JP2008295163A - Pwm inverter device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inverter device that can improve response to a current command even when a carrier frequency of PWM is low. <P>SOLUTION: A PWM inverter device includes a PWM power conversion means, which supplies electric power to a load connected; a current detection means (6), which detects a current flowing in a load (1); a current control means (8), which compares the detected current with a current command and calculates a voltage command which is supplied to the load, and a PWM control means, which sets a command to PWM power conversion according to the voltage command. In this device, a voltage command correction portion (11) is provided which calculates a voltage command correction amount based on a difference between the previous current command value and this time current command value and corrects a command to the PWM control means by adding the correction amount to the voltage command. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a PWM inverter device that controls a current of a load such as a motor.

  A conventional PWM inverter device will be described with reference to the drawings. FIG. 3 is a diagram showing a general configuration of the PWM inverter device. The carrier generator 4 generates a triangular wave based on the set carrier frequency, the comparator 3 compares the modulated wave command with the triangular wave and outputs a switching command to the inverter 2, and the inverter 2 turns on the power element according to the switching command. Turn OFF and supply power to load 1. The current detection start signal generator 5 outputs a current detection start signal to the current detector 6 so as to detect current at the position of the apex of the triangular wave, and the current detector 6 outputs a current based on the current detection start signal. Is detected. By doing so, it is possible to detect a current that is not affected by the switching of the inverter. The modulation wave command setting unit 7 outputs the inputted modulation wave command to the comparator 3 at the timing of the apex of the triangular wave. The current controller 8 compares the detected current with a current command and calculates a voltage command so that the current flowing through the load matches the current command. In general, proportional integral control (PI control) is used for calculation of current control. In addition, voltage FF compensation may be performed based on the current command or the current and load model simultaneously with PI control. The voltage compensator 9 calculates a modulated wave command by compensating for the DC voltage variation of the inverter circuit and the voltage variation in switching so that the voltage complies with the voltage command output by the current controller 8. The current command calculation unit 10 calculates a current that should flow through the load and outputs the current to the current controller 8.

FIG. 4 shows a timing chart of current detection, current calculation, and delivery of a modulated wave command. The current is detected at the peak of the triangular wave of PWM, the current is calculated and the voltage compensation is calculated, the modulation wave is set, and the modulation wave command is updated at the timing of the peak of the triangular wave.
Thus, since the current control period depends on the carrier frequency of PWM, the current control response is also limited by the carrier frequency. On the other hand, the carrier frequency becomes higher due to the characteristics of the power device as the capacity increases. In addition, in order to reduce the influence of external noise or the like, the carrier frequency may have to be lowered. In such an application, there is a problem that a sufficient current control response cannot be obtained.
In order to reduce the overshoot when the current control gain is high with respect to such a problem, a current command one sampling period before is given as a command to the current controller, and the model according to the rate of change of the current command There has been proposed a method for obtaining a voltage command by outputting a voltage in a feedforward direction and adding an output of a current controller and a feedforward amount (for example, see Patent Document 1). In some cases, when a current deviation occurs when the PWM frequency is low, the PWM frequency is changed to temporarily increase the current control response (see, for example, Patent Document 2). In addition, current is detected using a current at the apex position estimated from the detected current, and the dead time is reduced by issuing a voltage command at the apex position. Thus, a method of increasing the current control response has also been proposed (see, for example, Patent Document 3).

  In FIG. 5, a feedforward compensation calculation unit 105 receives a current command Ia *, outputs a current command one sampling period before as a model current Iam, and outputs a model voltage signal Vam. The current controller 107 inputs a deviation between the current Ia of the motor 101 and the model current signal Iam, and outputs a compensation voltage signal Va1. The PWM chopper 103 performs control so that the primary voltage Va of the motor 101 matches the primary voltage command Va * obtained by adding the model voltage signal Vam and the compensation voltage signal Va1. In this way, overshoot can be suppressed even when the gain is increased.

