JP3411995B2 - Power converter - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a power converter for converting direct current to alternating current or alternating current to direct current, and more particularly, to a PWM.
(Pulse width modulation) The present invention relates to a technique for controlling an inverter. 2. Description of the Related Art "Electric Vehicle Science" published by Electric Vehicle Research Society, 199
An example of the modulation method of the inverter is described in FIG. 1, page 14 of the April 2013 issue, "Evaluating Recent Inverter Control Technologies", page 14. In an inverter for an electric railway vehicle, when the output voltage fundamental wave frequency is low as shown in FIG. 2, control for maintaining a constant ratio between the magnitude of the output voltage and the fundamental wave frequency is performed. (The area where this control is performed is referred to as a variable voltage variable frequency area). When the output voltage fundamental wave frequency rises and the output voltage reaches its maximum, frequency control is performed while maintaining the maximum voltage. (A region in which this control is performed is referred to as a constant voltage variable frequency region). In the variable voltage variable frequency region, in order to adjust the output voltage by pulse width modulation control, a multi-pulse mode in which a half cycle of the output voltage is composed of a plurality of voltage pulses is used. On the other hand, in the constant voltage variable frequency range, in order to maximize the voltage utilization rate and reduce the size of the device, a one-pulse mode in which a half cycle of the output voltage is constituted by a single pulse is used. In a conventional inverter using a GTO thyristor as a switching element (hereinafter referred to as a GTO inverter), as shown in FIG. 3, as the output voltage fundamental wave frequency increases, the number of pulses included in one cycle is reduced. A multi-pulse mode of a pulse number switching method in which the number is gradually reduced by switching is used. This is because the upper limit of the switching frequency of the GTO thyristor is several hundred Hz. In this method, since the switching frequency becomes discontinuous when the number of pulses is switched, there is a problem that the timbre of the magnetic noise occurs due to the switching of the number of pulses, which is annoying. In the GTO inverter, the output voltage in the three-pulse mode including three voltage pulses in a half cycle of the output voltage and the output voltage in the one-pulse mode depend on the limitation of the minimum off time of the GTO thyristor. %, There is a problem that the torque generated by the motor fluctuates when switching between the three-pulse mode and the one-pulse mode. SUMMARY OF THE INVENTION It is an object of the present invention to provide a two-level inverter device which controls the magnitude of an output voltage from zero to a maximum voltage by a combination of a multi-pulse mode and a one-pulse mode. Eliminates the change in timbre of magnetic noise, and has a multi-pulse mode and 1
An object of the present invention is to reduce the gap of the output voltage in the pulse mode and control the entire output voltage almost continuously. In order to solve the above-mentioned problems, in a two-level three-phase power converter that outputs positive and negative binary voltage pulses, a carrier in pulse width modulation is asynchronous with a modulation wave. Assuming that the amplitude of the carrier is 1, the modulation rate, which is the ratio of the amplitude of the carrier to the amplitude of the modulation wave, exceeds 1, and the AC phase voltage as the output voltage becomes 9 of the maximum output voltage.
Synchronous with the output voltage fundamental wave and a single cycle in a half cycle of the output voltage fundamental wave, provided that the value has reached a value that is larger than 5.6% and at least one voltage pulse is secured near the zero cross of the output voltage fundamental wave. Means for controlling a plurality of switching elements by simultaneously switching over three phases to control in a one-pulse mode for outputting a voltage pulse of FIG. 1 to FIG. 17 show an embodiment of the present invention.
This will be described with reference to FIG. FIG. 4 shows the configuration of the inverter of the present invention in the PWM mode. It operates in the bipolar mode in the low output voltage range, the overmodulation mode in the high output voltage range, and the one-pulse mode in the maximum output voltage range. FIG. 1 is a block diagram showing an embodiment of the present invention.
Voltage type 2 as control converter for induction motor for electric car drive
This is an example using a level inverter. In FIG.
