JPS59110394A - Vector control system for induction motor - Google Patents

Abstract

PURPOSE:To reduce calculating steps of necessary data by calculating the output of a speed system obtained by compensating the deviation between a rotary speed instruction value and actual rotary speed data for the instruction value representing the torque current as slip angular frequency data. CONSTITUTION:A frequency command fref inputted from a frequency command input 1 is supplied to a subtractor 2 to obtain a deviation from the actual rotating frequency signal fM inputted from an actual rotating frequency input 16. This deviation is compensated for PI by a speed detecting system 3, which outputs data WS representing the slip angle frequency. On the other hand, the signal fM is supplied to a magnetic flux command generator 6, which generates a magnetic flux command phi*. The command phi* is processed by a subtractor 7, a magnetic flux control system 8 and a magnetic flux calculator 9 to obtain data Im representing the exciting current and data phi representing the secondary magnetic flux. The signal WS is multiplied by K3 times by an amplifier 5, and multiplied by a multiplier 20 by data phi to produce as data I2 representing the torque current component. An induction motor is controlled by vector via data I2, Im, thereby reducing the calculating steps.

Description

[Detailed description of the invention] [Field of application of promotion] The present invention is directed to i\? This article relates to a vector control method for induction motors that has excellent I-speed response, enables high-speed control, and is suitable for digitalization. FM and motor rotational speed control systems using stationary inverters for motors have become widely used, and many control systems employing the vector control method have also begun to be seen. For the control of the induced electric current!l!II orange using the mouth vector division 1 method, refer to Figure 1(a), ω, = [eI2/φ... Song... Song... Between songs.・
Song... Song... (1) ■u=, /"i", 2+1=
2・・・・・・・・・・・・・・・・・・・・・・・・
・・・・・・・・・・・・・・・(2Jθ= tall
-” (I 2 / Im)・・・・・・・・・・・・
...Song...Song...(3)0It is well known that the music is performed based on three formulas.

ωB: Slip angular frequency K: Constant determined by induction motor φ: Secondary magnetic flux Il knee current I2: Torque current ■m: Excitation current θ Knee current work 1 and excitation current work. phase angle between It is.

FIG. 2 shows an example of a conventional induction motor control method using such vector control.

In this Figure 2, 1 is a frequency command input, 2.7 is a subtracter, 3 is a speed control system, 4.11 is a divider, 5 is an amplifier,
6 is a magnetic flux command generator, 8 is a magnetic flux control system, 9 is a magnetic flux calculator, 10 is a root function generator, 12 is an arctangent function generator,
13 is a primary current phase calculator, 14.15 is a D/A (digital-analog) converter, and 16 is an actual rotation frequency input.Next, the operation will be explained.

The frequency command fref input from the frequency command input l is supplied to a subtracter 2, and the deviation from the actual rotation frequency signal fy input from the actual rotation frequency input 16 is calculated.

This deviation is determined by a Df constant compensator in the speed detection system (PI compensator), and is output as the torque current component 2, which is -H'-ta I2.

On the other hand, the signal fw from the input 16 is also supplied to the magnetic flux command generator 6, and a magnetic flux command φ0 that changes in a predetermined pattern according to the signal fM is generated. This command φ0 is processed by a subtracter 7, a magnetic flux control system (PI supplementary inno' unit) 8, and a magnetic flux calculator 9, and data Im representing the excitation current and data φ representing the secondary magnetic flux are obtained.

In this way, when data I 2 and Iz nφ are obtained,
The data I2 and φ are supplied to the divider 4, and the output is multiplied by amplifying device 5 to obtain the data ω8 representing the slip angular frequency (or the data representing the slip frequency) according to the above equation (1). fS) Obtain V.

Similarly, the data I2 and Im are input to the root function generator 1°, and data 11 representing the first-order ``fiL flow thick reference value'' is obtained according to equation (2).

Furthermore, the data 2 and Im are also input to the divider 11, and the output r-taper I2/Im is input to the arctangent function generator 12 to obtain phase angle data θ according to equation (3).

This data θ is then converted to the angle fMz from input 16.
It is inputted together with the data fs (ω,) to the primary current potential calculator 13 consisting of a microcomputer, etc., and the negative current phase 61 is calculated as [θ1=2π, t(Ja+j'u)dt
+θ], and further, using this phase θ1, the trigonometric function DELUN is used to obtain data sub01 and ■θ summer.

Then, these garter θ1 and μs01 are input to the D/A converter 14.15 along with data 11, and data
By multiplying by 1 and converting the analog coefficient, the two-phase primary power 6iY required for vector control and the reference instantaneous value signal iα are obtained.
, iβ are obtained and supplied to a drive circuit of a frequency converter such as an inverter, rotational speed control 7 of the visiting electric motor is performed by vector control in accordance with predetermined torque characteristics and output characteristics determined by the pattern of the magnetic flux command generator 6.
I can do what I do.