  In FIG. 6, a voltage command v1 * obtained by PI calculation of the current deviation between the current command i1 * and the current detection i1 by the current controller 203 is converted into three phases by the coordinate converter 204, and the comparator 210b outputs from the carrier generator 210a. In the inverter device that drives the motor M1 by controlling the three-phase inverter with PWM compared with the PWM carrier, the PWM carrier is set low, the absolute value of the current deviation is obtained by the absolute value calculator 211a, and the current deviation is obtained by the function unit 211b. The frequency modulation coefficient kTC that is almost inversely proportional to is made, the carrier frequency of the carrier frequency generator 210a is modulated by kTC to increase the carrier frequency according to the current deviation, and the current controller 203 and the integration using the frequency command as a phase command The response time is corrected by changing the integration constant of the multiplier 205 by kTC, and kTC is given to the multiplier 209 to give a voltage indication. A by adding the correction, it is possible to raise the current control response when the current deviation becomes large.

In FIG. 7, the current sample hold timing generation unit 319 instructs the current sample hold unit 312 to sample and hold current at a time position a predetermined time before the positive and negative vertex time positions of the triangular wave. A current control operation is performed by obtaining a current value obtained by adding the compensation by the current compensation unit 313 so that the detected current value is close to the current value at the apex time position of the triangular wave. That is, the dead time can be reduced by generating a PWM pulse pattern at the position of the next vertex after detecting the current before the vertex of the triangular wave and performing the current control calculation. In the current control calculation, the current at the estimated vertex position is used, and ideally the calculation is executed based on the current that contains almost no current ripple component. The dead time can be reduced and the current control response can be improved without increasing the time.
As described above, the conventional inverter device improves the current control response by improving the stability of the current control system and reducing the dead time.
Japanese Patent Laid-Open No. 11-18469 (page 6-14, FIG. 1) JP 2001-37248 A (page 3-5, FIG. 1) Japanese Patent Laying-Open No. 2005-31274 (page 3-5, FIG. 1)

  In the conventional inverter device, in the method using the current command model and the voltage command model output, since the current control gain can be high, the response to the fluctuation of the current detection can be increased, but the reflection of the current command is delayed by one sample, There was a problem that the response to the current command could not be raised. The method of operating the carrier frequency according to the current deviation cannot be used when the carrier frequency is limited by the device or the environment. In addition, when the carrier frequency is low, the current control cycle may not be sufficiently high with respect to the update cycle of the current command, and there is a problem that followability to changes in the current command is reduced. In the method of detecting the current before a predetermined time of the triangular wave, the detected current includes a PWM ripple component, and especially when it is close to the switching timing, a surge current due to switching is also detected, so the current at the apex is estimated. This is difficult, and when the carrier frequency is low, it is difficult to follow the change in the current command.

  The present invention has been made in view of such problems, and an object of the present invention is to provide an inverter device that can increase the response to a current command even when the carrier frequency is low.

In order to solve the above problem, the present invention is configured as follows.
The invention according to claim 1 compares the detected current with a current command by comparing PWM power conversion means for supplying power to a connected load, current detection means for detecting a current flowing through the load, and the current command. In a PWM inverter device comprising current control means for calculating a voltage command to be supplied to a load and PWM control means for setting a command for PWM power conversion in accordance with the voltage command, when the current command is given, A voltage command correction unit is provided that calculates a voltage command correction amount based on the difference between the current command value and the current command value and adds the voltage command correction amount to the voltage command to correct the command to the PWM control means.
According to the second aspect of the present invention, in contrast to the first aspect of the invention, the voltage command correcting unit calculates a voltage change amount based on an electrical model of a load.
Further, the invention according to claim 3 has a function comprising PI control means for calculating by PI control using the difference between the current command values as the voltage command correcting unit. It was added to.
According to a fourth aspect of the present invention, the PI control means according to the third aspect of the present invention is calculated using the output of the current control means, and the integration result of the integrator of the current control means is calculated using the calculation result. The value was corrected.
Further, the invention according to claim 5 is different from the invention according to claim 4 as a proportional gain of the PI control means, based on a proportional gain of the current control means and a ratio of a current control period and a current command calculation period. I wanted it.

