JPS63186596A - Controlling method for induction motor by pwm inverter - Google Patents

Abstract

PURPOSE:To know a correct magnetic flux component and a torque component by increasing a discriminating divisions equally divided from the circumference of magnetic flux vector direction discriminating means by a microprocessor as large as possible to calculate. CONSTITUTION:Magnetic vector direction discriminating means 8 outputs any of 12 types as magnetic vector position signal phii according to the clockwise rotating angle from the d-axis of a magnetic flux vector phi. Control signal generating means 9 selects the optimum voltage vector according to a magnetic flux increase/decrease signal phix from magnetic flux comparing means 6 aud a torque increase/decrease signal taux from torque comparing means 7, and turns ON, OFF switching elements so that a 3-phase PWM inverter 2 generates it.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a magnetic flux and torque control system for a three-phase induction motor using a high-speed digitally controlled PWM inverter.

[Conventional technology]

In recent years, inverter control has been common in controlling induction motors, but instantaneous space vector control, which will be described next, has recently been in the spotlight due to problems such as noise and transient response.

This instantaneous space vector control 1 problem was introduced in 19861.
It is also explained in the paper entitled "New high-speed torque control method for induction motors based on instantaneous slip frequency control" published in Volume 106, Issue 1, Page 9 and below of IEEJ Transactions B published in April, but the instantaneous space vector A method for controlling an induction motor will be explained using diagrams.

FIG. 1 is a block diagram showing a method of controlling an induction motor using instantaneous space vector control, and FIG. 2 is a main circuit connection diagram of the three-phase PWM inverter section in FIG.

The three-phase PWM inverter 2 is a large-capacity transistor, GT
Switching elements Q1 to Q6 that can be turned on and off, such as O thyristors or SI thyristors, and diodes D, -
Each of the switching elements Q1 to Q6 is connected in parallel to a diode h, respectively, and is also connected as a three-phase bridge circuit, and the switching elements Q1 to Q are connected by a switching signal sent from the control signal generation means 9.
6 turns on and off. The three-phase PWM inverter 2 configured in this way receives power from the DC voltage source 1 and operates the three-phase induction motor 3.

The voltage of each phase of the three-phase induction motor 3 is Vυ, My, v
When w and the currents of each phase are Lg, iy, and fw, to decompose these into two components of the d-axis and q-axis on the orthogonal coordinate axes, vd+V, and td, l, use the following equation. According to Oq component i (1 + Iq, voltage vector V and current vector i are v=vd+jvq...mark [old and old......
■t=id+jsq・・・・・・
・・・・・・・・・・・・・・・・・・■

Here, j indicates a vector product.

Voltage vector calculation means 4a, current vector calculation means 4
b and the magnetic flux calculation means 4 constitute means for calculating the instantaneous value of the magnetic flux vector. The voltage vector calculation means 4a uses the detected voltage value V of the DC voltage source 1 and the six switching elements Q1 to Q6 that constitute the three-phase PWM inverter 2.
Obtaining information on which one of them is on, the voltage of each phase of the three-phase induction motor 3 is determined.

Knowing My and VW, calculate the voltage vector V using equation (2).

In addition, the U-phase positive and negative switching elements Ql and Q4 of the three-phase PWM inverter 2, and the ■-phase positive and negative switching elements Q3.
Q6, W-phase positive/negative switching elements Qs, Q2,
It is configured to turn on one of the switching elements of at least two phases, which are different in positive and negative, or to turn on the switching element of one of the three phases, which are equal in positive and negative, and to turn off all the others.

When the six switching elements Ql-Qa constituting the three-phase PWM inverter 2 are turned on and off under the above conditions,
When the voltage vector v (k) depending on the on/off state is calculated using the equation (2), the following voltage vector table is obtained. In the table, 1 written in each switching element column indicates on, O indicates off, and the actual values of d component Vd and q-axis component v9 are the values in the table, L weight and DC voltage. It is multiplied by the detected voltage value V of the source.

Voltage vector table Here, the voltage vectors v (0) and v (13) whose magnitude is O are called zero voltage vectors, and when outputting this zero voltage vector, the switching elements constituting the three-phase PWM inverter 2 Among them, all of the positive side switching elements QhQs +Qs are turned on, or the negative side switching elements Q2. This is the case when all Q4-Qs are turned on, and corresponds to the case where the terminals of each phase of the three-phase induction motor 3 are short-circuited.