In addition, in the conventional example shown in FIG. 2, the signal processing is configured to be performed by hardware, or the large rlu 't+ signal processing 1 is performed using a microcomputer or the like.
．． 1i! ``Items that are designed to be performed using software are well known, and those that are constructed using analog circuits are also known.

By the way, in this conventional vector control method, there are 2 division processes in the signal processing, so if it is configured with analog circuits or digital circuit hardware, the divider is 2 i + i >I is required, which has the disadvantage of increasing costs, and if h7 is implemented as software using a microcomputer, it requires a lot of calculation processing time and the response speed of control tends to decrease. [Object of the Invention] The object of the present invention is to provide a vector GIjlJ control system for an induction motor that eliminates the drawbacks of the prior art described above, is low cost, and can provide sufficient control response characteristics [Summary of the Invention] Achieving this object. Therefore, the present invention extracts the output angular frequency data ω8 of the speed control system obtained by performing predetermined compensation for the deviation between the rotational speed command value and the actual rotational speed data, and calculates the basic vector control described above. By dividing the data IzY by 1 using the following equation (4), which is derived by modifying equation (1) of the equations, the division process is made % 9 so that only one division process is required.

1,=to, ・φ・ω8 ・・・・・・・・・4・・
・・・・・・・・・・・・・・・・・・・・・(4) However, K3: 1/K [Embodiment of the invention] Hereinafter, the induction motor-like vector control method according to the present invention will be explained. Examples will be described with reference to the drawings.

FIG. 3 shows an embodiment of the present invention, in which, like the conventional example shown in FIG. 2, most of the processing of digital signals is performed by hardware. In the figure, 20 is a multiplier; Other blocks are the same as the conventional example shown in FIG.

In this embodiment, the output of the speed detection system 3 is directly input to the primary current phase calculator 13 and processed as data ω representing the slip angular frequency.
The calculation of lnθl +Cmθ1 is performed from line 1.

As a result, the torque current is calculated as follows: 1-Data I
2 is such that the output of the speed detection system 3 is amplified three times by the amplifier 5 and then multiplied by the data turn by the multiplier 20 according to the above-mentioned equation (4), thereby obtaining the data (2).

Therefore, in this embodiment, as well as in the conventional example shown in FIG. Since it is only necessary to calculate the data x, 2 / I□ by
The drawbacks of the prior art can be eliminated.

Next, Figure 8TS4 shows another embodiment of the present invention, in which most of the data processing is done digitally using a microcomputer or the like. , 30° 35 is a subtraction process, 31 is a speed control process, 32 is an amplification process, 33 is a division process, 34
is magnetic flux processing, 36 is magnetic flux control processing, 37 is magnetic flux control processing, 38.39 is function table processing, 40 is integral processing, 4
1.46 is an addition process, 42.43 is a trigonometric function table process, 44.45 is a D/A converter, 47 is a rotational frequency command input register, and 48 is an actual rotational frequency input register. Therefore, in the embodiment of FIG. 4, the frequency command f re
From the acquisition of f, the integral of the primary angular frequency ωI - the primary current phase θ
The addition of the above data ω1 and the integral value of the data ω1, and the trigonometric function table processing of the tangent and cosine corresponding to the output data from this addition are performed by digital processing using a microcomputer, etc., and the subsequent conversion from deinotal data to analog data is In other words, in this embodiment, the speed control system and the field weakening control system are processed digitally by a major rouzo fft microcomputer, etc., and the minor luzo current, which requires high-speed processing, is processed in an analog manner. The control system uses analog processing.

Next, the operation of this embodiment will be explained.

The deviation (frefM) between the frequency I wave number command Jref inputted in the frequency command input register 47 and the actual rotational frequency fM inputted in the actual rotational frequency input register 48 is extracted by subtraction processing 30, and the content is set to PI assistant etc. The slip angular frequency ω8 is calculated in the quick control process 31.

On the other hand, the magnetic flux command processing 34 uses the input actual rotation frequency fs.
The magnetic flux command φ0 is generated in accordance with the magnetic flux command φ0, and the secondary magnetic flux φ obtained by calculation in the magnetic flux calculation process 37 is performed in the process 35. After that, the magnetic flux control process 36 with the same PI compensation is performed to excite the current I□. , put out.

Once the slip frequency ω8, magnetic flux φ, and excitation current Calculate the torque 'current' according to ◎ Up to this point, it is the same as the embodiment shown in Fig. 3, and as a result, the divider 4 in the conventional example shown in Fig. 2 can be omitted, reducing costs and improving response speed. will be 1(I).

