JPH02179277A - Method and device for controlling power converter - Google Patents

Abstract

PURPOSE:To easily and accurately perform compensation by adding an offset current to a current value in a discrete-time system discriminating a current direction and performing the dead time compensation. CONSTITUTION:A power converter control device is constituted of a control arithmetic part, dead time compensation part 2, PWM signal modulation part, dead time setter part and a current detector. This compensation part 2 is provided with registers 2a to 2b, zero cross detecting circuit 2c, current direction detecting circuit 2d, selector 2e and arithmetic parts 2g, 2m. As the result, being based on a voltage command value, by obtaining a PWM signal to be supplied to an inverter setting the dead time, switching action is performed obtaining a faithful output.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a power converter control method and device,
More specifically, the present invention relates to a power converter control method and apparatus for feedback-controlling a power converter configured by bridge-connecting a plurality of switching elements and converting direct current to alternating current using a discrete time system.

<Prior Art and Problems to be Solved by the Invention> Power converters that convert direct current to alternating current in order to drive electric motors and the like have been widely used. As a method of controlling the power converter, a pulse width modulation signal (hereinafter referred to as PW
Generally, a method is adopted in which a gate signal (abbreviated as M signal) is supplied as a gate signal.

In this case, the pair of switching elements to be selectively turned on, due to the delay in the switching operation,
In order to eliminate the inconvenience of both being rendered conductive, as shown in Figure 5, the conduction periods of both switching elements are slightly shifted during switching operation (hereinafter, this time difference is referred to as dead time). ).

In addition, FIG. 5A shows a triangular wave carrier waveform and a voltage command value, FIG. 5B shows a switching pulse waveform generated based on the triangular wave carrier waveform and voltage command value, and FIG. The switching states of a pair of switching elements are shown. And, Fig. 5C,
Td during D is dead time.

By controlling each switching element in this way,
Although it is possible to reliably prevent a short circuit between the switching elements, since both switching elements are in a cut-off state during the dead time Td, no current control can be performed at all.

As a result, distortion of the current waveform occurs during the dead time Td, and especially near zero cross where the current value is small, the distortion of the current waveform becomes impossible to ignore.

Further, the output voltage waveform obtained based on the switching state of each switching element is shown in FIG. 5E.

It will be either F. In other words, when current flows from the connection point between the switching elements toward the load, the ratio of the cutoff period becomes large, as shown in E in the figure, and the output voltage of one cycle, taking into account the conduction period and the cutoff period, becomes the voltage. becomes smaller than the command value. Conversely, when current flows from the load toward the connection point between the switching elements, the ratio of the conduction period increases as shown in F in the same figure, and the output voltage of one cycle, taking into account the conduction period and the cutoff period, becomes It becomes larger than the voltage command value. That is, in the former case, the output voltage V is V - VL - (VTd - VTL)
(1), and in the latter case V −VL + (VTd−VTL)
-=(2) (where VL is the voltage command value, VTd is the voltage drop due to dead time, and VTL is the voltage drop due to switching delay of the switching element).

Therefore, by changing the voltage command value VL so as to cancel the influence of the second term in each of the above equations, the output voltage can be accurately controlled regardless of the presence of dead time.

Specifically, ■ a method of detecting the PWM signal output voltage of each phase, comparing it with the PwM pulse signal voltage command value, and adding the deviation of the PWM signal output voltage detection value to the PWM output voltage command value; A method of detecting the output current polarity of and adding a predetermined value corresponding to the polarity of the output current that can cancel the influence of the second term of each of the above equations to the output voltage command value; A method has been proposed in which a compensation voltage value for compensating for output voltage distortion is calculated based on a current command value and a primary angular frequency command value, and is added to a voltage command value in a rotating magnetic field coordinate system.

However, when adopting method ■ above, analog
Since a digital converter (hereinafter referred to as an A/D converter) is required, there is a problem that the configuration becomes extremely complicated, and the PWM output voltage detection value is affected by switching noise etc. contains an error, and there is a problem in that it is no longer possible to accurately compensate for the effects of dead time.

