator flux change.

The magnitude of the stator flux change is proportional to the product of the supply voltage vector and the sampling period. The vector direction of the stator flux change remains the same as the direction of the selected voltage vector.

The stator and rotor flux can be expressed as Equation 2 below.

here,

The electromagnetic torque can be expressed as Equation 3 below.

It can be seen from Equation 3 that the torque is determined by the size of the stator flux and the phase angle of the stator flux with respect to the rotor flux.

Here, assuming that the stator flux vector is in the k-th sector, the selection of each stator voltage vector can be represented as in FIG.

or The choice of may increase the phase angle between the stator flux and the rotor flux.

As a result, the developed torque or It can be increased by the application of.

1, the stator flux is Is increased by the choice of It can be seen that it is reduced by the selection of.

If the half voltage vector is selected, the torque gradient can be reduced.

2 shows a pattern of three-level inverter torque slope.

The double torque hysteresis band method shown in FIG. 2 is applied to a three-level inverter, and when the torque falls to the negative upper hysteresis band, an appropriate large voltage vector is selected to increase the torque.

If the controlled torque approaches the positive lower hysteresis band, the full voltage vector is changed to a half voltage vector.

At this time, if the torque comes in contact with the positive upper hysteresis band, a zero voltage vector is applied to reduce the torque value.

The same applies to this rule for reverse driving, and the following switching lookup table is shown in Table 1 below.

Assuming that the stator flux (lambda_s) is located in the k-th sector, a possible voltage vector for increasing torque is , , , And to be.

Among these voltage vectors, one voltage vector is selected in consideration of the double torque hysteresis band and the stator flux condition.

However, in order to reduce torque, voltage vector Only is selected.

As described with reference to Fig. 1, there is no effective voltage vector that can reliably increase the stator flux at the two sector boundaries.

Thus, as the rotating stator flux vector is moved from one sector to another, another problem that degrades performance in low speed operation is the selection of the zero voltage vector.

In the low speed region, the zero voltage is not easily obtained due to the low stator frequency value of the resulant phase angle reduction, so that the torque reduction control cannot be effectively ensured.

In the present invention, as described above, to solve the problem of the DTC algorithm caused by the stator flux drop and the application of the zero-voltage vector during low-speed operation,

The demagnetization effect is mainly caused by the nonlinearity of the induction motor. This problem occurs immediately after the stator flux vector is changed from one sector to another.

To solve this problem, a reverse voltage vector is applied to reduce the torque instead of the zero voltage vector. However, this method can achieve a quick decrease in torque in the low speed range but increase the switching frequency and torque ripple.

Although the non-magnetization problem can be solved by the reference axis and switching sector rotation in the same way as in the two-level inverter, the phenomenon caused by the zero voltage vector application cannot be easily solved.

Therefore, to solve this problem, a new switching lookup table in the low speed operation range must be applied.

The intermediate voltage vector in the three-level inverter solves the non-magnetization problem without the reference axis and switching sector rotation applied in the two-level inverter.

The present invention is to control the three-level inverter of the induction motor by selecting the switching voltage vector according to the position of the stator magnetic flux and the state of the torque,

Detects the amount of magnetic flux reduction and compares the detected amount of magnetic flux reduction with a predetermined reference level to determine whether it is a low speed operation area, and if it is determined as a low speed operation area, selects an intermediate vector voltage according to the torque state to switch gating. And outputs a signal.

As shown in FIG. 3, the switching sector is divided into 12 sectors to accurately select the stator voltage vector.

Here, the k-th sector is divided into a lower subsector and an upper subsector, and each has a width of 30 °.

Stator voltage vector determined according to Table 1 in the lower subsectors Increases the stator flux.

At this time, Medium voltage vector instead By choosing, we can solve the non-magnetization problem.

In order to obtain the magnetization effect effectively at the two sector boundaries, we use a modified lookup table as shown in Table 2, especially in the low speed region.

Here, the switching from the basic switching lookup table as shown in Table 1 to the modified lookup table can be performed by detecting the amount of decrease in the stator flux magnitude.

If the stator flux size is below the minimum fixed level, the modified lookup table of Table 2 can be applied instead of Table 1 of the basic lookup table.

At this time, if the stator flux moves in the upper subsector, an appropriate voltage vector is determined according to the switching lookup table of Table 2.

As described above, the selection of the voltage vector in Table 2 is basically the same as in Table 1 except that the reverse voltage vector is selected instead of the zero voltage vector for torque control.

4 is a block diagram showing the configuration of the torque control device to which the above-described switching look-up table is applied.

The configuration is as follows.

