JPS62160089A - Vector controller for induction motor - Google Patents

Abstract

PURPOSE:To obtain a high speed torque response by performing a vector calculation by converting in polar coordinates to reduce the number of calculations. CONSTITUTION:A vector calculator 13 performs a vector calculation on the basis of a torque current command value IT* and an exciting current command value I0*, a slip angle frequency omegaS and the phase variation amount DELTAphi of a magnetic flux. A primary current waveform output unit 16 produces and outputs a sinusoidal data of the primary current stored in a data table of a CPU9 on the basis of an interrupt signal INTR input from a 1/n frequency divider 18, a primary angle frequency omega0 calculated by a primary angle fre quency calculator 12, and a phase variation amount DELTAphi input from the calculator 13. A multiplication type digital-to-analog converter 15 obtains U-phase and V-phase primary current command values IU* and IV* from the value I1* and the sinusoidal data.

Description

[Detailed description of the invention] Human and industrial applications The present invention relates to a vector control device for an induction motor.

B6 Summary of the Invention The present invention provides a vector control device for a single induction motor that controls a motor voltage by dividing it into a magnetic flux axis component and a torque axis component. Store the data in a data table and calculate the
The vector operation Frequency ωs and induction motor, rotational angular frequency ω of the machine
An Ichiya frequency signal is output based on n, a pulse signal that is a ratio of this output signal is divided according to the magnitude of the primary angular frequency signal, and when this frequency-divided signal is introduced, the Ichiya moon frequency signal is The U and V phase sine wave data according to the polarity is looked up and retrieved from the data table, and if the phase of the magnetic flux changes due to a change in torque, the lookup operation is stopped. Then, the sine wave data of the U and ■ phases corresponding to the phase change amount Δψ are extracted from the data table, and the extracted U and ■ phase sine wave data are
Based on the phase positive cW, tIjL data and the above-mentioned -arrow current indicator and the output ti detection amount of the inverter for driving the induction motor, the 8-phase tonnage pressure indicator vU, vvt vw t
``Determine the sum of these voltages Ill! VU*.

＊＊＊ * Pulse width modulation control of the induction motor drive inverter based on VV and V1/D not only provides high-speed torque response but also enables highly accurate speed control over a wide frequency range. In addition, it allows for a margin in calculation processing time and eliminates the need for a high-precision analog processing circuit.

C0 Prior Art Conventionally, for example, DC motors have been used as variable speed electric motors that require a faster response than servo motors and a wide range of speed control. It is used. However, this DC motor
The brushes wear out, so troublesome maintenance is required. For this reason, in recent years, AC variable speed generators, such as induction motors, which are a combination of a brushless AC motor and an inverter, have been used as servo motors. As a control method for this induction motor, the primary current of the motor is controlled by dividing it into an exciting current (a magnetizing current to generate all the magnetic flux) and a secondary current (a torque current that contributes to torque generation). By making the vectors of the secondary magnetic flux and the secondary Ef & always perpendicular to each other
A vector control method has been proposed to achieve responsiveness equivalent to that of a C motor.

Problems to be Solved by the Invention In the vector control system as described above, when the number of vector calculations increases, a coherent torque response cannot be obtained, and as a result, speed control with high-speed response is no longer possible.

The present invention has been made in view of the above-mentioned points, and is a vector control device for an dielectric motor that can obtain high-speed torque response and perform highly accurate speed control over a wide frequency range.
T-It is for the purpose of providing money.

Means for Solving Problem B The present invention provides a torque current obtained by proportional integration of the difference output between the rotational angular frequency ωs of the induction electric la machine and the rotational angular frequency command 1st rotation ω♂. Command value T* and excitation current command value. A vector calculation section which performs a predetermined vector calculation based on the rotational angular frequency ω and calculates the amount of phase change Δψ of the magnetic flux by calculating the current command value, *, and the angular frequency ωs. Ichiku angular frequency ω obtained by adding the Vesi angular frequency ωs shown below. The absolute value of |ω0| is calculated, and the calculated absolute value fl
I[Iω. When 1 is greater than the set value, the corresponding 1ω. a primary angular frequency calculation unit that outputs a signal of 1, and outputs a signal obtained by multiplying the |ω0| by n when the calculated absolute +1 [|ω0| is less than a set value; a pulse signal generating section that generates a pulse signal with a frequency proportional to the output signal of the -arrow angle frequency ω; When the absolute value of |ω0| is greater than or equal to the setting IIL, the corresponding 1ω. The frequency division ratio is determined according to the magnitude of 1, and the primary angular frequency ω. The absolute value of 1ω. a frequency division section that determines a frequency division ratio to be 1/D when I is less than a set value, and divides the output pulse signal of the pulse signal generation section according to the determined division ratio;

