JP3312237B2 - Power converter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a power converter for driving an AC motor by converting a direct current into an alternating current.
The present invention relates to a control technique for flattening a loss peak of a power converter.

[0002]

2. Description of the Related Art Japanese Patent Laid-Open Publication No.
No. 7085, JP-A-2-31119
The technology described in Japanese Patent Publication No. 7 is known. The technique disclosed in the former publication is a power converter for converting a direct current into an alternating voltage having a two-level potential, comprising multi-pulse generating means having bipolar modulation and overmodulation as pulse width modulation (PWM) means, and performing asynchronous modulation. It is characterized by the following.
Further, the technique disclosed in the latter publication increases the carrier (carrier) frequency when the magnitude of the modulation wave that determines the magnitude and frequency of the inverter output voltage is around 0.5, and then reduces the magnitude again, thereby reducing the motor current peak. It is intended to suppress a ripple component and reduce inverter capacity and heat loss.

[0003]

SUMMARY OF THE INVENTION The above-mentioned Japanese Patent Application Laid-Open No. 7-227 is disclosed.
No. 085, the output voltage is π / 4 * 100.
% (Approximately 78%), a semiconductor device constituting the power converter has a triangular peak loss. In particular, the size of the cooler of the device is determined by this peak loss, and there is no consideration for heat loss. There was a problem that the device became large. On the other hand, the technology described in Japanese Patent Laid-Open No.
Since the carrier frequency is gradually increased until the amplitude of the modulated wave is close to 0.5, and the carrier frequency is controlled to be reduced immediately before the one-pulse mode in which the output voltage waveform becomes a square wave, a triangular peak loss occurs. Also, there is a fear that the motor current ripple and noise increase due to a decrease in the number of pulses of the output voltage.

[0004] It is an object of the present invention to provide a power converter suitable for smoothly shifting from the asynchronous PWM mode to the one-pulse mode, and for suppressing and flattening the peak loss.

[0005]

The object of the present invention is to reduce the output voltage of the power converter to 100 when the output voltage is in the one-pulse mode.
%, Asynchronous modulation is performed in a region other than the one-pulse mode, and when the output voltage is around π / 4 * 100%,
The problem is solved by providing means for controlling the carrier frequency to be minimized. Also, if the power converter is 2
Carrier frequency setting means comprising a level inverter, comprising a bipolar modulation mode and an overmodulation mode, performing asynchronous modulation in the bipolar modulation mode and the overmodulation mode, and reducing and controlling a carrier frequency near a boundary between the bipolar modulation mode and the overmodulation mode. Is solved. Further, the power converter is a three-level inverter, and includes a unipolar modulation mode and an overmodulation mode in an asynchronous modulation region, and includes a carrier frequency setting unit that controls to reduce a carrier frequency near a boundary between the unipolar modulation mode and the overmodulation mode. This is solved.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a power converter according to an embodiment of the present invention. FIG. 2 shows a two-level IGBT (Insulate) for driving an electric vehicle as an example of a main circuit configuration.
d Gate Bipolar Transisto
r) A configuration example of an inverter is shown. First, the configuration and basic operation of the two-level inverter will be described. In FIG. 2, 10 is an overhead wire which is a DC voltage source, 20 is a filter capacitor installed on the DC side, and 301 and 302 are self-extinguishing switching elements having a rectifying element for freewheel (in this example, IGBTs are used). However, a GTO thyristor, a transistor, or the like may be used.), 30a is a switching arm for one phase of U phase, 30b and 30c have the same configuration as 30a, and are switching arms for V phase and W phase, respectively, and 40 is an induction arm. It is an electric motor. The switching arms 30a to 30c can operate independently for each phase, and generate two-level output voltages at the inverter AC side terminals (U, V, W) by ON / OFF control of the IGBT.
That is, an output voltage having two levels of potentials of + Ed / 2 and -Ed / 2 is obtained with respect to the ground point O virtually provided in the filter capacitor 20. This output voltage can be adjusted by pulse width modulation (PWM) control to obtain a predetermined voltage by adjusting each pulse width of a voltage pulse train forming an output voltage waveform. Hereinafter, this mode is referred to as a multi-pulse mode. In addition, as a control unique to electric vehicles, a one-pulse mode in which the output voltage waveform is a square wave is provided, making full use of the power supply voltage up to the maximum voltage utilization rate, thereby miniaturizing the device and increasing the efficiency. .

