JPH0787697B2 - Control method of PWM inverter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to a control method for a PWM inverter.

(Prior Art) FIG. 4 is a main circuit diagram of a PWM inverter. It is composed of a DC power supply 1, a capacitor 2, and an inverter 3 using a self-extinguishing semiconductor element. In the figure, an induction motor 4 is connected as a load. The DC voltage is constant in the PWM inverter, and the output voltage is controlled by the PWM control of the inverter 3. Recently, sine-wave PWM control is used to approximate the output line voltage to a sine wave.

The sine wave PWM control method includes "a method of performing 360 ° PWM and a method of performing PWM by turning on / off between 240 ° and not switching the remaining 120 °, and switching only two phases at the same time" (Kamiyama Naohiko "New Drive Electronics" Sho 57.7.25 Denki Shoin, p75). Here, the former is three layers
PWM and the latter will be called two-phase PWM. The PWM control method is classified into a synchronous type and an asynchronous type based on the relationship between the operating frequency and the modulation frequency (ibid. P70).

Comparing 3-phase PWM and 2-phase PWM with the synchronous method, 2-phase PWM
Is better. One is that the unstable phenomenon (vibration phenomenon) during the operation of the induction motor is less likely to appear, and the other is that the loss is small. This is shown in FIGS. 5, 6, and 7.
It will be described with reference to the drawings.

FIG. 5 is an explanatory diagram of the 2-phase PWM, and FIG. 6 is an explanatory diagram of the 3-phase PWM. In both figures, (a) shows the state of comparison between the voltage reference pattern and the carrier wave, and (b) shows U obtained thereby.
-Phase PWM signal, (c) is the U-phase positive arm switch element U
Drive signal given to P, (d) Drive signal given to switch element UN of negative arm of U phase, (e)
Is the U-phase output voltage, (f) is the V-phase output voltage, and (g) is the UV line voltage. In the voltage type inverter, the switching element of the positive side arm and the switching element of the negative side arm of the same phase are alternately turned on / off, but if the positive side and negative side switching elements of the same phase are turned on at the same time, the DC power supply is short-circuited. As a result, an overcurrent flows through the switch element and destroys the switch element. Therefore, the on timing is delayed with respect to the off timing so as not to turn on at the same time. Therefore, the U-phase drive signal is as shown by the solid lines in (c) and (d), and there is a period in which both the positive-side element and the negative-side element are off. Therefore, the U-phase output voltage (e) has a period in which the output voltage is indefinite. In the figure, a solid line indicates a period in which either the positive side or the negative side is on and the voltage is fixed, and a dotted line indicates a period in which both elements are off and the voltage is indefinite. The voltage during this irregular period is determined by the current direction at that time. For example, if a current flows into the induction motor 4 in FIG. 4, the current flows from the negative side of the DC power supply through the diode connected in antiparallel to the switch element UN. Therefore, the voltage becomes negative. On the contrary, if a current is flowing from the induction motor 4, the current flows into the positive side of the DC power source via the diode connected in anti-parallel to the switching element UP. Therefore, the voltage becomes positive.
Even though the voltage is controlled in this way, the voltage fluctuates under the influence of the current. The same applies to the V phase (f) and the W phase (omitted). (G) shows the UV line voltage, but the period during which the voltage is still fixed is shown by a solid line and the indefinite period is shown by a dotted line. In the part indicated by the arrow in (g), the U-phase reference and the V-phase reference are about the same size, so both are switched almost simultaneously and the U-phase voltage and V-phase voltage are both indefinite. Has become. This period will be described with reference to FIG. In FIG. 7, (I) is U
It is an enlarged view of the output voltage of the V phase and V phase. The U-phase voltage is negative until the timing t1, and is positive after the timing (t1 + Δt). The period from t1 to (t1 + Δt) is an indefinite period. V
Similarly, the phase voltage is negative until the timing t2 and the timing (t2 +
Δt) and thereafter is positive, and the period from t2 to (t2 + Δt) is an indefinite period. At this time, it is determined whether the voltage of the U-phase or V-phase during an irregular period becomes positive or negative in the direction of each phase current.
There are four combinations as shown in (II), (III), (IV) and (V). In (II), the voltage changes in the U and V phases without delay with respect to the control signal, and a positive voltage pulse having a width of (t2-t1) appears in the line voltage. Both (V) are the cases where the voltage changes after an irregular delay, and the width of the pulse appearing in the line voltage is (t2-t1). In this case, the phase of the pulse is only delayed by Δt with respect to the control signal. In (III), the U phase changes without delay and V
Only the phase changes with a delay of Δt. The line voltage is (t2-t1
A pulse with a width of + Δt) appears, and the width increases by Δt.
In (IV), the U phase changes by Δt and the V phase changes without delay. A negative voltage pulse having a width of (t1 + Δt−t2) appears in the line voltage. In this way, the position of the pulse changes depending on the current direction at that time, the pulse width changes, and the positive and negative of the pulse are reversed. The most unfavorable case is the case where the negative voltage pulse is output when the positive voltage pulse should be output (IV). Now, Fig. 5 and 6
Returning to the figure, we investigate which of two-phase PWM and three-phase PWM is more susceptible to this effect. In the case of FIG. 5 (two-phase PWM), each of the U-phase voltage (e) and the V-phase voltage (f) performs PWM of 240.
It does not PWM during the remaining 120 degrees. Moreover, PWM
Not 120 degrees are offset from each other. As a result, both U-phase and V-phase PWM at the same time only for 120 degrees,
Moreover, the line voltage is only near zero. For the rest of the period, only one of the phases performs PWM to generate the line voltage. Therefore, the period during which the voltage pulse opposite to that of the control signal is generated is small, and the period during which the voltage pulse is generated is near zero of the line voltage, so there is little effect on the line voltage. On the other hand, in the case of FIG. 6 (three-phase PWM), U-phase voltage (e), V
Both phase voltages (f) are PWMed for 360 degrees. Therefore, there is a possibility that a voltage pulse opposite to the control signal is generated in all switching. In FIG. 6, since the voltage is high, the values of the U-phase and V-phase voltage reference patterns are substantially the same, and switching is performed almost at the same time only near the line voltage of zero, but when the voltage is low, the voltage reference pattern is reduced. The range where the values of are about the same increases. When outputting a minute voltage, any switching between 360 degrees can output a line voltage pulse that is the reverse of the control signal. It can be said that the output voltage of the three-phase PWM fluctuates more easily depending on the motor current than the two-phase PWM.
Therefore, if the motor current fluctuates for some reason,
As a result, the output voltage fluctuates, and the fluctuation in the output voltage further fluctuates the electric motor current, which is so-called unstable development.