In the embodiment shown in FIG. 4, unlike the embodiment shown in FIG. Rangata I as the result of the calculation
2/Im Y is calculated, and this data is applied to the root function table as the 2/In1 variable, and further data ■.

A 40-table processing 38 is provided to obtain a linear value 11 by multiplication of , and this is for the following reason.

(・Ma, formula (2) of the above basic formulas 'a' f form ding, It=ImJ ding Tte]7f]=) r・river・naka・...・
...(5) is derived.

Therefore, as in the conventional example shown in Fig. 2 and the embodiment shown in Fig. 3,
It can be seen that instead of performing a table search using data E2 and Im to obtain data 11, first obtain the quotient data I 2 /Im, perform a table search using this as a variable, and then perform multiplication by data Im.

After the data Il e I 2/Imeω8 is calculated in this way, the data I2/Im is processed by the function table processing 39, as in the conventional example shown in FIG. 2 and the embodiment shown in FIG.
Find the primary current phase θ using the table. Also, data ωM
and ω8 become the primary angular frequency ρ1 in the subtraction process 46, are integrated in the integration process 40, and then are operated with the data θ in the addition process 41 to become the data θl=(ωit 10), and the trigonometric function table process 42° At 43, the data sinθl+cO
Find 8θl. Finally, data II l mountain θl
+Cmθl causes analog control signals tα and iβ for vector control to be taken out from the D/A converters 44 and 45. 11. From the above explanation, it can be seen that the integral processing 40, addition processing 41, trigonometric function table processing 42, 43, and addition processing 46 in the embodiment of FIG. 4 are the primary current phase calculator 13 in the embodiment of FIG. It is understood that this corresponds to the processing content of .

Therefore, according to this embodiment, as mentioned above, the removal '!
In addition to being able to reduce the number of processing steps and eliminating the drawbacks of the prior art, the following effects can also be obtained.

First, function table processing 38 to obtain data E1
is the arithmetic processing shown in equation (5), and accordingly the table is 100T, I2/IJ", variable force I2/Irn
Therefore, two Lj number table processing 38.3
9 becomes a search using the same variable, which speeds up the calculation (and reduces processing time. In other words, in the conventional example shown in Figure 2, data ■2 and ■□ are used as variables, and the root function table is searched from now on. Therefore, first, 1
Since the processing involves performing the table search after performing the calculation of '-data (Eng22+Irr1'), there are many calculation steps (resulting in a long processing time), but according to the above embodiment, , so the processing time can be shortened.

Next, if the method is to perform calculations on data (I2"+I♂) as data I2+ImY variables, as in the conventional example, then the bits (I2"+I♂) of data (I22+In12) are the respective bits of data ■2 and Since it is the sum of a, a huge amount of data is required to maintain control accuracy.
Maintaining precise control becomes difficult. According to the embodiment shown in FIG. 4, the number of bits of the f number I2/Im is equal to the number of bits of the data amount 2 # IIn, respectively.
If it is 6 bits, it will be the same 16 bits, so
There is no need to particularly reduce precision, and high control precision can be maintained.

In addition, in the conventional example shown in Fig. 2, data IxY is directly obtained from the table %, so if the rating of the induction motor to be controlled differs, the function table pattern must be changed accordingly. It is difficult to standardize and generalize tables, which tends to increase costs. According to the embodiment shown in FIG. 4, the root function table in the function table processing 38? Data 11 can be determined by normalizing the turns and adding 4 as a constant to the right side of equation (5) above.
According to , by simply changing the constant 4, it is possible to control induction motors of any rating with just one type of root function table, which reduces costs due to the generalization of equipment. You can fully expect it.

In this case, the above equation (5) becomes the following equation (6).

11=4 ・Im J 1 + (I, /
Im)2 ・---・--・・・・・・ (6 to 4: constant determined by the rating of the induction motor to be controlled) By the way, in such a vector control method, the output of the induction motor relative to its rotational speed is The control gloss turn is determined as shown in Figure 5, and when the speed is below N/l''-Na, it has constant torque characteristics, in the range of Na (N≦Nb, it has constant output characteristics, and in the range of N>Nb, it has output reduction characteristics. In many cases, it will be done.

If such a control pattern is adopted, the magnetic flux command φ0 in the constant torque control region will be maintained at a constant value, and therefore, in this case, each process 34, The amount of data obtained by 35°36.37 and φ are also constant values.

Therefore, in this case, since the variables φ and - in the above equation (4) have constant values, the variable I 2 / ImY K 1 ・ω can be set by setting 1 to K3・φ/Im=, and the above The root function in process 38 is J1+(K2
O2a and the arctangent function in process 39 are bn (KI
O3), respectively, and the division process 33 can be eliminated to further shorten the processing time.

Therefore, the operation of the embodiment shown in FIG. 4, including this case, is shown in FIG. 6 using a 70-chart.