FIG. 3 is a voltage vector diagram, which shows a voltage vector table, with ma (1) in the same direction as the d axis,
Then proceed 306 times□ clockwise from v(2) to v(1
2) and two zero voltage vectors v(13), V
(0) is output. However, except for the zero voltage vector, each voltage vector v (k) has the same absolute value when the value of k is an odd number, but when it is an even number, the absolute value is v/T/2 is the absolute value of

The current vector calculation means 4b calculates each phase current detection value lu, iv of the three-phase induction motor 3 detected by the current transformer OT.

From iw, a current vector 5 is calculated using equation 0.

The magnetic flux calculation means 4 calculates the magnetic flux vector J from the voltage vector V, the current vector i, and the primary resistance value R of the three-phase induction motor 3 using the following equation.

It can be thought of as extending in the direction of . magnetic flux vector φ
Of course, the d and q two-axis components φd and φq represent O. The torque calculation means 5 that calculates the instantaneous torque value calculates the instantaneous torque value f from the magnetic flux vector φ and the current vector l using the following equation. .

De=φ×i−φdX iq−φq X td...
・・・・・・・・・・・・・・・■The magnetic flux comparison means 6 calculates from the magnetic flux vector φ1φ1=sword■ ・・・・・・・・・・・・・・・
・・・・・・・・・・Absolute value of magnetic flux calculated by ■ 1φ1
and a magnetic flux command value φ given from another source, and generate a magnetic flux increase/decrease signal φ8 indicating whether the absolute value 1φ1 of the magnetic flux is too small, approximately appropriate, or too large.

The torque comparison means 7 compares the instantaneous torque value t with a torque command value f given from another source, and generates a torque increase/decrease signal f indicating whether the instantaneous torque value T is too small, almost appropriate, or too large. Figure 4 is a magnetic flux vector diagram, and when the three-phase induction motor 3 is operating stably,
The magnetic flux vector φ maintains a constant size and rotates clockwise at a constant speed as shown by the arrow, and the trajectory of the tip of the magnetic flux vector φ becomes a circle.

As explained in Equation 0, applying a voltage component in the same direction as the current magnetic flux vector φ will result in magnetization, and applying a voltage component in the opposite direction will result in demagnetization. That is, if a voltage vector v (k) having a magnetic flux component Φ is selected, the magnetization is increased.

On the other hand, 90 degrees clockwise from the direction of the current magnetic flux vector φ
The voltage component advanced by ° is the torque component element, and if a voltage vector V(k) having a voltage component in this direction is selected, a positive torque is generated and the voltage vector φ rotates clockwise.

In the case of the magnetic flux vector φ shown in FIG.
If it is reported that the absolute value 1φ1 of the magnetic flux is too small by the magnetic flux increase/decrease signal φ8 from the magnetic flux increase/decrease signal φ8, the switching element is turned on and off so as to apply the voltage vector v(2) with the largest magnetic flux component φ.

Torque by the torque increase/decrease signal from the torque comparison means 7! If it is reported that is too small, the switching element is turned on and off so as to apply the largest voltage vector v(5) of the torque component. If it is reported that both are too small, the switching element is turned on to apply voltage vector v(3) or v(4) having both magnetic flux component Φ and torque component.

Turn off.

In this way, the switching elements to be turned on and off according to requests from the magnetic flux comparison means 6 and the torque comparison means 7 are as follows:
Since it depends on the direction of the magnetic flux vector φ at that time, the direction of the magnetic flux vector must be calculated every calculation cycle.

The calculation of the clockwise angle θ from the d-axis of the magnetic flux vector φ is θ=tan(φq/φd) ・・・・・・・・・・・・
. . . . . . . . ■ However, directly calculating this using a microprocessor requires many calculation steps and takes too much calculation time.

Therefore, by simply dividing the circumference 3600 into equal integer parts and determining which part of the boundary line the magnetic flux vector φ belongs to, the calculation time can be greatly reduced.

As an example, if the circumference is divided into 6 equal parts and the boundary line is taken every 60° starting from 30°, the magnetic flux vector φ
By simply calculating the sign and absolute value of the ratio φq/φd of both axis components φd and φq and comparing it with the value on the boundary line, it is possible to know the current location division of the magnetic flux vector φ.