A data table storing data of U- and V-phase sine wave signals divided into a predetermined number, and the output number 1 of the pulse signal generation section whose frequency has been divided by the frequency dividing section by degrees Celsius are input. When, the -arrow frequency ω. When the sine inverse data of the U and V phases are extracted from the data table according to the polarity of a primary current waveform output unit that extracts U- and V-phase sine wave data from the data table according to the change in the voltage, and outputs the extracted U- and V-phase sine wave data with a delay of a predetermined time; - Multiply the output data of the secondary current waveform output section and the primary output from the Begnomit operator section by a current reference value of 11ゝ to calculate the current command value of the U and V phases U*r. A multiplier for calculating v*, and a three-phase voltage command value vU, vV for the output of the inverter for driving the induction motor by the current detection amount and the output of the multiplier.
The present invention is characterized in that it includes a phase voltage calculation unit that calculates the phase voltage, and a PWM control unit that performs pulse width modulation control of the inverter for driving the induction motor in accordance with the output of the phase voltage calculation unit.

F. Example Fl Explanation of the principle and the drawings'! ! An embodiment of the present invention will be described with reference to e-. First, the principle for making the vectors of the secondary magnetic flux and the secondary current always orthogonal will be described with reference to the vector diagram in Figure 42. What if the excitation flow command value 0* is kept constant on the a-q synchronous rotation coordinates shown in Fig. 2? X, the angular circumference angular frequency is determined by the above equation.

ω0=ωn+ωs ・・・・・・2) (However, ω
S is the vesi angular frequency, ω. is the rotation angle mantissa s
is the torque current command value, and τ2 is the second-order time constant.

Here, it is expressed by a field equation converted from a-q axes to 8-phase fixed coordinates.

r,=r. tψ ・・・・・・(311,=i了7T ・・・・・・(4) ψ:: tan (IT /ro) ”・・・
・(5) (However, 〒1 is the primary current, and ψ is the excitation current. It is used to sandwich * and -vct current i.

From the above formulas +31, +41, and 151, the three-phase fixed axis current "U + T, , I" can be calculated.The relationship between the -order current i1 and the three-phase current IU r "/ + machining is 7J.
f, as is clear from Figure 2. Here, 0 is the function made by the U-phase axis and the single-arrow wL flow vector i1, and since the d-q axes rotate at an angular frequency ω0, in a steady state where ψ does not change, σ=ωG. However, when the torque [flow command TT* changes, the phase angle ψ changes, so this change affects the phase angle 0, resulting in the following equation.

θ=ωot+Δψ Song...(7) (However, Δψ is the variation of ψ.) In the above equation (7), Δψ is the amount of phase change in one sample period in digital calculation, so it can be modified into the following equation. Desired.

Δψ=ψn−ψfi−1 ・・・・・・1.8)
(However, ψゎ is the current sample value ψ, and ψn-1 is the previous sample value ψ.) All equations 11) to (8) above are calculated, and the inverter for driving the induction motor is calculated based on the calculation results. If controlled, the vectors of the secondary magnetic flux and the secondary current can always be orthogonal, and the torque 'flow direction' can be directly controlled. The torque can be controlled in a curved manner.

Ft Description of the overall configuration The control system in which the calculations of equations (11 to (8)) are carried out will be explained in detail with reference to the circuit diagram in FIG. 1. In FIG. The operation is controlled by an inverter 2.The rotation speed of the induction motor 1 is converted into an electric signal proportional to the rotation angle by a resolver 3, and then a central processing unit t (hereinafter referred to as CPU) 9
is supplied to the speed calculation section 10 of. The speed calculation unit 10 calculates the rotation angular frequency ω based on the electric signal proportional to the rotation angle input from the resolver 3, and supplies it to the speed control p-factor calculation unit 11 and the angular frequency calculation unit 12. .