According to a sixth aspect of the present invention, in the first aspect, the voltage command correction amounts Δvd and Δvq are obtained by the equation (4) using the differences ΔId and ΔIq between the current command and the previous current command. It is a thing.
According to a seventh aspect of the present invention, in the sixth aspect, the voltage command correction amounts Δvd and Δvq are obtained by an equation (5) obtained by performing PI control on a difference between the previous current command and the current command. It is what.
The invention of claim 8 is the proportional gain K _Pd ′, K of the PI control of the d-axis and q-axis when the sampling period of the current control and the sampling period of the current command calculation unit for updating the current command are different from each other. _Pq ′ is characterized in that it is obtained by equation (6).
Further, in the ninth aspect of the present invention, when the current control period T _I is shorter than the calculation period T _v of the current command calculation unit in the seventh aspect, the integration of the current controller is performed when calculating the equation (5). The value is set as the initial integration value, and the integration value of the current controller is corrected using the integration amount changed after the calculation of Expression (5).

  According to the first aspect of the present invention, when the current command changes, the voltage command correction unit can immediately reflect the change in the PWM command, and the response to the current command can be improved. According to the second aspect of the present invention, since the voltage command correcting unit is operated based on the electrical model of the load, it is possible to correct the PWM command appropriately. According to the third aspect of the present invention, the PWM command correction amount can be adjusted by providing the voltage command correction unit with the PI control means. According to the fourth aspect of the present invention, the current response can be further improved by reflecting the PI control means in the current control means and reflecting it in the current control calculation from the next time. Further, according to the invention described in claim 5, it is possible to appropriately set the proportional gain of the PI control means.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a control block diagram of the inverter device of the present invention. In the figure, 1 is a load, 2 is an inverter, 3 is a comparator, 4 is a carrier generator, 5 is a current detection start signal generator, 6 is a current detector, 7 is a modulation wave command setter, and 8 is a current controller. , 9 is a voltage compensator, 10 is a current command calculation unit, and 11 is a voltage command correction unit.
This embodiment is different from the prior art in that a voltage command correction unit is provided with a current command as an input. The operation will be described below with reference to the block diagram of FIG.
In the figure, a carrier generator 4 generates a triangular wave based on a set carrier frequency, a comparator 3 compares a modulated wave command with the triangular wave and outputs a switching command for the inverter, and the inverter 2 is a power element according to the switching command. Turn on / off and supply power to load 1. The current detection start signal generator 5 outputs a current detection start signal to the current detector 6 so as to detect current at the position of the apex of the triangular wave, and the current detector 6 outputs a current based on the current detection start signal. Is detected. The modulation wave command setting unit 7 outputs the input modulation wave command to the comparator 3 at the timing of the apex of the triangular wave. The current controller 8 compares the detected current with the current command, and calculates a voltage command so that the current flowing through the load matches the current command. In general, proportional integral control (PI control) is used for calculation of current control. In addition, voltage FF compensation may be performed based on the current command or the current and load model simultaneously with PI control. For example, when the load is a three-phase synchronous motor, the following is performed. When a coordinate system is set at the magnetic pole position of the rotor, the magnetic flux component direction of the coordinate system is d-axis, and the torque component direction orthogonal to it is q-axis, the motor model can be expressed as equation (1).

However, Id and Iq are dq axis currents, vd and vq are dq axis voltages, R is a primary resistance, Ld and Lq are dq axis inductances, ω is a rotation speed (electrical angle), and φ is an induced voltage constant.
When controlling the current of such a model, dq-axis current Id and Iq are obtained by detecting at least two phases out of the three phases and coordinate-converting the detected current using the magnetic pole position of the motor. Control by comparing with commands Id * and Iq *. In general, the current is controlled by PI control as shown in Equation (2). The d-axis and q-axis PI control outputs vacrd and vacrq are obtained in (2).

Here, K _Pd and K _Pq are the proportional gains of the d axis and the q axis, respectively, and T _Id and T _Iq are the integration time constants of the d axis and the q axis, respectively. Further, in order to improve the responsiveness, the feed forward is obtained as shown in, for example, Expression (3) based on the model of Expression (1).