Therefore, as shown in FIG. 4, the magnetic flux vector direction determining means 8 determines the clockwise rotation angle θ from the d-axis of the magnetic flux vector φ.
Accordingly, the magnetic flux vector position signal φi is I, It
, II [, IV, V, VI NoifRekawo, it is assumed that the output is as follows.

- 30°≦0<30°; I 30≦θ<90°; ■ 90°≦θ<1500:11 1506≦8<210°; ■ 210'≦θ<270°; V 270'≦8 <330' : VI The magnetic flux vector direction determining means 8 stores the value of tan30° of 0.5774, and calculates the d-axis components φd and q of the magnetic flux vector φ.
Using the axis component φ, the magnetic flux vector position signal φl is determined from the sign of φd and the sign and absolute value of φ, /φd according to the following table.
Output.

The control signal generating means 9 generates each magnetic flux vector position signal φi
The center position of the angular range, i.e., &+:tO'
, I(7) ト＊Haro 0' , m(D ト ek は 120
', ■ (7) Tokiha180', V (7) To! Ha24
0', 11 (D to ljk c300'0:) direction is determined to be the direction of the magnetic flux vector φ, and the magnetic flux increase/decrease signal φ8
According to the torque increase/decrease signal and x, a switching signal is sent to the 3-phase PWM inverter 2, and the switching element Q
Turn on and off l to Q.

As an example of the magnetic flux increase/decrease signal φ8 which is the output of the magnetic flux comparison means 6, if the error Δφ is set and the absolute value of the magnetic flux *1φ1 is within the range of ±lφ/2 with respect to the magnetic flux command value φ, it is considered to be approximately correct. If it is smaller than that, a magnetic flux command signal φ8 is outputted, and if it is larger than that, a magnetic flux command signal φ8 is outputted.

In addition, as an example of the torque increase/decrease signal T which is the output of the torque comparison means 7, an error jt is set, and if the instantaneous torque value t is within the range of /2 with respect to the torque command value T, Output To as almost appropriate, *
* t-It≦t (should be within the range of r-Jr/2, exceed It* and when smaller, use f, and t +) t/2≦r
(It should be within the range of v''+it, and when it exceeds jt, T is output as the torque increase/decrease command value f.

In this case, the control signal generating means 9 selects the magnetic flux vector v (k) according to the amount of the magnetic flux vector position signal φ according to the magnetic flux increase/decrease signal φ and the torque increase/decrease signal T, as shown in the following switching chain. , turns on and off the switching elements Q1 to Q@ of the three-phase PWM inverter 2. The number in the switching cheeple indicates the value of k of the magnetic flux vector v (k) to be selected, and 0 indicates either the zero voltage vector v (0) or V (13), and the previous state Karasui.

When switching, it is sufficient to select the one with the smaller number of switching elements to be switched.

Switching table With the control method explained above, in normal operating conditions, the tip of the magnetic flux vector φ is located at φ8-noφ/2 and φ/2 inside and outside the circle of torque command value φ shown in Fig. 4. *It rotates in two circles of φ10 and φ/2, allowing smooth control of the induction motor.

[Problem that the invention seeks to solve]

However, in the conventional control method, there is a problem that an unstable operating state occurs when the torque command value t is small during low-speed* operation. Originally, the voltage vector v (k) should be selected according to the vector direction of the magnetic flux vector φ at each time, but due to the problem of calculation time, the circumference is divided into equal integer numbers and divided by the boundary lines. This is because the voltage vector v (k) is selected only at the center position.

For example, in the case of the magnetic flux vector φ shown in FIG. 4, if the magnetic flux increase/decrease signal φ is φ and the torque increase/decrease signal f is t, then if v (k) is selected from the instantaneous direction of the magnetic flux vector
(2), but according to the switching table that selects only the center position of this division, it becomes v(3).

Similarly, when the magnetic flux increase/decrease signal φ is φ, the torque increase/decrease signal t is
In this case, if selected from the instantaneous direction, it would be v(5), but according to the switching table, it would be v(6).

During low-speed operation of the three-phase induction motor 3, the circular motion of the tip of the magnetic flux vector φ is also slow. The locus of the tip of the tip will not be a circular locus and will be greatly distorted. , 90 clockwise from there
The voltage vector v (
k) should be selected by dividing the circumference into equal integer numbers, determining where the magnetic flux vector φ is located divided by the boundary lines, and determining the voltage vector only at the center position divided by the boundary lines. This is because P is selected according to the present invention. Since it is impossible to directly calculate the direction of the instantaneous magnetic flux vector φ every time due to the problem of calculation time.
The method for controlling an induction motor using the WM inverter I is characterized in that the number of discrimination sections that equally divide the circumference of the magnetic flux vector direction discrimination means is an even number, and the number of discrimination sections is made as large as possible by the microprocessor used. It is something to do.