Said PI calculation unit 11.7:1. , take the difference between the preset rotational angular frequency ωn and the rotational angular frequency ωno, and proportionally integrate the palate output to calculate the torque.
Calculate. That is, (ω♂−ωn)・Kp(1+−
') TL-8 to obtain IT (where K is the gain, TL is the time constant, and S is the Laplace variable). This torque flow command value T is input to the vector calculation section 13. This vector calculation unit 18 calculates an excitation current index value I set to a constant value. Based on * and the torque current command value IT, perform the calculation X shown in FIG.
angular frequency ωs. The phase angle ψ of the magnetic flux and the amount of phase change Δψ of the i flux are determined.

In other words, in Fig. 3, the excitation current command value is set. ' and the torque current command value IT to calculate the above-mentioned equation (4) to obtain the secondary current command II T , *e, and calculate the above-mentioned step T*. Using the quotient divided by , it takes 9 steps to calculate each of the above formulas (51) and 9. Calculate the number ω1 of the phase ψ, and use the phase angle ψ to calculate the equation 18 above. The amount of phase change Δψ is determined by . The above = connected velocity calculation, pr calculation, and vector calculation are CP
This is performed by U9, and the calculation procedure is as follows in the fourth
Steps 81 to S of the flowchart shown in the figure,
Work step S1. It is carried out in accordance with It is assumed that the vector calculation section 13 is activated periodically by, for example, an interval timer, and performs calculations at fixed intervals (for example, about 500 μsec) that do not affect the control response. The slip angular frequency ωS of the vector calculation unit 13 is inputted to the Ikkuku angular frequency calculation unit 12, and the next flow command value I is converted to an analog signal by the digital-to-analog converter 14. After that, it is input to the multiplication type digital-to-analog converter 15, and the phase change amount Δψ is input to the primary current waveform output section 16.

The -vcP1 frequency calculation section 12 adds the rotational angular frequency calculated by the speed calculation section 10 and the angular frequency ωs calculated by the vector calculation section 18 to obtain the first angular frequency ω. , and its absolute directivity |ω0| is calculated. Primary angular frequency ω.

absolute ([1ω.1 is the digital-to-analog converter 14
After being converted into an analog signal by the V/F converter 17
supplied to This V/F converter 17 has the absolute value 1ω.
Outputs a pulse signal with a frequency proportional to o1. v/F
The output pulse signal of the converter 17 is frequency-divided by an l/n frequency divider 18 whose frequency division ratio can be arbitrarily set in the software described later, and then outputted as an interrupt signal lNTR with a −arrow current waveform. The signal is supplied to the output section 16. Nao l/. The frequency divider 11 is composed of, for example, a programmable counter.

- The arrow current waveform output section 16 outputs the interrupt signal rNTR inputted from the 14 frequency divider 18 and the truncated circle frequency ω calculated by the primary angular frequency calculation section 12. Based on the amount of phase change Δψ inputted from the vector calculation unit 13, the sine wave data of the primary current stored in advance in the data table of the CPU 9 is retrieved and outputted. At that time, the procedure j@ is carried out according to the flowchart shown in FIG. That is, 1in.
The interrupt signal rNTR of the frequency divider 18 is used as an external interrupt signal C! When input to the PU 9, the routine shown in FIG. 5 is activated. First, in step S, the unoccupied circle frequency ω is determined. Determine the audibility of N: N in step St if positive.
+ n gold, if negative step 8. In N=:N
-n respectively, N is the call number of the data table,
n=1.2.4...), and two of these are the primary angular frequency ω. Determine the calling number of the data table according to the physical properties of. Meanwhile, in step S, 5L-n data corresponding to the call number N is extracted by referring to the sin data table tS. Then, in step S, U-phase current data I U =: 5 in (#H).

V-phase current data Iy = sin ('y--π), respectively, as s:ar.

Then, "Uk" is outputted as, for example, lower 8 bit data D0 to D7, and %Iv' is outputted as, for example, upper 8 bit data D8 to D7.
Output as IS. Incidentally, the sine wave data is divided into a predetermined number of divisions, for example 860, and stored in the order of call number H in the sine data table, as shown in FIG. 6, for example.