Next, the two are compared with respect to switching loss. The induction motor current has a phase delay of approximately 90 degrees with respect to the voltage when there is no load. However, the phase delay becomes smaller as the load increases, and becomes smaller at the rated load. Fifth
Looking at the phase voltage (e) in Fig. 6 and Fig. 6, the three-phase PWM always switches, whereas the two-phase PWM does not switch near the electrical angle where the basic number of phase voltages is maximum.
Switching is performed only while the phase voltage is small. 2-phase PWM
Means that switching does not occur near the peak of the phase current with a large loss per switching, but only with switching near the zero of the phase current with a small loss. Therefore, if the number of times of switching is the same, the loss is smaller in the two-phase PWM.

In this way, the two-phase PWM is superior in terms of stability when operating the induction motor and switching loss. Both of these have PW near the electrical angle where the fundamental wave of the phase voltage is maximum.
This is an advantage caused by not performing M.

(Problems to be Solved by the Invention) Although the two-phase PWM is excellent as described above, the circuit is complicated in the synchronous system. The circuit is simple and inexpensive, and another problem occurs when trying to do it with asynchronous two-phase PWM.

Figure 8 shows the asynchronous 2-phase PWM pattern generation method and the resulting output voltage waveform. In FIGS. 8 I, II, and III, (a) shows the comparison of the reference wave and the carrier wave,
(B) is a U-phase output voltage, (c) is a V-phase output voltage, (d)
Is the UV line voltage waveform. As in the case of (a), the carrier wave has a continuous waveform, unlike the case of the synchronous type. Further, since the reference wave and the carrier wave cannot be fixed in a specific phase relationship as in the case of the synchronous type, the phase relationship between the reference wave and the carrier wave is I.
→ It gradually shifts like II → III. UV line voltage (d)
Looking at, it can be seen that the pulse width is irregular at the time of switching to the phase that does not perform PWM every 60 degrees. Moreover,
The width of the irregular pulse changes in I, II, and III. The irregular pulse indicated by the arrow in the figure occurs at the electrical angle at which the magnitude of the fundamental wave component of the line voltage is maximum, so the change in pulse width has no effect on the magnitude of the fundamental wave component of the line voltage. large. Therefore, this pulse not only causes the current to protrude during the induction motor operation, but also causes the unstable development during the induction motor operation described above, making the two-phase PWM that was stable in the synchronous system unstable. turn into.