Is an induction motor, 5 is a two-level three-phase PWM that drives it
Inverter, 9 is a DC overhead wire that is a power supply for the inverter,
Reference numerals 7 and 8 denote filter reactors and capacitors on the inverter DC input side. The multi-pulse generating means 2, one-pulse generating means 3, and PWM mode selecting means 4 in FIG. 1 are obtained by integrating the output voltage command E ^* of the inverter and its frequency command Fi ^* by the integrator 1. Phase output voltage fundamental wave phase θx
(The subscript x is a general term for the subscript indicating the phase.
Represents any phase of u, v, w. ) To generate a control signal for the inverter. Of the control signals of the inverter, S1x, S2x, Sx are called switching functions,
It is defined as 1 when the positive arm of the inverter is on and 0 when the negative arm is on. First, a method of generating a control signal for the inverter will be described. FIG. 5 shows an example (one phase) of the multi-pulse generating means 2 of FIG. Here, the switching functions of the bipolar mode and the overmodulation mode are generated by the same means. The output voltage command → modulation rate conversion means 21 obtains the modulation rate A, that is, the amplitude of the modulated wave from the output voltage command E ^* . Assuming that the carrier amplitude is 1, 0 ≦ A ≦ in the bipolar mode.
1, A> 1 in overmodulation mode. In order to make the magnitude of the output voltage fundamental wave coincide with the voltage command, E ^* and A are made to correspond to each other by (Equation 1) in the bipolar mode and (Equation 2) in the overmodulation mode. ## EQU1 ## ## EQU2 ## In the function y = sin (x) 22, the sin of the output voltage fundamental wave phase (equivalent to the modulation wave phase) θx is obtained. A product obtained by multiplying this by the modulation factor A is a modulated wave ax. The modulation wave ax and the carrier frequency (equivalent to the switching frequency in the bipolar mode) Fc are supplied to the switching function calculation means 24 to obtain the switching function S1x. The switching function calculating means 24 generates a carrier wave, which is a triangular wave having an amplitude of 1 and a frequency Fc, and compares the carrier wave with the value of the modulated wave to generate a switching function. Alternatively, the switching function may be obtained by calculation from the modulated wave ax and the pulse interval without using a triangular wave. FIGS. 6 and 7 show examples of waveforms of the switching function in the bipolar mode and the overmodulation mode obtained by the triangular wave comparison. In the inverter device of the present invention, a device capable of switching at several kHz, such as an IGBT or a large-capacity power transistor, is used as a switching element (herein, collectively referred to as an IGBT inverter hereinafter). And the carrier are asynchronous. In the bipolar mode shown in FIG. 6, since 0 ≦ A ≦ 1, the carrier 242 corresponds to the switching function 243, and the carrier 242 and the modulation wave 241 are not synchronized. Furthermore, since A> 1 in the overmodulation mode shown in FIG. 7, a switching function 246 of a wide pulse is obtained in a portion where A exceeds 1, and a comparison between the carrier 245 and the modulated wave 244 is performed in other portions. A switching function 246 is obtained. Also in this overmodulation mode, the carrier wave 245 and the modulated wave 244 are generated asynchronously. As described above, in the figure, the method of obtaining the switching function by comparing the carrier wave and the modulated wave is shown for the sake of understanding, but the switching function can also be obtained by calculation from the modulated wave ax and the pulse interval. As a result, the switching frequency becomes constant in the bipolar mode, and can gradually approach the switching frequency in the one-pulse mode described below in the overmodulation mode. In this multi-pulse mode, since the modulation wave and the carrier are asynchronous, the carrier frequency needs to be sufficiently higher than the modulation wave frequency.
Desirably higher than about twice. FIG. 8 shows an example of the waveform of the switching function generated by the one-pulse generating means of FIG. The value of the switching function S2x is 1 when the sign of the fundamental wave 31 (the amplitude is arbitrary) of the output voltage is positive, and S when the sign is negative.