Here, the digital-to-analog converter 14 is used to output the absolute directivity 1ωo1 of the primary angular frequency, e.g.
oo : To control the speed in the range of 1, the converter 1
4 must be configured with 13 bits or more (12-bit digital-to-analog converters are common),
Equipment prices have soared and construction is cumbersome. Also, if the digital to analog converter output at rated time is 5V, the a range by speed is xo
ooo: If it is 1, the output will be 0.5a+V, which is a small level. Therefore, the analog output signal of the digital-to-analog converter 14 may be affected by noise. Therefore, in the present invention, when the primary angular frequency is less than a predetermined value and the rotation speed is low, |ω0| supplied to the digital-to-analog converter 14 is nXIω. 1, and the frequency division ratio of the 1-inch frequency divider 18 is determined to be 61/n. I decided to work on O! The analog level of the output signal of the digital-to-analog converter 14 is n times higher9.
, the shadow of noise 19 is no longer received (in this case, the call number N of the data table is set to NUN±1). The determination as to whether or not the omitted circle frequency is less than a predetermined value of 1 is based on the absolute value of the primary angular frequency, 1ω, as shown in Table 1 below. 1 and the rated hourly angular frequency ωRatl (1ω, r) is, for example, 1151
Test 1 to see if it is smaller than 2.

Table 1 In Table 1 above, when n is less than 11512, n is set to n=16, for example, but it is not limited to this and may be set to other values. The above-mentioned calculation of 1ω0bi, determination operation, and determination of the frequency division ratio are performed in accordance with steps S--Sta of the flowchart of FIG.

That is, first step S, smell τω. and ωstl”
Add it up to get the unoccupied circle frequency ω. and its absolute platform 1ωo1, and then calculate the absolute [1ω. 1 and the rated hourly angular frequency ωRa
The ratio 1ω of t and e, l'=lω""Ra te is calculated. In step S6, it is calculated as described above and it is determined whether or not 91ω(I is smaller than a predetermined value, for example, 11512. Said 1ω.I'l)X11
If it is smaller than 512 degrees, the division ratio of the l/n frequency divider 18 is set to k''/1s in step S, and a signal with a primary angular frequency of 1ω.lx16 is output to the digital-to-analog converter 14 in step S. If 1ω.1' is 11512 or more, in step 8, 1, the signal of the angular frequency a|ω0| is output as it is to the digital-to-analog converter 14. Then, in τ step 81, Ifol = lω01/ 2π (lIL calculation is performed to calculate the absolute value 1f01 of the primary frequency of the induction motor. In step StS, the -9 frequency division ratio is set in advance according to the magnitude of 00 at the -missing frequency as shown in Table 2). At the same time, the 1/n frequency divider 18 is set to the determined division ratio.

Table 2 If 1ωIF is above '1512u as shown above,
- The number of doji conversions f of the reference sine wave according to the increase in the missing frequency f0
t360/n (n = 1, 2, 4'...),
At the same time, the output side of the V7F converter 17 performs frequency division by 1/n to create the interrupt signal lNTR, so -
It is not the case that when the frequency 10 increases and the frequency of the output pulse No. 1 (signal including 1llJ) rNTR of the 1/n frequency divider 18 increases and the period of this signal becomes shorter, the frequency of the output pulse 1 of the 1/n frequency divider 18 increases. For example, the frequency f lNTR of the interrupt signal lNTR is determined by ωG and the quantization number of the sine wave data, so 'lNTR'
86”n
0==25Hz, n=lk substituting frNT
R=9 KHz. Also, f is added to equation (9) as shown in Table 2. =100Hz.

D: 4 When f is substituted, the above f0=25H
As in the case of zNTR, the frequency is 9 KHz, and the interrupt signal
The period of R remains the same at 111 μsec. Therefore, the processing time for executing the flow NTR chart shown in FIG. 5 in which the interrupt signal fK is activated is 15 ttse.
Since it is about c, it is a -missing frequency. It can be seen that even if the amount increases, the entire processing time can be ensured sufficiently.

Work on this C! No overhead phenomenon of PU9 occurs.

In addition, the rotation speed of the induction motor is extremely low (1ω(I
is less than 11512), - 1 shift f D ! Therefore, digital A
The analog output signal level of the t-analog converter 14 can be kept large. For example, when the output of the digital-to-analog converter 14 at the rated time is 5, and the speed control range is 10,000:1, ni is as shown in Table 1 (o,
= 16, the analog output signal Vbell of said converter 14 is 5"10000X, 1ft=8mV
becomes. Also~. Immediately after the value of 1' exceeds 11512, if the rated output is 5V, the analog output signal level of the digital-to-analog converter 14 can be increased in this way. The construction 510 consists of a digital-to-analog converter.
Speed control can be performed within a range of about 000:1.

Note that the reference sine wave data is 3flO/n (360, 180.