Hereinafter, this embodiment will be described with reference to FIG. In FIG. 1, reference numeral 1 denotes three-phase modulated waves Ymu and Ym based on a frequency command Fi * and a voltage command E * of an inverter output voltage.
Modulated wave calculating means for generating v and Ymw, 2 is a voltage command E *
Is a carrier frequency setting means for outputting a preset carrier frequency command Fc * from the CPU. Here, the voltage command E *
, The carrier frequency command Fc * is output, but the frequency command Fi * may be used instead of the voltage command E *.
Reference numeral 3 denotes a carrier computing means for generating a triangular carrier Yc in accordance with a carrier frequency command Fc *, and 4 compares the three-phase modulated waves Ymu, Ymv, and Ymw with the carrier Yc to determine a magnitude relationship. Inverter main circuit IG
PWM operation means for generating a multi-pulse signal which is an on / off signal to be given to the BT, and an on / off signal for generating a square wave (one pulse) according to the inverter frequency command Fi *.
One-pulse generating means 6 for generating an off-signal (one-pulse signal) selects the output of the PWM calculating means 4 and the one-pulse generating means 5 according to the voltage command E * and outputs it as a final on / off signal. PWM mode selection means. The IGBT of the inverter main circuit shown in FIG. 2 is turned on / off by the on / off signal via a gate driver (not shown).

FIG. 3 shows an example of an inverter output voltage waveform in an electric vehicle driving inverter. Fig. 3 for low-speed range
The bipolar modulation mode of FIG. 3A, the medium speed range is covered by the overmodulation mode of FIG. 3B, and the high speed range is covered by the one-pulse mode of FIG. In this example, the multi-pulse mode of the bipolar modulation mode and the over-modulation mode is changed to a triangular-wave carrier Yc.
And a sine wave modulated wave Ym, the pulse width is determined so that the inverter phase voltage becomes + Ed / 2 when Ym> Yc and −Ed / 2 when Ym <Yc. are doing. In the multi-pulse mode, the magnitude and frequency of the inverter output voltage can be controlled by controlling the amplitude and frequency of the modulated wave Ym. In particular, the bipolar modulation has a feature that the output voltage can be linearly controlled with respect to the amplitude of the modulation wave Ym, that is, the modulation rate. At this time, the output voltage fundamental wave component E1 is theoretically given by the following equation. E1 = (A / 2) Edsin (2πFit) (1) where A: amplitude of modulated wave (0 ≦ A ≦ 1) Ed: DC voltage Fi: inverter frequency (frequency of modulated wave) On the other hand, FIG. In the one-pulse mode shown in (1), the output voltage waveform becomes a square wave synchronized with the inverter frequency. At this time, the output voltage fundamental wave component E1 is determined only by the magnitude of the DC voltage, and is given by the following equation: E1 = (2/2 / π) Edsin (2π Fit) (2) Therefore, the theoretical controllable range of the output voltage in the bipolar modulation mode is π / 4 * 100% (about 78%) when one pulse is 100%. In this example, the waveform at the time of synchronous modulation in which the cycle of the modulated wave Ym and the carrier Yc is an integral multiple is used.
Even if asynchronous modulation does not cause the period of
The M control method and output voltage are exactly the same. Modulated wave Ym
When the amplitude becomes larger than the amplitude of the carrier Yc, the mode shifts to the overmodulation mode shown in FIG.
In the overmodulation mode, since there is no intersection with the carrier Yc near the peak of the modulated wave Ym, there is a feature that the pulse period near this peak is extended. Thereby, π / 4 * 100%
The voltage can be controlled from (about 78%) to almost 100%, and the voltage gap between the bipolar modulation mode and the one-pulse mode can be covered. In this region, the modulation rate and the magnitude of the output voltage are not linear, but linearization is possible by reading the modulation rate in response to the output voltage command. In this example as well, the waveform at the time of synchronous modulation is used for convenience, but similar control is possible even when asynchronous. However, in the case of asynchronous modulation, if the number of pulses appearing in one cycle is significantly reduced, the imbalance of the output voltage may occur, and the motor current may oscillate. Not in.). Therefore, when overmodulation is used for asynchronous modulation, it is necessary to set the carrier frequency so as to secure a certain number of pulses or more.