[For production]

The larger the number of discrimination divisions that equally divide the circumference of the magnetic flux vector direction discrimination means, the more accurate the direction of the magnetic flux component Φ and the direction of the torque component element can be known. For example, when it is set to infinity, It is clear that the same result as when directly calculating the direction of the instantaneous magnetic flux vector φ can be obtained.

However, increasing the number of discrimination categories leads to an increase in calculation time, and if the calculation time is too long, the absolute value of magnetic flux 1φ1 or instantaneous torque T may exceed the allowable error range for the respective command value φ or t. Also, since it becomes impossible to change the on/off states of the switching elements Ql-Qs of the three-phase PWM inverter 2, the three-phase induction motor 3 cannot be operated smoothly.

In order to operate the three-phase induction motor 3 smoothly, even when the microprocessor's calculation cycle is maximum, it must be 100 μS.
As a result of experiments, it is known that the temperature should not exceed about ee.

As previously explained as a conventional example with reference to FIG. 4, the boundary line that equally divides the circumference for determining the direction of the magnetic flux vector φ is made to be symmetrical with respect to both the d-axis and the q-axis. If this is done, the sign of the two components φd and φq of the magnetic flux vector can be used to determine which of the 1st to 4th quadrants they belong to, and the boundary line t can be determined based on the value of φq/φd by making each quadrant common.
By comparing with ana, it becomes possible to determine the magnetic flux vector position.

In order to make the boundary line symmetrical with respect to both the d-axis and the q-axis, the number of discrimination sections that equally divide the circumference should be an even number, and the distance between adjacent boundary lines of - It is only necessary to align the center position of the d-axis with the d-axis.

If the boundary line is defined in this way, one boundary line will be added per quadrant every time the number of discrimination sections increases by 4, so one memory for storing the tanθ of the boundary line and one memory for the comparison operation will be required. The number of steps will increase by one. In addition, the number of columns for switching cheaples will increase, so the memory for this will also increase slightly.

Therefore, as long as the calculation time required for one calculation cycle, which is determined by the memory capacity and calculation speed of the microprocessor used, is permissible, by increasing the number of discrimination sections that equally divide the circumference, the magnetic flux component Φ and torque can be more accurately determined. Since the direction of the component elements can be known, the three-phase induction motor 3 can be operated even more smoothly.

〔Example〕

Hereinafter, an embodiment of the method for controlling an induction motor using a PWM inverter according to the present invention will be described in the case where the number of discrimination sections equally dividing the circumference of the magnetic flux vector direction discrimination means is 12.

The only difference in operation from the conventional control method described in FIGS. 1 and 2 is the magnetic flux vector direction determining means 8 and the control signal generating means 9, so only these operations will be described in detail.

FIG. 5 is a direction discrimination classification diagram of the magnetic flux vector direction discrimination means, and FIG. 6 is an explanatory diagram of the voltage vector selection method when the magnetic flux vector direction coincides with the d-axis. Since the number of discrimination sections that equally divides the circumference of the means 8 is 12, the angle between the boundary lines is 3.
06, and the angle between the boundary line closest to the d-axis and the d-axis is 15. The magnetic flux vector direction determining means 8 determines the magnetic flux vector position signal φ1 according to the clockwise rotation angle θ of the magnetic flux vector φ from the d-axis. As follows, one of the 12 types from 1' to 'store' is output as follows.

=15°≦θ<15°;I'15°≦fl<45' : II'45°≦θ<75°;■'75°≦#<105°; γ 105°≦θ<135°; y 135 °≦θ<165°;■'165°≦θ<195°;■'195°≦θ<225°;■'225°≦θ<255':IK'255°≦θ<285° +, X/ 285'≦θ<315': XI'315°≦e<
345° ; W These magnetic flux vector position signals! In order to send out the magnetic flux vector direction determination means tan 15°.

Memorize the values of tan45° and tan75°,
Using the d-axis component φd and the q-axis component φ of the magnetic flux vector φ, from the sign of φd and the sign and absolute value of φq/φd,
Output the magnetic flux vector position signal φi according to the table below. When performing these calculations, if φd or φq is 0, the direction of the voltage vector φ can be determined without calculation, so
Of course, some processing is required.