There is no need to have a data table for each step S! As in es1, the total X of data table number N is N=N:l:n (normal rotation; +,
Reverse S-). Example λban = 2 (1/2 frequency division)
In this case, the data table number N is 360 as shown in FIG.
Since each piece of doji-ized data is multiplied by two at a time, it is the same as having 180 data tables.

Here, if a rainbow phase change amount Δψ occurs due to the change in the torque T8, and the primary is input to the linear output section 16, the interrupt signal (I NTR) is The activated routine of FIG. 5 is stopped, and in response to this, the routine from step StS of FIG. 4 is executed. That is, vector 11i['! Phase change amount Δ from A part 13
When ψ is input, the input of the interrupt signal rNTR from the 17n frequency divider 18 is prohibited in step 815, and N=N+Δψ is calculated in step 816, and two "
A call number for the data table is determined according to the amount of phase change Δψ. After that, as described above, the sin data corresponding to the calling signal N is generated.Description of the -order current waveform output section F,-order current waveform output section As described above, the first-order', TFI waveform output section 16 outputs the first-order angular frequency ω. The amount of phase change Δψ is determined by the sin data according to the resistance of the signal. Although the sLn data can be output, in this case, a delay occurs in the processing time of the CPU 9 when the CPU 9 performs arithmetic processing according to the routines shown in FIGS. 4 and 5. As a result, U-phase sine wave data sin (ωot+Δψ),
There is a risk that the v-phase sine wave data will be lost. Therefore, in the present invention, the primary IIL flow waveform output section 1flJ: for example, the seventh
By configuring as shown in the figure, it is possible to completely prevent variations in data extraction time and improve frequency accuracy. 7th
In the figure, the primary waveform output section 16 is a 1/n frequency divider 1
Triggered at 9 above the output pulse of 8, then 80μ
88CM/<Rus t- An AND condition that takes the AND condition of the monostable multivibrator 41 that is generated, the CPU 9, the decoder 42 that decodes the instructions of the CPU 9, and the Q terminal output of the monostable multipipe sorter 41 and the output of the decoder 42. A circuit 43 and an AND circuit 44 with an inhibit terminal into which one read/write (WR) word of the CPU 9 and the output of the decoder 42 are input; It consists of latch circuits 45 and 46 which double the sine wave data to be output, and latch circuits 47 and 48 which latch the output data of the latch circuits 45 and 46 when the output signal of the AND circuit 48 is at the rHJ level. has been done. Note that the output signal of the latch circuit 47, that is, the U-phase sine wave data 5 inches
(ω.t+Δψ) is the multiplication type digital-to-analog converter 15aK! The output signal is then converted into an analog signal and multiplied by the flow command value. As a result, the U-phase current indicator U*=resin(ω.t+Δψ) is obtained.

Further, the output signal of the latch circuit 48, that is, the V-phase sine wave data log converter 15t) converts the signal into an analog signal, and then multiplies the signal by -arrow current +I. +Δψ
) is required.

The output of the 1/n frequency divider 18 (a), the Q1 output of the monostable multivibrator 41 (b), the output of the AND@ circuit 44 (C), and the output of the latch circuit 45.46 (dl, t
dl'.

The output of the decoder 42 (e), the output of the AND circuit 43 (
f), the output fgl of the latch circuit 47.48, (gl
FIG. 8 shows the signal waveforms of the output +h+ of the multiplier digital-to-analog converter 15a, and the operation of the primary current waveform output section 16 will be described with reference to FIG. First, time t1
When a pulse word such as fat is inputted from the 1/n frequency divider 18, the monostable multi-pipe node 41 generates a pulse signal lNTR' such as 30 μ5 ecl 1 bl from the output of the pulse signal, External input terminal E of CPU9
Supply to XT INTR. Then, the CPU 9 calculates the call number N of the sine wave data according to the flowchart shown in FIG. 5, and extracts the sine wave data t corresponding to the call number N from the data table shown in FIG.

Here, the time from when the pulse signal 1NTR is input to the CPU 9 until the sine wave data is extracted is approximately 15 μSec.
It is. For this reason, time t! At time t, 16 μsec has elapsed since then, the output signal of the decoder 42 changes and the AND condition of the AND circuit 44 is satisfied, so that the AND circuit 44 outputs a pulse signal as shown in FIG. 8 +cl. This pulse signal is input to the gate terminal of the latch circuit 45.46, and the latch circuit 45.46
The output of changes. Next, at time t, 30 μSec has elapsed from time t1, the smell sensor and monostable multi-pipe node 41
Since the Q4 child output of is inverted and the AND condition of the AND circuit 43 is satisfied, the output signal ffl of the AND circuit 43 is rH
” becomes a novel.