Next, the conventional operation and the operation of the present embodiment will be described with reference to FIGS. Here, the voltage command E * has a relationship substantially proportional to the inverter frequency. 4 to 8, the horizontal axis represents the inverter frequency. FIG. 4 shows the carrier frequency and the corresponding switching (SW) frequency of the main circuit element (which can be approximated by the frequency obtained by multiplying the number of pulses included in one cycle of the output voltage by the inverter frequency) and the element in the conventional method. It shows the characteristics of loss. In this example, the case where asynchronous modulation is performed in the multi-pulse mode at a constant carrier frequency Fco has been described. The element loss has a characteristic having a peak loss Ppk at the inverter frequency Fi1 which is a boundary between the bipolar modulation and the overmodulation. The output voltage at this time substantially coincides with π / 4 * 100% (about 78%) when the voltage in the one-pulse mode is 100%. I
Since the temperature rise of the GBT chip junction largely depends on the peak loss, by reducing this peak loss,
The cooling system can be made smaller and lighter.

In this embodiment, the carrier frequency is set by the carrier frequency setting means 2 shown in FIG. 1 so as to be reduced from Fco to Fc1 in the vicinity of the inverter frequency Fi1, that is, from Fi3 to Fi1, as shown in FIG. Is controlled so that the carrier frequency in the vicinity of the output voltage π / 4 * 100% (about 78%) is minimized. As a result, the switching (SW) frequency in the vicinity of the inverter frequency Fi1 can be reduced from Fswo to Fsw1, and the element loss can be improved from a triangular shape having peak characteristics to a flat characteristic with low peak loss. Further, by gradually returning the carrier frequency to Fco, a decrease in the number of pulses of the output voltage during overmodulation is suppressed. For this reason, the motor noise generated by the decrease in the number of pulses of the output voltage can be reduced. Naturally, if the carrier frequency is sufficiently high and a sufficient number of pulses can be secured in overmodulation even at the carrier frequency Fc1 during the reduction control, it is not necessary to increase the carrier frequency.

Further, as shown in FIG. 6, the carrier frequency may be set back by the inverter frequency Fi4 by the carrier frequency setting means 2 of FIG. 1 to speed up the operation of returning the carrier frequency. In this case, when the inverter frequency is in the range of Fi4 to Fi2, characteristics equivalent to those of the conventional overmodulation control can be obtained. On the other hand, when the initial value Fco of the carrier frequency is set relatively low by the carrier frequency setting means 2 in FIG. 1, the carrier frequency in the overmodulation mode is raised to a higher frequency Fcp as shown in FIG. Set to. This allows switching (S
W) It is possible to improve a flat element loss characteristic with a low peak loss without increasing the frequency and almost without increasing the element loss.

Further, when the thermal time constant of the cooling system is taken into consideration in the embodiment of FIG. 5, the peak temperature occurs at the inverter frequency Fi1, but as shown in FIG.
If the carrier frequency at the time of the reduction control is set slightly lower from Fc1 to Fc1 'by the carrier frequency setting means 2 and a slight gradient is provided in the element loss, it is possible to adjust the peak occurrence point of the temperature rise. Becomes

As described above, in this embodiment, in addition to the effect of the smooth PWM control by the conventional asynchronous modulation, the control of the carrier frequency near the boundary between the bipolar modulation and the overmodulation allows the flat element loss having a low peak loss to be achieved. Characteristics can be obtained, and a compact, lightweight, and low-noise device can be realized.

Although the two-level inverter has been described as an embodiment of the present invention, a unipolar modulation mode for outputting a pulse of a single polarity pulse-width-modulated at a constant period during a half period of an output voltage fundamental wave; A similar effect can be obtained for a three-level inverter having an overmodulation mode in which the output cycle of the pulse near the peak of the output voltage fundamental wave is longer than the output cycle of the pulse near the fundamental wave zero cross. Also,
The same control can be realized by software using a microprocessor or the like.