In Figure 6, in the constant torque region of Figure 5, %
Since Im #φ is a constant value, the process 55 is not performed. Therefore, the processing of 56 I 2 /I□=3・φ・ω
Since Im and φ are constant in s/Im, 3・φ/I
Set m=3, set the variable (I 2 /Im) to ω3
Perform the process of converting to .

Moreover, FIG. 7 shows Itj in the field weakening region of FIG.
: Detailed flowchart of L output. Here, 4 in 573 is 1, and since the function table is standardized, 4 can be adjusted depending on the rated current value of the induction motor to be used, and the number of compatible models can be expanded.

Furthermore, FIG. 8 shows 11 in the constant torque region of FIG.
It is a detailed flowchart of calculation. In this region, the calculation is simplified by setting I2/Irrl=Kloω8 as described above.

Similarly, since Im at 572 in FIG. 7 is constant, 4@Im=2 is used to simplify the process as at 542 in FIG. 8.

According to an actual measurement result using the embodiment shown in FIG. 6, the processing time is reduced by about 10% compared to the conventional control method.
Particularly in the constant torque region, the processing time was reduced by as much as 50%, and the response characteristics were significantly improved.

〔Effect of the invention〕

As explained above, according to the present invention, it is possible to reduce the number of arithmetic processing steps for data required for vector control, and it is also possible to avoid an increase in the number of data bits required for arithmetic processing when digitized. Therefore, excluding the drawbacks of the conventional technology, the configuration is simple and can be easily reduced in cost, and there is no drop in control accuracy during digitalization, making it suitable for digitalization, and the control response speed is high (and high-speed control is possible). Possible attraction; easy to provide vector control method of motive (part)
Be able to do that.

[Brief explanation of drawings]

Figure 1 is an explanatory diagram showing the current vector of an induction motor, Figure 2
FIG. 3 is a block diagram showing a conventional example of a vector control method for an induction motor, FIG. 3 is a block diagram showing an embodiment of the vector control method for an induction motor according to the present invention, and FIG. 4 shows another embodiment. Block diagram, Figure 5 is an explanatory diagram showing an example of the control mode corresponding to the rotation speed of the induction motor, Figure 6
The figure is a flowchart showing the operation of a practical example of the present invention, and FIGS. 7 and 8 are flowcharts showing a part of FIG. 6 in detail. ■... - Frequency command input, 2, 7... Subtractor, 3...
・Speed control system, 4.11... Divider, 5... Amplifier, 6... (Kurato command generation profit, 8... Magnetic flux control system, 9
. . . Magnetic flux calculation generator, 13 . . . Primary current phase calculator, 14. 15 . . . El/A converter, 16 . 35... Subtraction processing, 31... Speed control processing, 32... Amplification processing,
33... Division processing, 34... Magnetic flux command processing, 36.
... Magnetic flux control processing, 37... Magnetic flux calculation processing, 38.3
9... Function table processing, 40... Integral processing, 41
．． 46... Addition processing, 42° 43... Trigonometric function table processing, 44.45... D/A converter, 47...
- Rotation frequency command large sword register, 48...Actual rotation frequency input register. Kaya 6 eyes Kaya 7 eyes $8 Mouth

Claims (1)

[Claims] 1. An induction motor in which the primary current of the induction motor is separated into a torque current component and an excitation current component, and each is controlled according to the deviation between a rotation speed command value and actual rotation speed data. In the vector control method, the deviation between the rotational speed command value and the actual rotational speed data is determined, and the output data of the speed control system is calculated by performing correction χ such as PI (proportional + integral) compensation and outputting the result. Berry angular frequency data ω8
The vector of an induction motor is characterized in that the command value (2) representing the torque current is calculated from this alignment angular frequency data ω8 using a basic vector control formula (I2=ni・φ・ω8). Open side 10,000 types. K: Constant determined by the induction motor to be controlled. φ=M” Im/(1+TS) M Mutual inductance between the primary and secondary paths of the induction motor to be controlled Im: Excitation current command value T: Secondary time constant S of the induction motor to be controlled: Differential operator 2. In claim 1, the data 1 representing the reference value B of the secondary current is the excitation current command value Im and the division value of the torque current command value 2 by this command value Im. Daple data (t
A vector control method for a zigzag motor, characterized in that the vector control method is configured to be obtained by calculating +('xz/'mdoor). 3. In claim 2, when the control state of the induction motor to be controlled is in the constant torque control region, the variable (Iz/Im') of the table data is equal to the angular frequency data ω8. The i number Kl・ω. 1. A vector control system for an induction motor, characterized in that it is configured so that: However, K1 = *φ1・Imi φ1 2.φ IrrI! when the induction motor to be controlled is in the constant torque control region. -Similarly, I when in the constant torque control region
m