The value of 2x is 0. Next, a description will be given of a combination of the multi-pulse mode and the one-pulse mode for controlling the high output voltage range. As a document written on the overmodulation method, Proceedings of the 1991 IEEJ National Conference on Industrial Applications No.1
06 “Voltage-type three-phase PWM inverter overmodulation control method”
There is. According to this, the operation of the 6-step inverter, that is, the one in which the modulation rate of the overmodulation
It is stated that this is a pulse mode operation. However, when the one-pulse mode is realized by extending the overmodulation mode (the one-pulse mode is realized by making the modulation rate extremely large), the following inconvenience occurs. First, the point at which the overmodulation mode and the one-pulse mode are switched depends on the switching frequency and cannot be set arbitrarily. Second, when the modulated wave in the overmodulation mode and the carrier wave are asynchronous (hereinafter referred to as asynchronous PWM), the modulated wave near the boundary between the overmodulation mode and the one-pulse mode is affected by the turn-on and turn-off times of the element. A pulse near the zero cross may or may not appear. As a result, an imbalance occurs between the positive and negative of the output voltage, and a beat phenomenon occurs in which a low-frequency pulsation is superimposed on the load current of the inverter. Third, in the overmodulation mode, as shown in FIG. 7, the output voltage waveform (equivalent to the waveform of the switching function described later) is a modulated wave (equivalent to the output voltage fundamental wave) and the pulse interval near the zero cross becomes uniform. In other words, the pulse generation period is divided into a uniform pulse portion (equally-spaced pulse) and a wide pulse portion centered on the peak of the modulated wave. Can occur. In this case, the load current of the inverter may be disturbed, and the switching element may be destroyed due to the overcurrent or the generated torque of the motor may fluctuate significantly. To solve these problems, a voltage for switching between the overmodulation mode and the one-pulse mode (hereinafter, referred to as a transition voltage) and a phase of the output voltage fundamental wave (hereinafter, referred to as a transition phase). Manage. First, management of the transition voltage will be described.
When the transition voltage is set and the overmodulation mode and the one-pulse mode are switched on the basis of the transition voltage, it is desirable that the set value of the transition voltage be as close as possible to the output voltage of the one-pulse mode, that is, a value close to 100%. This is because the smaller the difference from the maximum value of the output voltage in the overmodulation mode, the smaller the fluctuation of the generated torque of the motor at the time of switching. However, in the asynchronous PWM, the width of each voltage pulse included in one cycle of the fundamental wave of the output voltage is different in each cycle, and the output voltage is changed in the overmodulation mode.
As the number approaches 100%, when the number of pulses near the zero cross of the output voltage fundamental wave decreases, this effect becomes apparent, an imbalance occurs between the positive and negative of the output voltage, and a beat phenomenon occurs in the load current of the inverter. FIG. 9 shows an example of this state. FIG. 10 shows an example of the relationship between the average number of pulses near the zero cross of the output voltage fundamental wave and the current pulsation due to the beat phenomenon. As shown in FIG. 7, the absolute value of the modulated wave is 1.0.
Since the following portions correspond to equally-spaced pulses, the average number of pulses is given by Expression (3). The current pulsation rate was defined by (Equation 4). [Equation 3] (Equation 4) From FIG. 10, if at least one pulse is not secured near the zero cross of the output voltage fundamental wave, the low frequency pulsation of the load current of the inverter due to the beat phenomenon becomes extremely large. Therefore, the set value of the transition voltage is set to a value that secures at least one or more voltage pulses near the output voltage fundamental wave zero cross. Since this value depends on the output voltage fundamental wave frequency Fi ^* and the carrier frequency Fc in the multi-pulse mode, a means for calculating these values by calculation may be provided, or the upper limit of the output voltage fundamental wave frequency Fi ^* may be used in advance. It may be set by calculation. Next, management of the transition phase will be described. Depending on the phase of the output voltage fundamental wave when switching between the overmodulation mode and the one-pulse mode, the state of the transient fluctuation of the load current of the inverter and the generated torque of the motor immediately after the switching differs. FIG. 11 shows an example of the current fluctuation. FIG.
