JPH06197550A - Controlling method for switching frequency - Google Patents

Abstract

PURPOSE:To make an electromagnetic sound generated by a carrier at the mode changing time of a pulse exert an acoustic effect matching the accelerating feeling of a vehicle by making the carrier to have such a characteristic that the frequency of the carrier rises together with the frequency of an inverter. CONSTITUTION:A frequency command F* to an inverter is inputted to the variable-frequency sine wave generating means of a block 2 and a voltage command V* to the inverter is inputted to the sine-wave amplitude increasing/ decreasing means of a block l. The block generates a modulation factor alpha*by normalizing a given voltage command and further generates a three-phase sine-wave signal V3* by adding the factor alpha* to the block 2 and multiplying a sine-wave signal generated in the block 2 by the factor alpha*. A carrier frequency command Fc* is inputted to the constant-amplitude triangular wave generating means and generates a constant-amplitude triangular carrier signal Vc* having a frequency proportional to the command Fc* and, at the same time, supplies a one-pulse mode switching command Fg* to the carrier wave control means of a block 4. The signals V3* and Vc* are supplied to a block 5 and a comparing means outputs a signal Si after comparing the signals with each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of controlling a switching frequency of an inverter by switching a pulse mode, and more particularly to a sudden change in voltage when a high frequency switching state shifts to a low frequency switching state and a corresponding rush current. The present invention provides a switching frequency control method that suppresses and reduces fluctuations in motor noise during acceleration.

[0002]

2. Description of the Related Art Since the allowable switching frequency of a switching element that constitutes an inverter has a limit determined by the performance of the element and a cooling device, when the load of the inverter is an electric motor, PWM is performed by high frequency switching in the starting and low frequency regions. A method of performing voltage control and switching to constant voltage control without PWM in the medium to high speed range has been put into practical use. In this constant voltage control state, a half cycle of the line voltage of the inverter becomes one uncontrolled pulse, which is called a one-pulse mode.

A switching frequency control method in the conventional inverter control method will be described with reference to FIG. FIG. 4 shows a conventional inverter frequency Fi and switching frequency Fs.
An example of the relationship of w is shown. Here, when the switching frequency by PWM at a low frequency is sufficiently high, a carrier wave that determines switching is an asynchronous carrier wave irrelevant to the phase of the inverter fundamental wave. In general, the number of carrier waves included in an inverter half cycle is at most 1 regardless of the phase relationship between the carrier and the fundamental wave.
There are only individual differences. Therefore, when the frequency of the carrier is sufficiently high, the difference of one carrier wave has almost no effect on the waveform distortion of the inverter, and there is no need to restrain the synchronization relationship between the two waveforms.

However, when the inverter frequency rises and approaches the carrier frequency, the difference of one carrier wave in the inverter half cycle gradually increases the influence. Therefore, as shown in the figure, the inverter frequency Fi is FN
When it reaches 1, the mode is switched to the N1 pulse mode in which the carrier frequency is N1 times the inverter frequency so that N1 carrier waves are included in the half cycle. The pulse mode is switched sequentially as the switching frequency increases,
The 1-pulse mode is reached via the N1, N2 and N3 pulse modes. N1, N2, and N3 are waveform symmetry,
Numbers such as 21, 15, 9, 3, etc. are often used. The method described here is incorporated in the switching frequency control of almost all electric vehicle drive inverters.

The broken line in FIG. 4 shows the change in the effective value of the inverter output voltage. As you can see, PWM
In the region, the voltage is almost proportional to the inverter frequency and is a constant voltage in the 1-pulse mode. The 1-pulse mode is
The voltage waveform contains a large amount of low-order harmonics, but the fundamental wave has the maximum amplitude, so it is used to apply the maximum voltage to the motor. If the 1-pulse mode is not used, the current of the electric motor and the inverter increases as the voltage decreases, and the size of the device increases.

[0006]

A conventional inverter that switches the pulse mode according to the inverter frequency is
It has been pointed out that when the pulse mode is changed, the magnetic noise of the electric motor changes discontinuously, which causes discomfort to passengers in the electric vehicle.