When adopting the above method (■), it is extremely difficult to accurately detect the current polarity, and as a result, compensation may be made that increases the effect of dead time, which may actually increase the ripple. There is a problem with letting it happen. In particular, when a discrete time system is adopted, the timing at which current polarity is detected and the timing at which compensation is performed based on the detected current polarity inevitably lag, and as a result, the effects of dead time are reduced. Compensation may be made in an accretive manner.

When adopting method (■) above, feedforward compensation is adopted in which the occurrence of output voltage distortion is predicted based on the current command value and the primary angular frequency command value, and compensation is performed based on this prediction. There is no guarantee that accurate compensation can be provided.

<Object of the invention> This invention was made in view of the above problems,
It is an object of the present invention to provide a power converter control method and apparatus that can extremely easily and accurately compensate for the effects of dead time when controlling a power converter using a discrete time system.

Means for Solving the Problems> In order to achieve the above object, the power converter control method of the present invention applies an offset current to the discretely obtained current value to make the current direction equal to the actual current. In addition, this method determines the direction of the current based on the current value to which an offset current is added, and performs dead time compensation based on the direction of the current.

However, the offset current is preferably set after a zero cross is detected based on discretely obtained current values.

In order to achieve the above object, the power converter control device of the present invention includes a current value correction means for adding an offset current to the discretely obtained current value to make the current direction equal to the actual current; The present invention includes a direction determining means for determining the direction of current based on the current value to which the current is applied, and a dead time compensating means for performing dead time compensation based on the direction of the current.

Effect〉 With the above power converter control method, when a power converter that converts direct current to alternating current is feedback-controlled by a discrete time system, the current value obtained discretely by the discrete time system is By adding an offset current that makes the current direction equal to the actual current, the direction of the current can be accurately determined based on the current value to which the offset current is added, eliminating the influence of time delays and the like. Therefore, by performing dead time compensation based on the current direction determined in this manner, the influence of dead time can be completely canceled.

If the offset current is set after a zero crossing is detected based on discretely obtained current values, it is possible to alternately set offset currents with different polarities each time a zero crossing is detected. Therefore, dead time compensation can be performed to completely cancel out the effects of dead time in a predetermined range following the detected zero crossing.

With the power converter control device having the above configuration, when a power converter that converts DC to AC is feedback-controlled by a discrete time system, the current value correction means is used to adjust the discretely obtained current values. , adding an offset current that makes the current direction equal to the actual current, and accurately determining the direction of the current by a direction determining means based on the current value to which the offset current is added, eliminating the influence of time delay, etc.;
By performing dead time compensation by the dead time compensating means based on the determined direction of the current, the influence of dead time can be completely canceled.

In addition, highly accurate dead time compensation can be achieved with a relatively simple overall configuration.

<Example> Hereinafter, embodiments will be described in detail with reference to the accompanying drawings showing examples.

FIG. 2 is a schematic block diagram showing the electrical configuration of an embodiment of the power converter control device of the present invention, and shows the current command value! *
By manually performing the necessary calculations, the voltage command value V
A control calculation section (1) that obtains the voltage command value V*, a dead time compensation section ■ that performs dead time compensation on the voltage command value ■* and obtains the voltage command value V* after compensation, and a dead time compensation section ■ that obtains the voltage command value V* after compensation. a modulation section (3) for obtaining a PWM signal; a dead time setting section (4) for setting a dead time for the PWM signal;
An inverter (5) that performs a switching operation based on a PWM signal with a dead time set;
5) and a current detector (6) inserted between the electric motor. The current detection signal obtained discretely by the current detector (6) is fed back to the dead time compensator (2) and also fed back to the input side of the control calculation unit (1).