[0015]

As described above, according to the present invention,
By providing a carrier frequency setting means for managing the carrier frequency, asynchronous modulation is performed at a carrier frequency determined by this output in a region other than one pulse, and when the output voltage is near π / 4 * 100%, the carrier frequency is minimized. And the power converter is a two-level inverter or a three-level inverter, each near the boundary between the bipolar modulation mode and the overmodulation mode or near the boundary between the unipolar modulation mode and the overmodulation mode in the asynchronous modulation region. Thus, the carrier frequency can be reduced and controlled, so that the peak loss of the semiconductor element constituting the power converter can be suppressed and flattened, and the device can be reduced in size, weight, and noise.

[Brief description of the drawings]

FIG. 1 is a power converter according to an embodiment of the present invention.

FIG. 2 is a typical main circuit configuration diagram.

FIG. 3 is a diagram illustrating an inverter output voltage waveform.

FIG. 4 is a diagram for explaining a conventional loss characteristic.

FIG. 5 is a diagram illustrating a method of setting a carrier frequency according to the present invention.

FIG. 6 is a diagram illustrating a method of setting a carrier frequency according to the present invention.

FIG. 7 is a diagram illustrating a method of setting a carrier frequency according to the present invention.

FIG. 8 is a diagram illustrating a method of setting a carrier frequency according to the present invention.

[Explanation of symbols]

1: Modulated wave calculating means 2: Carrier frequency setting means 3: Carrier calculating means 4: PWM calculating means 5: 1 pulse generating means 6: PWM mode selecting means 10: Overhead wire 20: Filter capacitors 30a to 30c: Switching arms 301 to 302 ： Switching element 40 ： Induction motor

Continuation of front page (56) References JP-A-7-227085 (JP, A) JP-A-3-3695 (JP, A) JP-A-2-311197 (JP, A) (58) Fields studied (Int .Cl. ⁷ , DB name) H02P 5/408-5/412 H02P 7/628-7/632 H02P 21/00 H02M 7/42-7/98 WPI (DIALOG)

Claims (6)

(57) [Claims] 1. A power converter comprising a power converter for converting a DC voltage into an AC voltage by pulse width modulation, and a motor driven by the power converter, wherein the output voltage waveform of the power converter is a square wave. Output voltage when
When it is set to 00%, asynchronous modulation is performed in a region other than the region where the output voltage waveform is a square wave, and the output voltage is π / 4 * 100.
%, Comprising means for controlling the carrier frequency to be a minimum when the power conversion value is around%. 2. A power converter comprising a power converter for converting a DC voltage to an AC voltage by pulse width modulation, and a motor driven by the power converter, wherein the power converter is a two-level inverter. One pulse mode that outputs a square wave synchronized with the inverter frequency, a bipolar modulation mode that outputs a pulse width-modulated pulse at a constant cycle, and the output cycle of the pulse near the peak of the output voltage fundamental wave Carrier frequency setting means comprising an overmodulation mode longer than an output cycle of a nearby pulse, performing asynchronous modulation in the bipolar modulation mode and the overmodulation mode, and reducing and controlling a carrier frequency near a boundary between the bipolar modulation mode and the overmodulation mode. A power conversion device comprising: 3. The carrier modulation apparatus according to claim 2, wherein the carrier frequency setting means sets the carrier in the original overmodulation mode before shifting the carrier frequency reduced near the boundary between the bipolar modulation mode and the overmodulation mode to the one-pulse mode. A power conversion device characterized by returning to a frequency. 4. The carrier frequency setting means according to claim 2, wherein when the initial value of the carrier frequency is set relatively low, the carrier frequency in the overmodulation mode is set to be raised to a higher carrier frequency. A power converter characterized by the above-mentioned. 5. The electric power according to claim 2, wherein the carrier frequency setting means sets a carrier frequency slightly lower than a carrier frequency reduced near a boundary between the bipolar modulation mode and the overmodulation mode during the reduction control. Conversion device. 6. A power converter including a power converter for converting a DC voltage to an AC voltage by pulse width modulation and a motor driven by the power converter, wherein the power converter is a three-level inverter. In the asynchronous modulation region, a unipolar modulation mode that outputs a pulse of a single polarity that is pulse-width-modulated at a fixed cycle during the half cycle of the output voltage fundamental wave, and a pulse output cycle near the peak of the output voltage fundamental wave A power conversion device comprising: an overmodulation mode longer than an output cycle of a pulse near a wave zero crossing; and carrier frequency setting means for controlling carrier frequency reduction near a boundary between the unipolar modulation mode and the overmodulation mode.