In recent years, elements having excellent frequency characteristics have been developed and put into practical use, and in the PWM control mode at a low frequency, a switching frequency higher than before has been adopted. For example, a conventional transistor that was switched at 1 to 2 kHZ can be switched at 10 to 15 kHZ by using an IGBT. An inverter employing such an element can not only propose harmonics of current, but also can raise the carrier frequency sufficiently higher than the inverter frequency. No need to sync to.
That is, the asynchronous PWM mode, which is sufficiently higher than the fundamental wave, can be used up to the 1-pulse mode. However, P
The maximum value of the inverter output voltage during WM control is only (π / 2√3) (about 90%) of the fundamental wave in 1-pulse mode.
% The inverter output voltage increases rapidly and rush current flows.

[0008]

A means for solving the problems will be described with reference to FIGS. 1 and 2. FIG. 1 is a diagram showing carrier frequency characteristics with respect to an inverter fundamental frequency when the present invention is implemented.
In the case of an inverter using a high-speed switching element such as an IGBT, it is not necessary to switch the pulse mode in the PWM area as described above, and an asynchronous carrier may be used. Further, in the transient state to the 1-pulse mode, the pulse frequency is not set according to the frequency, but the carrier frequency is continuously lowered at a certain short time τ as shown in FIG. 2 to reach the 1-pulse mode. Good.

When the asynchronous carrier frequency in FIG. 1 is not very high and electromagnetic noise is still audible from the motor, the carrier frequency does not have to be constant, and the electromagnetic noise generated by the carrier is increased by the characteristic that the carrier frequency increases with the inverter frequency. To provide an acoustic effect that matches the acceleration feeling of the vehicle. Alternatively, it can be used in combination with a known technique such as modulating the carrier frequency using a random number table to disperse the electromagnetic sound frequency spectrum and change the audibility of the electromagnetic sound.

[0010]

The electromagnetic sound is hardly generated in the case of the asynchronous high frequency carrier, but is generated in the low pulse mode such as 3, 9 because the carrier frequency is lowered. Therefore, according to the present invention, there is no pulse mode switching during acceleration, and it is possible to eliminate an unpleasant fluctuation of electromagnetic sound.
Further, in the present invention, the switching to the 1-pulse mode is finished in a short time period, the change of the electromagnetic sound is also promptly finished, and the voltage jump is suppressed to the minimum. Even when switching from a low pulse mode of about 3 and 9 pulse modes, a jump in the voltage due to the minimum pulse width determined by the switching element (about 5 microseconds in the case of the IGBT) occurs, but there is almost no problem. For example, when switching from the 3-pulse mode to the 1-pulse mode at an inverter frequency of 50 Hz, the half cycle period of the line voltage is 10 milliseconds and the number of cuts in the voltage waveform is 2. Therefore, the jump by the minimum pulse width is 10 microseconds. It is 0.1%.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in more detail with reference to the accompanying drawings. FIG. 3 is a block diagram showing an embodiment to which the present invention is applied. That is, it shows a function for generating an energization signal to each element of the inverter from a voltage command and a frequency command to the PWM inverter.

In FIG. 3, when the frequency command F * to the inverter is used, the frequency command F * is input to the variable frequency sine wave generating means of the block 2. The block 2 is means for generating a three-phase sine wave signal of a constant amplitude having a frequency proportional to the commanded signal.
And a sin θ data table corresponding thereto are stored, and this is read out in accordance with the frequency command F *. However, the information phase θ is obtained by time integration of the frequency command F * of the inverter.

Next, assuming that the voltage command to the inverter is V *, the voltage command V * is input to the sine wave amplitude amplification / attenuation means of the block 1. The block 1 normalizes the commanded voltage command to obtain a modulation factor α * (0 ≦ α * ≦
Make 1). The modulation factor α * is applied to block 2, which produces a three-phase sinusoidal signal V3 * by multiplying the amplitude of the internally generated sinusoidal signal by α *. The three-phase sine wave signal V3 * is the phase voltage Vu *, Vv that should be generated by the inverter.
Includes * and Vw *.