In the control signal generating means 9, it is assumed that the magnetic flux vector φ is at the center position of the range indicated by each magnetic flux vector position signal φ, that is, at every position of θ from 06 to 30′.
According to the magnetic flux increase/decrease signal φ from the magnetic flux comparison means 6 and the torque increase/decrease signal T from the torque comparison means 7, each switching is performed so that the optimum voltage vector v(k) is selected and the three-phase PWM inverter 2 generates it. Elements Q1 to Q6
Turn on the old man.

For the case where the magnetic flux vector position signal φl is ■',
The method of selecting the voltage vector v (k) will be explained with reference to FIG. In this case, consider that the magnetic flux component Φ coincides with the d-axis, that is, the direction of the voltage vector v(1). Therefore,
By combining the magnetic flux increase/decrease signal φ8 and the torque increase/decrease signal T, the voltage vector v (k) shown in the following table can be selected. - The magnetic flux increase/decrease signal φ8 is φ0, and the torque increase/decrease signal! 8 also selects the zero voltage vector in the case of t, but if it is necessary to switch from this momentary state, select the voltage vector of the switching element to be switched from among the voltage vectors v(0) and v(13). The one with the smaller number should be selected. The above idea is applied to any case where the magnetic flux vector position signal φi is from ■' to summer. As shown in Fig. 5, the directions of the magnetic flux component Φ and the torque component differ depending on the magnetic flux vector direction discrimination area. Therefore, the voltage vector v (k) is selected according to the following new switch, Each switching element of the PWM inverter 2 is turned on and off.

The numbers in the table indicate the value of k of the voltage vector v (k) to be selected, and 0 indicates either the zero voltage vector v (0) or v (13). , when switching from the previous state, it is sufficient to select the one with fewer switching elements to be switched.

〔Effect of the invention〕

In the above embodiment, the case where the number of discrimination sections for equally dividing the circumference of the magnetic flux vector direction discriminating means 8 was 12 was explained, but the number of discrimination sections is an even number and depending on the microprocessor used, By increasing the capacity and calculation time as much as possible, the method for controlling an induction motor using a PWM inverter according to the present invention can ensure smooth and stable operation.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing a method of controlling an induction motor using instantaneous space vector control, and Figure 2 is a three-phase PWM shown in Figure 1.
The main circuit connection diagram of the inverter section, FIG. 3 is a voltage vector diagram, FIG. 4 is a magnetic flux vector diagram, and FIG. 5 is a magnetic flux vector direction in an embodiment of the method for controlling an induction motor using a PWM inverter according to the present invention. The direction discrimination classification diagram of the discrimination means, FIG. 6, is an explanatory diagram of the voltage vector selection method when the magnetic flux vector direction coincides with the d-axis. 1...DC voltage source, 2...3 phase PWM
Inverter, 3...3-phase induction motor, 4...
... magnetic flux calculation means, 4a ... voltage vector calculation means, 4b ... current vector calculation means, 5.
... Torque calculation means, 6 ... Magnetic flux calculation means, 7 ... Torque comparison means, 8 ... Magnetic flux vector direction determination means, 9 ... - Control signal generation means.

Claims (1)

[Claims] means for calculating the instantaneous value of the magnetic flux vector by converting the voltage and current of the three-phase induction motor into space vector values; means for calculating the instantaneous value of the torque from the magnetic flux vector and the current vector; magnetic flux comparing means for comparing the magnitude of the magnetic flux vector; torque comparing means for comparing the torque command value and the instantaneous value of the torque; magnetic flux vector direction determining means for determining the direction of the magnetic flux vector;
Utilizing a microprocessor comprising control signal generating means for generating switching signals to six switching elements constituting a three-phase PWM inverter based on the outputs of the magnetic flux comparing means, torque comparing means and magnetic flux vector direction determining means. The control device turns on one of the positive and negative switching elements of at least two or more phases, which are different, among the three-phase positive and negative switching elements constituting the three-phase PWM inverter, or turns on one of the three phases, which is equal to either positive or negative. In the PWM control method in which an element is turned on and all others are turned off, the number of discrimination divisions equally dividing the circumference of the magnetic flux vector direction discrimination means is an even number and is as large as possible by the microprocessor. A method of controlling an induction motor using an inverter.