JPS60226786A - Digital speed detecting method - Google Patents

Abstract

PURPOSE:To eliminate a speed ripple or a torque ripple even at a low speed by adding a motor speed and a slip frequency by a rate multiplier. CONSTITUTION:The addition omegaM+omegaS of a motor speed omegaM and a slip frequency omegaS is not performed in a counter, but executed in a microcomputer 15 and a rate multiplier 9. Thus, since the omegaS becomes constant in the steady state even at a low speed, rough and dense states do not occur in the frequency of omegaM+ omegaS. Even if the slip is varied, omegaM+omegaS smoothly changes. Accordingly, a reference sinusoidal wave output does not transiently vary in the steady state.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a method for digitally detecting speed using a pulse oscillator attached to an electric motor.

[Prior art]

In accordance with No. 1N, the operation of an induction motor drive slip frequency control inverter device using speed detection using a conventional pulse oscillator will be described. Control is performed using a microcomputer.

That is, in FIG. 1, a DC power supply rectified by a diode converter (1) and smoothed by a capacitor (2) is converted into an AC current flowing to a modele (4) by a power transistor (3), which is generated through a DA converter (111). The PWM circuit 4 operates to match the phase reference current waveform (generally a sine wave).When the modele (4) operates, the pulse oscillator (5) generates a C refeedback pulse.The feedback pulse is usually the 8th A in the diagram
As shown in FIG.

As shown in the acm waveform, forward rotation pulse (CW), reverse rotation pulse (C
aW) and input to the counter (7)O't).

The pulse train input to the counter α is the moder speed ω
Becomes I. Further, the pulses input to the counter (7) are used to calculate the deviation from the command speed V ref by means of a microcomputer.

The method for calculating the deviation will be described below.

The CPU generates an interrupt at regular intervals due to the sampling pulse generated by the timer α6, which generates pulses at regular intervals, and at the same time reads the speed command V ref''4c from the input interface 00, the sampling frequency of the counter (7) is changed. The count pulse number V7Jl of the period is read into the microcomputer and stored in RAMcJ2.

To calculate slip, usually Δε. =V ref ts
If V pBn, speci111J=kp(Δg,
+ΣΔIIn)τ− is calculated and P-operation is performed.

Then, input the calculated Ns to the rate multiplier (9), and set the number of digits of the rate multiplier to 12 bits in binary.
, if the output of the oscillator (8) is fo, then ωS=J −η11·fo is input to the reversible counter α to generate f(ωS+ωn)dt1 necessary for slip frequency control. Also, the DA converter CL41 is normally
BooI! By microcomputer as = f
f〒o" (However, IO is constant) and gives the amplitude of the reference free current. Step 2 is DC! [In kth machine, it is the armature current, and IO is the field current. Also, Step 2 The problem is that the output of the phase angle 0=-1Ti- is 6ROM exposed and the digital value is 5ln(f(ω+ωs)dt
10) is output, and W! which is the output of the DA converter C14) in an analog manner by the DA converter α→ Width work and riding work W = work 5 ill (J” (ωH + ωB) at ten θ ten east π)
is given to control the three-phase current of the induction generator.

However, although this method operates without delay because the ~ period velocity from the mote ~ is directly input to the basic sine wave counter Q''f), it has drawbacks as shown in (6) of Section 4 □□□. That is, 0M and ωB are simultaneously reversible counters αη
It is synchronized by the internal clock so that it operates normally even if it is input to 0M in FIG. 4.

This is because the input is out of phase as shown by ωS.

The density occurs at the frequency of ωI(-ω8, and when observed with the pulse train I V -F converter of ωM+ωS, the 4th (
Since ripples occur as shown in b), the DA converter (9
) In extreme cases, the output becomes as shown in FIG. 7, which causes speed ripples and torque ripples in the moder, especially at low speeds. This effect is even greater in voltage type inverters than in current type inverters.

[Summary of the invention]

This invention has been made to eliminate these drawbacks.

[Embodiments of the invention]

An example is shown in @2□□□. In this diagram, the difference from Figure 1 is that the moder speed ωM is directly calculated by counter α
The difference is that the ω8+ωu'k rate multiplier (9) is used to generate a fault without inputting the ω8+ωu′k value. If the input cabalne during a certain sampling time is Nsp, then the 011 minute generated from the rate multiplier (9) in Fig. 2 is normally N, +r=Nsp・( ) (N, M, l integers).