Figure 1 is a block diagram showing in detail the configuration of the dead time compensator (2), including registers (2a) (2b) that hold two types of voltage command compensation values, and a zero-cross detection circuit (2C). , a current direction detection circuit (2d) that detects the direction of the current before zero crossing, a selector (2e) that selects a corresponding register based on the current direction detected by the current direction detection circuit (2d), and zero crossing detection. Based on the zero cross detection signal output from the circuit (2c), a register (2r) temporarily holds the contents of the register selected by the selector (2e), one voltage command value and a register (2f). It has a calculation section (2g) that performs compensation calculation based on the content. Furthermore, there are registers (2h) (21) that hold two types of offset values, a selector (2j) that selects the corresponding register (2h) (21) based on the detected current direction, and zero cross detection. Based on the signal, the register (2k) temporarily holds the contents of the register selected by the selector (2j), and the zero-cross detection circuit (2c) adds the sampled current value and the contents of the register (2k). ).

The operation of the power converter control device having the above configuration is as follows.

The corresponding voltage command value 1 is obtained by supplying the value obtained by subtracting the detected current value I from the current command value I* to the control calculation unit (1).
By performing dead time compensation on this voltage command value V* by the dead time compensator 0, a compensated voltage command value V* is obtained. Therefore, by obtaining a PWM signal based on this voltage command value V* and supplying it to the inverter 6) with a dead time Td set,
A switching operation can be performed to obtain an output faithful to the voltage command value V*. As a result, the electric motor (
7) can be operated faithfully to the command value.

That is, one voltage command value above corresponds to vL in the above equation (1).
Td-V TL). Therefore, one voltage command value output from the dead time compensator ■ is v * −V
* If it is (VTd - VTL), voltage fluctuations caused by dead time setting can be reliably compensated for.

This will be explained in more detail with reference to the waveform diagrams shown in FIGS. 2 and 3.

A detected current waveform shown by a broken line in FIG. 3 is obtained based on the current detection signal discretely obtained by the current detector (6). This detected current waveform lags behind the actual current (see the solid line in FIG. 3) by a predetermined time determined by the sampling time and the like. Therefore, as it is, it is impossible to detect the zero crossing of the actual current in real time, and it is also impossible to accurately detect the polarity of the actual current in the vicinity of the zero crossing. However, the relationship between the detected current and the actual current is only delayed by a predetermined time. Therefore, once the zero cross is detected, the zero cross can be realized by adding a predetermined offset current to the detected current. A correction current (see the two-dot chain line in FIG. 3) that matches the current can be obtained.

That is, the registers (2a) (2) are set in correspondence with the polarity of the actual current.
b) It is sufficient to store the voltage command compensation value (V Td - V TL) in advance, and accurate dead time compensation can be performed as follows.

First, the polarity of the current is detected by the current direction detection circuit (2d) at a timing before the zero crossing is detected,
The corresponding register is selected by operating the selector (2e) based on the detected current direction. Also,
The sampling current value is supplied to the zero-crossing detection circuit (2c) with the offset value selected by the selector (2j) added thereto. Thereafter, at the timing when the zero-crossing detection signal is output from the zero-crossing detection circuit (2C), the contents of the register selected by the selector (2e) are temporarily held in the register (2f), and the other offset value is is selected by the selector (2j).

Therefore, by performing the necessary compensation calculation v*-V*±(V Td-V TL) in the calculation unit (2g) based on the voltage command value V* and the contents of the register (2r), the influence of dead time Td can be reduced. It is possible to obtain a voltage command value ■* that can cancel out.

In addition, the above series of operations is performed every half cycle of the current waveform, and zero crossings are detected based on the sampled current value with the offset value added, so there may be a time delay. The precise current polarity can be identified without any problems, and good dead-time compensation can be achieved over the entire range.

Figure 3 shows the case where an ideal sine wave-like real current is flowing, but since the real current is generally influenced by noise etc., it occurs multiple times near the zero cross of the ideal waveform. Although there is a high possibility that the polarity will be reversed, by employing the power converter control device described above, dead time compensation can be performed accurately even in such a case.