The present invention uses an asynchronous method that can make the circuit simple and inexpensive.
The purpose is to obtain a PWM control method that has less loss than phase PWM and does not generate irregular pulses as in the case of two-phase PWM.

[Structure of Invention]

(Means for solving the problem) In the synchronous system, the two-phase PWM has less loss and is more stable than the three-phase PWM because the PW is close to the electrical angle where the fundamental wave of the phase voltage is maximum.
This is because there is a period in which M is not performed. On the other hand, the reason why the irregular pulse is generated in the asynchronous two-phase PWM is that the phase in which PWM is not performed is instantaneously switched. Therefore, in the vicinity of the electrical angle where the fundamental wave of the phase voltage is maximum, there is a period in which PWM is not performed as in 2-phase PWM, but that period is shorter than 2-phase PWM, and all three phases are PWM during the remaining period. , Gradually switch the phases without PWM.

(Function) If PWM is performed as described above, irregular pulses are not generated, and switching does not occur near the electrical angle at which the fundamental wave of the phase voltage is maximum, so stable and low-loss PWM control is achieved.

(Exemplary Embodiment) Configuration of Embodiment FIG. 1 shows a configuration diagram of an embodiment of the present invention. In FIG. 1, 1-4 are the same as in FIG. 10 is a frequency commander that outputs a voltage proportional to the operating frequency of the inverter, 11
Is a V / F converter that outputs a pulse having a frequency proportional to the output voltage of the frequency commander 10, 12 is a counter that integrates and outputs the pulses that the V / F converter 11 outputs, and 13 is an output of the frequency commander 10. The A / D converters 14a, 14b, 14c for converting the voltage into a digital amount are ROMs which are addressed by the integrated value output by the counter 12 and the digital amount output by the A / D converter 13, and the voltage reference pattern is It has been written.

Reference numerals 15a, 15b, 15c are D / A converters for converting the digital quantities output from the ROMs 14a, 14b, 14c into analog quantities. The outputs of the D / A converters 15a, 15b, 15c are comparators 16a, 16b, 16c, respectively.
Entered in. The comparator 16a, 16b, 16c includes a triangular wave generator 1
The triangular wave output by 7 is also input, and the D / A converter 15
The magnitudes of the outputs of a, 15b, and 15c are compared and the result is output as a logic signal. 18a, 18b, 18c are negation theoretical circuits, and take the negative logic of the outputs of the comparators 16a, 16b, 16c. Comparator 16a, 1
The outputs of 6b and 16c drive the switching elements of the positive arm of the three phases of the inverter 3, and the negative logic circuits 18a and 18b,
The switching element of the negative arm is driven by the output of 18c.

The voltage reference pattern written in the ROMs 14a, 14b, 14c in the above configuration will be described with reference to FIG. The voltage reference pattern is a function of the electrical angle and the voltage, but the functional expression differs depending on the electrical angle. To make this easier to explain, one cycle of the function is divided into 24 equal parts and numbers are given to each period. This is shown in (a). The electrical angle θ = 0 at the beginning of the period and the function FU (θ) for the U phase in the period (O <θ <π <12) is FU (θ) = (12 / π) · (V · sin (π / 4) −1/2) ・ θ
(1) The function is In the period and (π / 12 <θ <π / 4), FU (θ) = V · sin (θ + π / 6) −1/2 (2) In addition, the period, (π / 4 <θ <5π / 12) Then FU (θ) =-(12 / π) ・ (V ・ sin (π / 4) −1/2) ・
θ + 4 ・ (V ・ sin (π / 4) -1/2) + V ・ sin (θ-π /
6) (3) In the period (5π / 12 <θ <π / 2), FU (θ) = 1/2 (4). With this, since a waveform having a quarter period is defined, the remaining three quarter periods are determined from the symmetry of the waveform. For the period ~ (π / 2 <θ <π), the function is such that the waveform is symmetrical with the period ~ with respect to the electrical angle π / 2. The period-is a function whose polarity is opposite to that of the waveform of the period-. The waveform is shown in (b). In the period before and after the electrical angle at which the fundamental wave of the phase voltage takes the maximum value, and, the function value is a constant value 1/2 or −1/2 regardless of the value of the voltage command V. The reference function FV (θ) for the V phase and the reference function FW (θ) for the W phase are 2π /
The function has the same waveform with different phases by 3, 4π / 3. This is shown in (c). When the voltage-based functions FU (θ), FV (θ), and FW (θ) are thus determined, the function value values {FU (θ) −FV (θ)} and {FV (θ) −FW (θ) )},
{FW (θ) -FU (θ)} is the voltage command value V as shown in (d).
It becomes a sine wave function waveform whose amplitude is.