When this HJV bell signal is input to the gate terminal of the latch circuit 147.48, the latch circuit 47.48 latches the output data of the launch circuit 45.46, and outputs the output signal of the latch circuit 47.48. is shown in Figure 8 [gl, tgl
'become that way. As mentioned above, the latch circuit is provided in two stages, and the data latch metal wheel stabilizes the multivibrator 41.
Since the configuration is such that the Qgs output is performed after a certain period of time, even if there are variations in the sine wave data extraction time in the CPU 9, the sine wave data is transferred from the primary current waveform output section 18 to the multiplication type digital-to-analog converter. 15 is kept constant. This significantly improves frequency accuracy. In addition, the pulse generation time of the monostable multivibrator 41 is not limited to 80 μsec, and the pulse generation time of the CPU 9
The sine wave data extraction time may be set arbitrarily as long as it is long. When the sine wave data extracted from the data table of FIG. 6 as described above is multiplied by the one-fail standard value knee in the multiplication type digital-to-analog converter 15a, the output signal is as shown in FIG. It changes like +h+. Here, when the Δψ outputted from the vector calculation section 13 changes, the signal novel of each section changes as shown by the broken line in FIG. That is, when Δψ changes at time t4, the CPU 9 executes step 8. in FIG. .

~S. Execute all the steps to find the calling number N=N+Δψ, and
Extract sine wave data corresponding to N. At this time, the CPU
A command is issued from 9 and the output signal le of the decoder 42
t becomes "LJV bell. Therefore, at 1% of the time when the output signal ffl of the AND circuit 48 becomes "L" and then becomes "rH", the latch circuit 47.48 has a signal corresponding to the change in Δψ. The sine wave data is latched. In the case of FIG. 8, the primary angular frequency ω. is in positive phase and Δψ=20°, so the sine wave data changes from υ1 to 6. at time t. This indicates that the output signal (hl) of the multiplier digital-to-analog converter L5a becomes a U-phase current command signal corresponding to a change of Δψ=20°.

The U-phase current command value IU* obtained as described above is matched in the first matching circuit 25 with the single word Lu detected by the U-phase output '1 current full current transformer 26 of the inverter 2. Further, the V-phase Kairyu finger spear 1 straight knee is matched with the signal -Lv detected by the V-phase output current metal flow device 28 of the inverter 2 in the second matching circuit 27 . The U-phase voltage reference value VUk is obtained by carrying out proportional integration in the first adjusting circuit 250 and the differential output voltage VL flow control amplifier 29.

Further, the V-phase voltage command value vV* is obtained by proportionally integrating the entire fan difference output of the second matching circuit 27 in the flow control amplifier 80. Furthermore, current control Anne 7'29, 30
The W-phase voltage index '+V[Vvr''t- is obtained by passing the signal obtained by calculating the power of the 0 output signal to the third matching circuit 81 through the inverting amplifier 82.

Next, these voltage command boxes “U * r ” I/” t ”
l'l' is compared with the triangular wave output of the triangular wave generator 86 in comparators 83 and 84.35, and the imperme 2i PWM control is performed based on the comparison output. By working on this, it is possible to control the vectors of the induction motor 11-12vc magnetic flux and the secondary current so that they are always perpendicular to each other.

Effects of the G0 Invention As described above, if the present invention is applied, other effects can be obtained. In other words, since the vector calculation is performed by converting the Yamaya coordinates, the number of calculations is small.
As a result, highly accurate single current commands can be created at high speed. This provides high-speed torque response.

(21) Highly accurate speed control is performed over a wide frequency range.

(Even if the phase angle ψ of the 31ai interval changes, the secondary current command value can be obtained according to the amount of phase change Δψ, so
It is possible to control the height of 4 degrees.

Since the 14+-11Kt flow waveform output section has a function of delaying the sine wave data extracted by cpσ by a predetermined period of time and outputting it, it is possible to completely correct the variation in CPH processing time. This eliminates the occurrence of frequency errors and significantly improves the frequency product.

15) Since the output signal of the pulse signal generator is frequency-divided according to the magnitude of the monthly cycle mantissa, sufficient calculation processing time can be secured even when the dielectric motor rotates at high speed. Therefore, CPU overhead can be eliminated.