FIG. 12 shows a case where all three phases are switched at 0 ° in the phase of the U-phase output voltage fundamental wave as shown in FIG. 12, and a transient fluctuation is observed in the current immediately after the switching. On the other hand, FIG. 13B shows a case where the three phases are switched at 90 ° with the phase of the U-phase output voltage fundamental wave as shown in FIG. 13, and there is almost no transient fluctuation of the current. FIG. 14 shows an example of the relationship between the phase of the output voltage fundamental wave (U-phase reference) and the transient current fluctuation when switching from the overmodulation mode to the one-pulse mode collectively in three phases.
Here, the current fluctuation rate is defined by (Equation 5). [Equation 5] In FIG. 14, the phase of the output voltage fundamental wave is 60
The current fluctuation rate increases for each ゜. This is a case where the overmodulation mode and the one-pulse mode are switched when any one of the three phases is an equally-modulated pulse in the overmodulation mode. Since the unbalance increases, the transient current fluctuation also increases. Therefore, as shown in FIG. 15, by setting the transition phase to a portion where all phases become wide pulses in the overmodulation mode, transient current and torque fluctuations can be suppressed. Here, in order to switch overmodulation mode and one-pulse mode collectively in three phases, there must be a section in which the output voltage of the overmodulation mode in all three phases becomes a wide pulse. For this purpose, the intersection (U
30 °, 90 °, 150 ° with respect to the phase of the phase modulated wave
{, 210}, 270 °, 330 °), the absolute value of the modulated wave must be greater than 1.0. Considering the case of 30 °, au = Asin30 ゜> 1.0, A> 2, and in the overmodulation mode, the correlation between the modulation factor A and the output voltage E ^* is given by (Equation 2), so E ^* > 95. Must be 6%. Therefore, in order to switch the overmodulation mode and the one-pulse mode at the same time in three phases, the transition voltage is larger than 95.6%, and a value for securing at least one voltage pulse near the zero crossing of the output voltage fundamental wave by overmodulation. Become. FIG. 16 shows an example of the configuration of the PWM mode selection means 3 for realizing the management of the transition voltage and the transition phase. The mode selection command generation means 32 compares the transition voltage Ec set in the transition voltage means 31 with the voltage command E ^* , and generates a mode selection command Mc indicating whether to select the multi-pulse mode or the one-pulse mode. . [0034] Although it was decided to determine the mode select command Mc based on the output voltage command E ^* in this case, the output voltage command E ^*
Corresponds uniquely to the modulation rate A, so that the modulation rate Ac corresponding to the transition voltage is set in advance, and this and the modulation rate A
May be compared to generate a mode selection command Mc. In the variable voltage variable frequency region, the output voltage command and the output voltage fundamental frequency also uniquely correspond to each other. Therefore, the output voltage fundamental wave frequency Fic corresponding to the transition voltage is set in advance, The mode selection command Mc may be generated by comparing Fi ^* . The transition phase management means 44 refers to Mc,
If mode switching is required, the phase θ of the output voltage fundamental wave
x is compared with the transition phase θc set in the transition phase setting means 43. If θx has reached θc, the mode selection signal M is switched. The mode selection switches 45, 46, and 47 select either the output S1x of the multi-pulse generation means or the output S2x of the one-pulse generation means according to the mode selection signal M, and determine the switching function Sx. The transition phase may be managed by the following method. Take the absolute value of the modulated wave of each phase,
If all three phases are greater than 1.0, then all phases are at the point of the overmodulated wide pulse.