FIG. 4 is a flowchart showing an embodiment of the power converter control method of the present invention, in which the current value supplied from the inverter (5) to the electric motor (6) is sampled in step Determine whether polarity is identified.

If the polarity is not identified, the polarity of the current is identified based on the sampling data in step 2, and in step 2, one of two offset values to compensate for the delay in the detected current is selected based on the polarity. , In step (2), a dead time is set for the voltage command value to control the switching timing of the inverter 6), and the process in step (2) is performed again.

Furthermore, if it is determined in step ■ that the polarity of the current has already been detected, a zero cross is detected in step ■ based on the value obtained by adding an offset value determined based on the polarity of the current to the sampling data. If zero crossing is not detected,
In step ■, the corresponding offset value is selected based on the polarity of the previous current, and in step ■, the dead time compensation voltage value ±(V
Td - V TL) is added to calculate V voltage command values with dead time compensation, and in step 2, the dead time Td is set and the above voltage command value V*
The switching control of the inverter (5) is performed based on the following. After that, the process of step (2) is performed again. If it is determined in step (2) that a zero cross has been detected, the other offset value is selected in step [phase], and then the process in step (2) is performed.

Therefore, by changing the offset value every time the polarity of the current is reversed, it is possible to accurately detect the zero-crossing timing of the actual current. As a result, although it is a discrete system, it is possible to eliminate the detection delay when the polarity of the current changes, and it is possible to perform accurate dead time compensation.

Note that the present invention is not limited to the above-described embodiments; for example, after detecting the polarity of the current, it is possible to automatically switch the offset value selected each time a zero cross is detected. , it is possible to make various changes to the configuration without changing the gist of the invention.

<Effects of the Invention> As described above, the first invention adds an offset current that makes the current direction equal to the actual current to the current value discretely obtained by the discrete time system, and the offset current is added. The direction of the current can be determined based on the current value determined in this way, and by performing dead time compensation based on the determined current direction, it is possible to completely cancel out the effects of dead time. It has a unique effect.

The second invention is capable of alternately setting offset currents with different polarities each time a zero cross is detected, and performs dead time compensation to completely cancel out the effects of dead time in a predetermined range following the detected zero cross. It has the unique effect of being able to

The third invention adds an offset current that makes the current direction equal to the actual current to the current value discretely obtained by the discrete time system, and determines the current direction based on the current value to which the offset current is added. By performing dead time compensation based on the direction of the current determined in this way, the effect of dead time can be completely canceled out, and the overall configuration is relatively simple. It has the unique effect of being able to

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing in detail an embodiment of the configuration of the dead time compensator, which is the most important part of the present invention, and FIG. 2 shows the electrical configuration of an embodiment of the power converter control device of the present invention. 3 is a current waveform diagram illustrating the principle of eliminating the delay in current polarity detection. FIG. 4 is a flowchart showing an embodiment of the power converter control method of the present invention. Timing chart explaining time and its effects. ■...Dead time compensation section, (2C)...Zero cross detection circuit, (2d)...Current direction detection circuit, (2g) (2m)
...Arithmetic unit, 6)...Inverter

Claims (1)

[Claims] 1. In a power converter control method that performs feedback control of a power converter (5) configured by bridge-connecting a plurality of switching elements and converts direct current to alternating current using a discrete time system, An offset current that makes the current direction equal to the actual current is added to the current value obtained by the method, the direction of the current is determined based on the current value with the offset current added, and dead time compensation is performed based on the current direction. A power converter control method characterized by performing the following. 2. The power converter control method according to claim 1, wherein the offset current is set after a zero crossing is detected based on discretely obtained current values. 3. Current values obtained discretely in a power converter control device that is configured by bridge-connecting multiple switching elements and performs feedback control of a power converter (5) that converts direct current to alternating current using a discrete time system. , current value correction means (2m) that adds an offset current to make the current direction equal to the actual current, and direction determination means (2c) (2d) that determines the direction of the current based on the current value to which the offset current is added. ) and dead time compensation means (2g) for performing dead time compensation based on the direction of current.