The problem here is that there is a condition that NN must be an integer. Of course, if the integer is an integer, Nap will be an integer because it is the number of pulses of the counter, but in general, for the reason mentioned above, m > n and it is rarely an integer. Various conditions are required to make it an integer. For example, when calculating slip, it is necessary to match the range with the speed command [Vref, so a coefficient is required.

So I sent it to the microcomputer (Nsp Xl)/
Calculate k. If the darkness of this calculation is NM and the remainder is α, then (N8'p X l ), 4(= N, sr+α(N,
M: Integer α (basket integer). and) generally l)k.

Here, if α is omitted or only rounded off, the number of feedback pulses is not all input to the counter αη as shown in FIG. 1, which results in an ωIK error compared to the method shown in FIG. In the slip frequency control of the induction motor, there is a difference ωK>ωθ, so it is necessary to reduce the margin of ωy.

To prevent this, when sampling at tE1, when calculating the various l as NSp&il, NM-αli,
If the calculation at j +1 + 1 is performed every moment after the sampling period using (N8p(n+1) x 1+α)/N = NM (n-1-1) + α(n+1), the delay compared to Fig. 1 will be reduced. However, since ωI in FIG. 2 is completely equal, the above-mentioned drawback disappears. Further, if the five sampling periods are set to a predetermined value or more, the delay can be ignored in terms of control.

In the method of this invention shown in FIG. 2, the moder speed ωI
Then, the addition of the total frequency ωS to ω +ω8 is not done by a counter but by the microcomputer αO and the rate multiplier -■, so the output of +ω8 to ω is shown in the 5th part), and it is steady even at low speeds. In this state, ωS becomes constant, so ωM+ωS no longer becomes as shown in Figure 4).

When ωI + ωS is observed with an F-'V converter, Figure 5 (
6), the reference sine wave output does not fluctuate transiently in a steady state as shown in FIG. 6.

〔Effect of the invention〕

In the induction acid motor control using the digital speed detection method combined with microcomputer calculation described above, there is no 0M fluctuation due to frequency addition/subtraction using a counter, and the advantage of the method shown in Figure 1 (i.e., n-period The point that the detection of the velocity component ωI is extremely accurate) will not be impaired. Further, according to this method, the number of orders of magnitude of the counter α force and the number of pulses per revolution of the pulse oscillator (5) are increased, and there is an advantage that the design becomes easier.

[Brief explanation of the drawing]

Figure 1 is an example of a conventional microcomputer-based vector control PWM inverter using a pulse oscillator for speed detection. 2 is a diagram showing a simple embodiment of the present invention, FIG. 8 is a diagram showing a feedback pulse waveform from a pulse oscillator and a forward/reverse discrimination pulse waveform, and FIG. 4 is the same as that shown in FIG. 1. Figure 5 is a diagram showing an example of the waveform generated, and Figure 5 is a diagram showing an example of the waveform generated by the waveform shown in Figure 2.
The meat is a diagram showing an example of the sine wave output of the ODA converter in FIG. 2, and the seventh factor is a diagram showing an example of the sine wave output of the ODA converter in FIG. In the figure, (1) is an 8-phase diode bridge that rectifies 8-phase AC, (2) is a smoothing capacitor, (3) is an 8-phase power transistor bridge, (4) is an induction motor, and (5) is an induction motor. (6) is a circuit that generates forward pulses and reverse pulses from two-phase pulses with a 90° phase difference, (7) is a reversible counter, and (8) is the source of the rate multiplier. Transmitter, (9) is rate multiplier, αQ is microcomputer pass interface, (b) is OR gate, (2) is RAM% (to) is ROM%α or DA converter V (to) is microcomputer CPU, αQ is the timer, αη is the counter, (to) is the ROM, and Ql is the DA.
Converter, wire is PWM circuit, (2) is current detection CT. (A) is a phase angle latch circuit. Note that the same reference numerals in the figures indicate the same or equivalent parts. Agent Masuo Oiwa Figure 7 Figure 2 Figure 3 Figure 4 (a (b) Figure 5 Figure 6 Figure 1s7

Claims (1)

[Scope of Claims] A pulse oscillator coupled to a Mosyl axis and a device for counting feedback pulses from the pulse oscillator are provided, and the counting pulse (i-
Nn, and l and k are arbitrary integers. When the integer value of the answer to the operation is αn, the remainder V is βn, and the remainder V from the calculation at the previous sampling time tn-1 is βul, then CNIIxl+βn-1)/z-αi+βn(k'>β
A digital speed detection method characterized by executing the calculation n) and setting the answer αn to a systemized mho)/7 speed.