The carrier frequency command to the inverter is Fc.
Then, the carrier frequency command Fc * is input to the constant amplitude triangular wave generating means of block 3. The block 3 generates a constant amplitude triangular wave carrier signal Vc * having a frequency proportional to the input Fc *. The carrier frequency may be a constant value or a characteristic that increases with the inverter frequency as described above, or a modulated wave using a random number table. When the 1-pulse mode is set, the command value Fc * is decreased toward zero in a short time as described above.

At the same time, the one-pulse mode switching command F
g * is supplied to the carrier wave control means of block 4. 1
The switching logic signal of the pulse mode switching command Fg * is (Fg =
The PWM mode is represented by 1), and the 1-pulse mode is represented by (Fg = 0). And (Fg =
In the case of 1), the output of the block 4 allows the block 3 to pass Vc * of the triangular wave output as it is, but when it is changed to (Fg = 0), after the time limit as shown in FIG. To zero.

The two systems of signal waveforms V3 * and Vc * are applied to the block 5. The block 5 compares both signals by a comparison means, and when (V3 *> Vc *) (Si = 1),
When (V3 * ≦ Vc *), (Si = 0) is output. However, Si is a switching signal output from the block 5, which is information for instructing the switching state of each phase of the three-phase inverter, and is composed of three ON / OFF signals corresponding to each phase voltage of V3 *. In particular, in the case of (Vc * = 0), since V * of the voltage signal becomes (Si = 1) in the positive half cycle and (Si = 0) in the negative half cycle, the frequency that does not include the slit by the PWM. The F * three-phase rectangular wave signal (1 pulse mode energization signal) is obtained.

In FIG. 3, four command signals V
Since the control means at the front stage having the function of generating *, F *, Fc *, Fg *, etc., and the control means at the rear stage based on the switching signal Si are not directly related to the present invention, various known methods are known. It goes without saying that it may be a system type. Further, the method of decreasing the carrier frequency for shifting from the high-frequency PWM control state to the 1-pulse mode may use a temporally continuous function without taking the pulse mode determined as described above. Low fixed pulse modes such as 9 and 3 may be passed stepwise in a short time period. The time required for this may be set so that the transient phenomena of the current due to the minute jumps at the respective stages do not overlap. That is, it may be about several milliseconds to several tens of milliseconds. When switching between pulse modes other than 1-pulse mode, the individual pulse width can be controlled even if the number of pulses in a half cycle changes, so basically (pulse width x number of pulses) is This is because if it does not change, the voltage jump can be almost suppressed.

[0018]

According to the present invention, it is possible to reduce the inverter output voltage jump at the time of switching the one-pulse mode, which is a problem when the high frequency switching is applied to the inverter driving vehicle.

[Brief description of drawings]

FIG. 1 is a diagram showing an inverter frequency and a carrier frequency characteristic shown to facilitate understanding of the present invention.

FIG. 2 is a diagram showing the inverter frequency and carrier frequency characteristics shown together with FIG.

FIG. 3 is a block diagram showing an embodiment to which the present invention is applied.

FIG. 4 is a switching frequency characteristic curve diagram showing a relationship between a conventional inverter frequency and a switching frequency.

[Explanation of symbols]

 Fi Inverter frequency Fsw Switching frequency F * Frequency command V * Voltage command α * Modulation rate V3 * Three-phase sine wave signal Fc * Carrier frequency command Vc * Triangular wave carrier signal Fg * 1 Pulse mode switching command Si switching signal

Claims (2)

[Claims] 1. An inverter control method for performing switching control from asynchronous PWM control to one-pulse control mode with a predetermined inverter fundamental frequency as a boundary, in which a carrier frequency is continuously gradually reduced at a preset time period and then one-pulse mode is set. A switching frequency control method, characterized in that the switching frequency is switched to. 2. An inverter control method for performing switching control from asynchronous PWM control to one-pulse control mode with a predetermined inverter fundamental frequency as a boundary, in which a synchronous pulse control mode of an intermediate carrier frequency is stepped several times at a preset time period. The switching frequency control method is characterized in that the mode is switched to the 1-pulse mode after being automatically passed.