This is added to the PI control output of equation (2) to obtain dq axis voltage commands vd ^* , vq ^* . Here, there is a case where a current command is used as a current used for calculation of feedforward. Moreover, the term regarding primary resistance may be omitted. The motor current can be controlled by converting the obtained vd ^* , vq ^* into a three-phase voltage command using the magnetic pole position of the motor and outputting it. However, when the characteristics of the switching elements of the inverter and the DC voltage input to the inverter are different from the assumed voltage, it is not possible to output a voltage as specified by the voltage command. Therefore, it is necessary to compensate for these voltage fluctuations. The voltage compensator 9 calculates a modulated wave command by compensating for the DC voltage variation of the inverter circuit and the voltage variation in switching so that the voltage complies with the voltage command output by the current controller. The main causes of the voltage fluctuation include a voltage drop of the switching element when a current flows, a voltage fluctuation due to a dead time for preventing a short circuit during switching, and a power supply voltage fluctuation, which are compensated for. A modulation wave command is created by performing these compensations, and is output to the modulation wave command setting unit 7.

The current command calculation unit 10 calculates a current to be supplied to the load, and outputs the calculated current command to the current controller 8 and the voltage command correction unit 11. The current command in the case of a motor is a torque command given by a user or another controller, or in the case of speed control, the torque command is calculated so that the actual motor speed matches the command speed.
Based on the difference between the input current command and the previous current command, the voltage command correction unit 11 obtains the amount of change in voltage and corrects the output of the voltage compensator 9. Output to.
When the three-phase synchronous motor is controlled, for example, the voltage change amounts Δvd and Δvq are obtained from the equation (3) as follows.

Here, Id _n, Iq _n is the current d-axis current command of the _{q-axis, Id n-1, Iq n} -1 the previous d-axis using the dq-axis rotating coordinate system that rotates in synchronization with the motor rotor , Q-axis current command.
By adding the obtained voltage change amounts Δvd and Δvq to the previous output of the voltage compensator 8, the voltage command is corrected and output to the modulation wave command setter.

In the calculation of the amount of voltage change, the method according to the first embodiment may not be able to accurately compensate when there is an error between the model and the actual parameter. In order to cope with this, PI control is performed on the difference between the previous current command and the current command. In Equation (5), K _Pd ′ and K _Pq ′ are the proportional gains of the d-axis and q-axis PI control, T _Id and T _Iq are the d-axis and q-axis PI control integration constants, respectively, and s is the Laplace operator. It is.

Furthermore, if the sum of Equation (4) and Equation (5) is used as the amount of voltage change, no excessive integration is performed and the response is improved. Here, it is desirable to set the proportional gains K _Pd ′ and K _Pq ′ of PI control to be equal to the current control gain, but the sampling cycle of the current control and the sampling cycle of the current command calculation unit 10 for updating the current command are different. Therefore, the current control may not be stable depending on the difference between these periods. In order to eliminate this, the proportional gain multiplies the proportional gain of the current controller 8 by the ratio of the current control period TI and the calculation period TV of the current command calculation unit 10;

Can be stabilized.
When the current control cycle is faster, the response can be further improved by linking the integration of the equation (5) with the integration of the current controller 8. That is, when the calculation of equation (5) is performed, the integration value of current controller 8 is used as the initial integration value, and the integration value of current controller 8 is corrected using the integration amount changed after calculation of equation (5). To do.

FIG. 2 shows a timing chart of current command setting, current detection, current calculation, voltage command correction calculation, and modulation wave command delivery in the present invention. The current is detected at the peak of the triangular wave of PWM, the modulation wave is set after the current control calculation and the voltage compensation calculation, and the modulation wave command is updated at the timing of the peak of the triangular wave. FIG. 2 shows an example in which the current command is set in an asynchronous cycle that is later than the current control cycle. In FIG. 2, the current command set at the time t1 is used for the current control calculation started from the time t2 to obtain a voltage command, and the modulation wave is set and the voltage is output at the time t3. . In the worst case, the output reflecting the command is delayed by two cycles of current control. On the other hand, the voltage command correction calculation is calculated immediately after the current command changes, and the modulation wave command is set immediately. It is reflected at time t4 with respect to the command at time t1. That is, since it is set at a timing earlier by one cycle of current control, it is possible to immediately respond to a change in current command.
Recently, a plurality of processors are used for load distribution, and current control and current command calculation may be executed by different processors. In such a case, when the current command is written from the current command calculation unit, an interrupt is generated for the current control processor, and the current control processor is provided with the voltage command correction function of the present invention.

  According to the present invention, by correcting the voltage command at the timing when the current command is updated separately from the current control, the current command can be reflected immediately and the response to the current command can be improved.