Function of the embodiment Functions FU (θ), FV output from ROM 14a, 14b, 14c
(Θ) and FW (θ) are D / A converters 15a, 15b, 15c, respectively.
Is converted into an analog amount at and is compared with the output of the triangular wave generator 17 at the comparators 16a, 16b and 16c. The state of this comparison is shown in FIG. The triangular wave repeatedly increases and decreases between +1/2 and -1/2. As a result, by comparing the U-phase reference waveform with the triangular wave, the output of the comparator 16a becomes the waveform shown in (b). U
The phase reference waveform has the same value as the maximum value +1/2 and the minimum value -1/2 of the triangular wave only for the period of 30 electrical degrees, regardless of the voltage reference. Therefore, there can be a period without PWM as shown in (b). Similarly, by comparing the V-phase reference waveform and the triangular wave, the output of the comparator 16b becomes the waveform shown in (c). Although not shown, the output of the comparator 16c has a similar waveform. The outputs of the comparators 16a, 16b, 16c are directly used as the drive signal of the switching element of the positive side arm of the inverter 3, and the signals inverted by the negative logic circuits 18a, 18b, 18c drive the negative side arm of the inverter 3. Signaled. Therefore, the U-phase output voltage has the same waveform as the output of the comparator 16a, and the V-phase output voltage has the same waveform as the output of the comparator 16b. The line voltage between the U phase and the V phase has a waveform shown in (d). As shown in FIG. 2, since the reference waveform of each phase is a function that becomes a sine wave when the difference from the other phases is taken, the line voltage is controlled in a sine wave. At this time 2
In the vicinity of the electrical angle where the irregular pulse was generated in the phase PWM, the voltage reference continuously changed, and all of the U phase, V phase, and W phase were PW.
The irregular pulse does not appear because it is M.

〔The invention's effect〕

As described above, the present invention is a system that performs PWM for 300 degrees in one cycle and does not perform PWM for the remaining 60 degrees.
This is an intermediate PWM control method between WM and 3-phase PWM. By this method, the output line voltage can be controlled by a sine wave like the conventional two-phase PWM and three-phase PWM. This method uses PWM for 30 degrees before and after the electrical angle where the magnitude of the fundamental wave component of each phase voltage becomes maximum.
Since it does not, like the two-phase PWM, it does not switch during the period when the current value is large, and switches only during the period when the current value is small, so that an efficient PWM inverter can be obtained. Also, since the voltage reference waveform does not become discontinuous unlike the conventional two-phase PWM, an irregular pulse does not appear even in an asynchronous PWM control circuit, and stable PWM control can be performed with a simple circuit.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an embodiment of the present invention, FIG. 2 is an explanatory diagram of a voltage reference waveform for a voltage control method of the present invention, and FIG.
Fig. 4 is a diagram for explaining the operation of the voltage control method of the present invention. Fig. 4 is a PWM.
Inverter main circuit configuration diagram, Fig. 5 is an explanatory diagram of 2-phase PWM control, Fig. 6 is an explanatory diagram of 3-phase PWM, Fig. 7 is an explanatory diagram of irregular voltage in PWM control, and Fig. 8 is an asynchronous type 2 phase PW
FIG. 7 is an explanatory diagram of irregular pulses generated when M control is performed. 1: DC power supply, 2: Capacitor 3: Inverter, 4: Induction motor 10: Frequency commander, 11: V / F converter 12: Counter, 13: A / D converter 14a, 14b, 14c: ROM 15a, 15b, 15c : D / A converter 16a, 16b, 16c: Comparator 17: Triangle wave generator 18a, 18b, 18c: Negative logic circuit

Claims (1)

[Claims] 1. A PWM inverter control method for PWM controlling a three-phase inverter, wherein the three-phase voltage reference signal is an electrical angle.
It changes while maintaining continuity from the positive side constant voltage period of less than 60 degrees to the negative side constant voltage period of less than 60 degrees in electrical angle,
In addition, each of them has a phase difference of 120 degrees in electrical angle, and each interphase signal is given as a function that changes sinusoidally, and the three-phase inverter is operated by using the three-phase voltage reference signal and the triangular wave signal of a predetermined frequency. A PWM inverter control method characterized by PWM control.