(6) When the primary angular frequency is less than the set value (for example, during extremely low speed rotation), DXIω. 1 and outputs it, the voltage level of the analog IM part can be raised, the noise shadow W can be reduced, and a high-level analog processing circuit can be made unnecessary, which can reduce the cost of the device. .

[Brief explanation of drawings]

1 to 8 show an example of the present invention, FIG. 1 is a block diagram showing the overall configuration, and FIG. 2 is a polar prefecture conversion process t-
86 is a vector diagram for explanation purposes, FIG. 3 is a block diagram for explaining details of the vector calculation section, and FIGS. 4 and 5 are both flow charts showing the routine of the CPU.
7 is an explanatory diagram showing an example of a sine wave data table, FIG. 7 is a circuit diagram showing details of the Ikku's current waveform output section, and FIG. 8 is a time chart showing signal waveforms at various parts in FIG. 7. 1°°° induction motor, 2...PWM inverter, 9.
...CPU, 10...Speed calculation unit, 11...PI calculation unit, 12...Primary angular frequency calculation unit, 1g...Vector calculation unit, 14...Digital-analog converter,
15...X'X type digital-to-analog converter, 16
...-arrow current waveform output section, 17...v/F converter,
18...1/n frequency divider, 25, 27.81... Matching circuit, 29.30... It flow control amplifier, 33
, 34, 35... comparator, 36° triangular wave oscillator. Figure 4 Flowchart

Claims (2)

[Claims] (1) Torque current command value I_T^* and excitation current command value I_O obtained by proportional integration of the deviation output between the rotation angular frequency ω_n of the induction motor and the rotation angular frequency command value ω_n^*
a vector calculation unit that performs a predetermined vector calculation based on the rotational angular frequency ω_s and the slip angular frequency ω_s to obtain the primary current command value I_1^*, the slip angular frequency ω_s, and the phase change amount Δψ of the magnetic flux; Absolute value of the primary angular frequency ω_0 obtained by addition | ω_0
|, and the calculated absolute value |ω_0
When | is greater than the set value, the signal of |ω_0| is output,
a primary angular frequency calculation unit that outputs a signal obtained by multiplying |ω_0| by n when the calculated absolute value |ω_0| is less than a set value; and a pulse having a frequency proportional to the output signal of the primary angular frequency calculation unit. a pulse signal generator that generates a signal; and when the absolute value |ω_0| of the primary angular frequency ω_0| Absolute value of ω_0 | ω_
0| is less than a set value, a frequency dividing section that determines the frequency division ratio to be 1/D, and divides the output pulse signal of the pulse signal generation section according to the determined frequency division ratio; When the data table storing the data of the divided U and V phase sine wave signals and the output signal of the pulse signal generation section whose frequency has been divided by the frequency dividing section are input, the primary angular frequency U and V phase sine wave data corresponding to the polarity of ω_0 are extracted from the data table, and when the amount of phase change Δψ of the magnetic flux determined by the vector calculation unit changes, the extraction operation is stopped and the a primary current waveform output unit that extracts U- and V-phase sine wave data according to the change in Δψ from the data table, and outputs the extracted U- and V-phase sine wave data with a delay of a predetermined time; a multiplication unit that multiplies the output data of the primary current waveform output unit and the primary current command value I_1^* output from the vector calculation unit to obtain primary current command values I_U^*, I_V^* of the U and V phases; , 3-phase voltage command value V_U^*, based on the output current detection amount of the inverter for driving the induction motor and the output of the multiplier section,
An induction device characterized by comprising a phase voltage calculation unit that calculates V_V^* and V_W^*, and a PWM control unit that performs pulse width modulation control of an inverter for driving an induction motor according to the output of the phase voltage calculation unit. Vector control device for electric motors. (2) The vector calculation unit calculates the torque current command value I_T
Based on ^* and excitation current command value I_O^*, √(I
_T^*^2+I_O^*^2) is performed to obtain the primary current command value I_1^*, and (I_T^
*)/(I_O^*)×1/τ_2 (τ_2 is the second-order time constant of the induction motor) and calculates the slip angular frequency ω.
Find __s, and ψ=tan^-^1(I_T^*/I
_O^*) and Δψ=ψ_n−ψ_n_−_1(ψ_
2. The vector control device for an induction motor according to claim 1, wherein Δψ is determined by performing an operation in which n is a phase at the time of the current sample and φ_n_−_1 is a phase one sample before.