Control block diagram of inverter device showing an embodiment of the present invention Time chart of current control in an embodiment of the present invention Control block diagram of conventional inverter device Time chart of current control in conventional inverter device Example 1 of a current control block diagram in a conventional inverter device Example 2 of a current control block diagram in a conventional inverter device Example 3 of current control block diagram in conventional inverter device

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Load 2 Inverter 3 Comparator 4 Carrier generator 5 Current detection start signal generator 6 Current detector 7 Modulation wave command setter 8 Current controller 9 Voltage compensator 10 Current command calculation part 11 Voltage command correction part 101 Electric motor 102 Current Detector 103 PWM chopper 104 Current controller 105 Feedforward compensation calculator 106 Comparator 107 Current controller 108 Adder 110 Current command generator 201 Current command generator 202 Comparator 203 Current controller 203a Proportional calculator 203b Integration calculation Unit 203c Adder 204 Coordinate converter 205 Integrator 206 Coordinate converter 209 Multiplier 210a Carrier generator 210b Comparator 211a Absolute value calculator 211b Function unit 301 Inverter unit 302 AC motor 311 Current detector 312 Current sample hold unit 313 Current compensation 314 current control calculation unit 315 PWM pulse pattern generating section 316 PWM inverter 317 triangular wave generating unit 318 voltage command update timing generator 319 current sample-and-hold timing generator

Claims (9)

PWM power conversion means for supplying power to the connected load, current detection means for detecting current flowing through the load, and comparing the detected current and current command to calculate a voltage command to be supplied to the load In a PWM inverter device comprising current control means and PWM control means for setting a command to PWM power conversion according to the voltage command,
When the current command is given, a voltage command correction amount is obtained based on a difference between the previous current command value and the current current command value, and the voltage control correction amount is added to the voltage command, and the PWM control means A PWM inverter device comprising a voltage command correcting unit that corrects the voltage command to the motor.   2. The PWM inverter device according to claim 1, wherein the voltage command correction unit calculates a voltage change amount based on an electrical model of a load.   3. The PWM inverter device according to claim 1, wherein the voltage command correcting unit includes PI control means for calculating by PI control using a difference between the current command values.   4. The PWM inverter device according to claim 3, wherein the PI control means calculates using an output of the current control means, and corrects an integral value of an integrator of the current control means using a result of the calculation.   5. The PWM inverter device according to claim 4, wherein the proportional gain of the PI control means is determined from a proportional gain of the current control means and a ratio of a current control period and a current command calculation period. 2. The PWM inverter device according to claim 1, wherein the voltage command correction amounts [Delta] vd and [Delta] vq are obtained by the following equations using differences [Delta] Id and [Delta] Iq between the current command and the previous current command.

Here, Id _n, Iq _n is the current d-axis current command of the _{q-axis, Id n-1, Iq n} -1 the previous d-axis using the dq-axis rotating coordinate system that rotates in synchronization with the motor rotor , Q-axis current command, R is primary resistance, Ld and Lq are d-axis inductance, q-axis inductance, and ω is rotational speed, respectively. 7. The PWM inverter device according to claim 6, wherein the voltage command correction amounts [Delta] vd, [Delta] vq are obtained by the following equation in which PI control is performed on the difference between the previous current command and the current command.

Here, K _Pd ′ and K _Pq ′ are proportional gains for PI control of d-axis and q-axis, respectively, T _Id and T _Iq are integration constants for PI control of d-axis and q-axis, respectively, and s is a Laplace operator. . When the sampling period of the current control and the sampling period of the current command calculation unit that updates the current command are different, the proportional gains K _Pd ′ and K _Pq ′ of the PI control of the d-axis and the q-axis are obtained by the following equations: The PWM inverter device according to claim 7.

Here, K _Pd and K _Pq are the proportional gains of the current controller, T _I is the current control period, and T _v is the calculation period of the current command calculation unit. If towards the current control period T _I is shorter than the calculation period T _v of the current calculation unit, at the time of the calculation of the equation (5), the integral value of the current controller as the integral initial value, the formula (5) 8. The PWM inverter device according to claim 7, wherein the integration value of the current controller is corrected using the integration amount changed after the calculation.