Speed setpoint received ) And the rotor speed of the motor driven by the estimated inverter 120 ( Torque command value and calculates the torque command value and the estimated torque ( A speed control unit 10 for calculating the torque of the control target by calculating

Flux setpoint ) And the estimated magnetic flux size of the motor ( Magnetic flux control unit 20 for calculating the magnetic flux by calculating the error of the),

A torque comparison unit 30 for comparing the torque output from the speed control unit 10 to the reference torque pattern,

A magnetic flux comparator 40 for comparing the change in magnetic flux output from the magnetic flux control unit 20 with a reference magnetic flux change pattern,

Stator flux size ( Of the electric motor input from the torque lookup section 30 and the magnetic flux comparison section 40 and the switching lookup table 51 for the selectable voltage vectors, applying an intermediate voltage vector in the low speed operation region. Output torque and magnetic flux and detected stator flux of the motor ), From the switching lookup table 52, a switching controller 50 configured to include a voltage vector selecting unit 52 for selecting _sk ) and outputting a gating signal of the inverter 120;

An inverter voltage calculator 70 for detecting inverter voltages Vds and Vqs,

A current sensor unit 80 for detecting currents Ids and Iqs from the inverter 120;

The stator magnetic flux (S) is calculated using the calculated inverter voltages Vds and Vqs and the measured currents Ids and Iqs. ), Rotor flux ( ), Rotor speed ( ), Stator resistance ( Adaptive adaptive observer 90 for estimating

Stator flux detected through adaptive full-dimensional observation unit 90 ( ) And rotor flux ( Torque from Torque estimating unit 100 for estimating) and

Observed stator flux from adaptive full-dimensional observer 90 ), The magnitude of the magnetic flux ( It is configured to include a magnetic flux size estimator 110 for estimating ().

And, the speed control unit 10 is a speed command value ( And estimated rotor speed from the adaptive full-dimensional observation section 90 Torque command value and torque determined from the PI (12) controller, PI (Proportional Integral) controller (12) which determines the torque command value from the results of the first operator (11) Torque estimated from estimator 100 It is configured to include a second operator 13 for calculating the error with).

The flux control unit 20 is a flux command value ( ) And the magnitude of the magnetic flux estimated by the magnetic flux size estimation unit 110 ( It is configured to include a third operator 21 for calculating an error with the ().

Unexplained reference numeral 91 denotes a flux observer and 92 an adaptive scheme.

The apparatus of the present invention as described above is the stator flux magnitude of the switching voltage of the inverter 120 for driving the induction motor ( ), A lookup table is applied to the intermediate vector voltage in the low speed region, and the non-magnetization phenomenon in the low speed region is solved.

The operation of the apparatus of the present invention as described above is as follows.

The voltage calculator 70 calculates inverter voltages Vds and Vqs from the gating signal input to the inverter 120.

In addition, the current sensor unit 80 measures the current supplied from the inverter 120 to the induction motor serving as a load and outputs the inverter currents Ids and Iqs to the adaptive full-dimensional observation unit 90.

The adaptive all-dimensional observer 90 uses the calculated inverter voltages Vds and Vqs and the measured inverter currents Ids and Iqs to generate the stator magnetic flux. ), Rotor flux ( ), Rotor speed ( ), Stator resistance ( ) Is estimated.

In the torque estimator 100, the stator magnetic flux estimated from the adaptive full-dimensional observation unit 90 ( ), Rotor flux ( ) To the torque ( ) Is estimated.

The estimated torque ( ) Is input to the speed control unit 10, and the rotor speed of the current electric motor (estimated from the adaptive full-dimensional observation unit 90) ) Is also input to the speed control unit 10.

Here, the magnetic flux magnitude estimator 110 estimates the stator flux (estimated from the adaptive full-dimensional observation unit 90). Of magnetic flux from ) Is estimated and output to the flux control unit 20.

In the first operator 11 of the speed controller 10, the speed command value ( ) And the rotor speed received from the adaptive full-dimensional observation unit 90 ) And calculates an error with the PI controller 12, and outputs it to the PI controller 12, which determines and outputs a torque command value for driving the inverter 120 therefrom.

The torque output value thus output is the torque estimated from the torque estimator 100 through the second operator 13 ( ) Error is obtained.

In the magnetic flux control unit 20, the magnetic flux command value (the third operator 21) ) And the magnitude of the magnetic flux of the motor estimated by the magnetic flux magnitude estimation unit 100 ( The magnetic flux is determined by calculating the error with.

The magnetic flux determined as described above is input to the magnetic flux comparing unit 40, and the magnetic flux comparing unit 40 compares the reference magnetic flux pattern and outputs the result to the voltage vector selecting unit 50.

In addition, the final torque command value obtained through the second operator 13 of the speed controller 10 is compared with the reference torque pattern through the torque comparator 30, and the result is compared with the voltage vector selector ( 51).

The voltage vector selector 50 is a stator flux obtained from the adaptive full-dimensional observer 110 ( In consideration of the voltage vector from the torque comparator 30 and the flux comparator 40 _sk )

Stator flux ) And the voltage vector (from the switching lookup table 52 as shown in Table 2) _sk ) to output the gating signal of the inverter 120.

Here, the voltage vector selection unit 50 is a stator magnetic flux ( When the position of) is the lower sub-sector, the voltage vector ( _sk ) is the intermediate vector depending on the state of torque or To be.

In the present invention as described above, in the direct torque control of an induction motor driven by a high-capacity three-level inverter, by applying the intermediate vector in the low speed operation region, the non-magnetization problem can be solved without the reference axis and switching sector rotation applied in the two-level inverter. have.

Claims (1)

In controlling the three-level inverter of the induction motor by selecting the switching voltage vector according to the position of the stator flux and the state of the torque, Detects the amount of magnetic flux reduction and compares the detected amount of magnetic flux reduction with a predetermined reference level to determine whether it is a low speed operation area, and if it is determined as a low speed operation area, selects an intermediate vector voltage according to the torque state to switch gating. Torque control method of a large-capacity induction motor driven by DTC (DC) characterized in that to output a signal.