Therefore, at such time, the output of the multi-pulse generating means and the output of the one-pulse generating means are switched. As described above, the gap between the output voltages in the multi-pulse mode and the one-pulse mode is reduced from about 10% in the conventional GTO inverter to about 1-2%, and the magnitude of the output voltage is reduced from zero to the maximum voltage. Control almost continuously until
Further, it is possible to configure a two-level inverter device that can smoothly perform switching without changing the current or the generated torque of the motor when switching between the multi-pulse mode and the one-pulse mode. The relationship between the output voltage fundamental wave frequency and the switching frequency in the present invention is as shown in FIG. 17, and there is no large discontinuity as in the conventional inverter modulation system of FIG. A great tone change can be eliminated. According to the inverter device which controls the magnitude of the output voltage from zero to the maximum voltage by the combination of the multi-pulse mode and the one-pulse mode, it is possible to eliminate the discontinuous change of the magnetic noise. The output voltage range can be controlled almost continuously.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a configuration diagram showing one embodiment of the present invention. FIG. 2 is a diagram showing operating characteristics of a vehicle inverter. FIG. 3 is a diagram illustrating an example of a conventional inverter modulation method. FIG. 4 is a diagram showing operating characteristics of the inverter according to the present invention. FIG. 5 is a diagram showing an example of a configuration of a multi-pulse generation unit. FIG. 6 is a diagram showing a modulation wave, a carrier wave, and a switching function in a bipolar mode. FIG. 7 is a diagram showing a modulated wave, a carrier wave, and a switching function in an overmodulation mode. FIG. 8 is a diagram showing a fundamental function of an output voltage and a switching function in a one-pulse mode. FIG. 9 is a diagram showing how a beat phenomenon occurs. FIG. 10 is a diagram showing a relationship between an average number of pulses near an output voltage fundamental wave zero cross and current pulsation. FIG. 11 is a diagram showing that the state of transient current fluctuation immediately after mode switching differs depending on the transition phase. FIG. 12 is a diagram showing the switching timing of FIG. FIG. 13 is a diagram showing the switching timing of FIG. 11 (b). FIG. 14 is a diagram showing a relationship between a transition phase and a transient current fluctuation immediately after mode switching. FIG. 15 is a diagram showing a transition phase settable section. FIG. 16 is a diagram showing an example of a configuration of a PWM mode selection unit. FIG. 17 is a diagram showing a relationship between an output voltage fundamental wave frequency and a switching frequency of the inverter according to the present invention. [Description of Signs] 1 ... Integrator, 2 ... Multi-pulse generating means, 3 ... 1-pulse generating means, 4 ... PWM mode selecting means, 5 ... 2-level three-phase P
WM inverter, 6 induction motor, 7 filter reactor, 8 smoothing capacitor, 9 DC overhead wire, 21 frequency command → modulated wave amplitude reference conversion means, 22 function y = sin
(x), 23: switching frequency, 24: switching function calculating means, 241, 244: modulated wave ax, 24
2,245 carrier wave c, 243,246 switching function S1x, 31 output voltage fundamental wave, 41 transition voltage setting means, 42 mode selection command generating means, 43 transition phase setting means, 44 transition phase management means , 45, 46, 4
7 Mode switch.

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Mutsuhiro Terunuma 1070 Ma, Katsuta-shi, Ibaraki Pref. Mito Plant of Hitachi, Ltd. (56) References JP-A-5-146160 (JP, A) JP-A-5-146 161364 (JP, A) JP-A-3-32391 (JP, A) JP-A-5-316735 (JP, A) JP-A-62-163589 (JP, A) JP-A-62-131791 (JP, A) JP-A-63-287372 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H02M 7/48

Claims (1)

(57) [Claim 1] An AC phase voltage of a variable frequency variable voltage from a DC voltage by controlling a plurality of switching elements by a signal which is pulse width modulated based on a modulated wave of each of three phases. In a two-level three-phase power converter that outputs positive and negative binary voltage pulses, the carrier in the pulse width modulation is asynchronous with the modulated wave, and when the amplitude of the carrier is 1, the amplitude of the carrier is The modulation rate, which is the ratio of the amplitudes of the modulated waves, exceeds 1, the AC phase voltage as the output voltage is greater than 95.6% of the maximum output voltage, and at least one voltage pulse is near the output voltage fundamental wave zero crossing. Is synchronized with the output voltage fundamental wave, and is switched to the one-pulse mode control in which a single voltage pulse is output in a half cycle of the fundamental wave. Power conversion apparatus characterized by comprising means for controlling a plurality of switching elements.