JP4448351B2 - Control device for power converter - Google Patents

Description

  The present invention relates to a power converter control device, and more particularly to a power converter control device capable of accurately correcting an error in a voltage command for commanding an output voltage of a PWM inverter or the like.

  In a PWM inverter (hereinafter referred to as an inverter), the positive side and negative side elements of the inverter are alternately turned on and off, and the output voltage is controlled according to the voltage command by changing the pulse width of the voltage command. At this time, in order to prevent both elements from being turned on at the same time and causing a DC short circuit, in consideration of the turn-off time of the elements, a period in which both elements are turned off when on / off changes is provided. For this reason, output voltage distortion occurs. In the electric motor driven by the inverter, torque ripple occurs due to this output voltage distortion, and the control accuracy of torque and speed deteriorates. In order to suppress the output voltage distortion, there are some which estimate and correct the voltage command error of the inverter by the voltage command component orthogonal to the fundamental wave component vector of the output current and the output current component (variation component other than the fundamental wave). (For example, refer to Patent Document 1).

However, some of the loads driven by the inverter include harmonics in the induced voltage, such as a permanent magnet motor. When the driving power factor is not 1 with such a load, that is, when there is a phase difference between the output voltage (voltage command) and the output current, there is a voltage command component that is orthogonal to the fundamental component vector of the output current. The orthogonal voltage command component and the output current component also include components due to the induced voltage harmonics.
In addition, in the correction of the voltage command error in the conventional inverter control device, the correction value is common to all phases. However, in an actual inverter, there is a characteristic variation for each element, and therefore, the voltage due to the conductive element is Command error may be different.

JP-A-08-126335 (see paragraph numbers 0007 to 0009, (Equation 2), (Equation 3), FIG. 1 and FIG. 2)

  In the correction of the voltage command error in the conventional power converter control device, the voltage command component orthogonal to the fundamental wave component vector of the output current and the detected output current component are used. Therefore, the load is applied to these signals as described above. When an error component associated with the generated induced voltage harmonic is included, the voltage command error cannot be corrected with high accuracy. Further, in the correction of the voltage command error in the conventional power converter control device, the correction value is common and the same value in all phases, and therefore, when the voltage command error for each phase differs due to characteristic variation for each element. The voltage command error cannot be corrected with high accuracy.

  The present invention has been made to solve the above-described problems, and a first object is to accurately correct an output voltage command error even when a load including a harmonic in an induced voltage is driven. It is an object of the present invention to obtain a control device for a power converter that can be performed. In addition, a second object is to obtain a control device for a power converter that can appropriately correct an output voltage command error even when the output voltage command error differs from phase to phase due to, for example, characteristic variations among elements. With the goal.

  The power converter control device according to the present invention includes an output voltage command means for giving an output voltage command value to the power converter, a current detection means for detecting an output current value of the power converter, an output voltage command value and an output current. The direction orthogonal to the induced voltage of the load is obtained based on at least one of the values and the value orthogonal to the induced voltage of the output voltage error of the power converter estimated from the output voltage command value and the output current value is obtained. Output voltage command value correcting means for correcting the output voltage command value based on the direction value and the output current value.

  In the control device for the power converter according to the present invention, the output voltage command means for giving the output voltage command value to the power converter, the current detection means for detecting the output current value of the power converter, the output voltage command value and the output A direction orthogonal to the induced voltage of the load is obtained based on at least one of the current values, and a value orthogonal to the induced voltage of the output voltage error of the power converter estimated from the output voltage command value and the output current value is obtained. Output voltage command value correction means that corrects the output voltage command value based on the value in the orthogonal direction and the output current value is provided, so even when driving a load that includes harmonics in the induced voltage, output is accurate. The voltage command error can be corrected.

Embodiment 1 FIG.
1 to 4 show a control apparatus for a power converter according to Embodiment 1 of the present invention. FIG. 1 is a configuration diagram of the control apparatus, and FIG. 2 is a detailed configuration of a voltage command error feature signal generator. FIG. FIG. 3 is a waveform diagram of signals for explaining the operation, and FIG. 4 is a vector diagram. In this embodiment, the current control system of a conventional permanent magnet motor is improved by adding a correction system for a voltage command error. First, the configuration of the current control system of the permanent magnet motor will be described. In FIG. 1, an angle detector 2 is attached to the rotating shaft of the permanent magnet motor 1, and current control is performed based on the phase angle θ of the permanent magnet magnetic flux of the rotor of the permanent magnet motor 1 output from the angle detector 2. Is called.

  A current flowing through a three-phase winding (not shown) of the permanent magnet motor 1 is detected as an output current by the current detector 3, and current detection values iu, iv, and iw are obtained. This current detection value is converted by the coordinate converter 4 into a d-axis current detection value idd which is a component in the phase angle θ direction and a q-axis current detection value iqd which is an orthogonal component thereof. Deviations between the dq-axis current detection values idd and iqd and the current command values idr and iqr are respectively calculated by the adders 5 and 6, and the d-axis current controller 7 and the q-axis current controller 8 respectively calculate the dq-axis voltage commands vd and vq. Is calculated. The dq axis voltage commands vd and vq are converted into a three-phase voltage command by the coordinate converter 9.

  Next, the configuration of the voltage command error correction system will be described. The phase calculator 10 obtains a current phase difference Δθ with respect to the d axis from the dq axis current detection values idd and iqd. The adder 11 adds the current phase difference Δθ and the rotor phase angle θ to obtain the current phase θi, and the signal generator 12 generates a voltage command error detection signal verrdt synchronized with the current phase θi. On the other hand, the voltage command error feature signal generator 13 generates a characteristic voltage command error feature signal verrrft caused by a voltage command error generated in synchronization with the current from the dq axis current detection values idd, iqd and the dq axis voltage commands vd, vq. Is calculated. The voltage command error feature signal verrrft is multiplied by the voltage command error detection signal verrdt by the multiplier 14 to extract an error signal verr proportional to the voltage command error.

  The error signal verr is proportionally integrated by the calculator 15 to obtain a voltage command error correction signal vcomp. The sign detector 16 detects the sign of each of the current detection values iu, iv, and iw and outputs each current sign. The current sign and the voltage command error correction signal vcomp are multiplied by the multiplier 17 and each phase voltage is detected. The voltage command correction values vucomp, vvcomp, vwcomp corresponding to are calculated. This voltage command correction value is added to the three-phase voltage command output from the coordinate converter 9 by the adder 18 to obtain final voltage commands vu, vv, vw. By this voltage command, the PWM inverter 19 is subjected to switching control to drive the permanent magnet motor 1. The voltage command error feature signal generator 13, the multiplier 14, the arithmetic unit 15, the sign detector 16, the multiplier 17 and the adder 18 are voltage command value correction means in the present invention.

Next, details of the voltage command error feature signal generator 13 will be described. FIG. 2 is a diagram showing a detailed configuration of the voltage command error feature signal generator 13. In FIG. 2, the magnetic flux phase calculator 20 uses the dq-axis current detection values idd and iqd as the armature interlinkage flux d. A phase difference α with respect to the axis is calculated. Specifically, for example, the calculation according to the following equation (1) is performed. However, in the formula (1), Ld is a d-axis inductance, Lq is a q-axis inductance, and Φf is a permanent magnet magnetic flux.
α = arctan (Lq · iqd / (Ld · idd + Φf) (1)
On the other hand, the voltage calculators 21 and 22 calculate voltages Δvd and Δvq generated by the dq-axis current detection values idd and iqd, respectively. Specifically, for example, the calculation according to equation (2) is performed. However, in Formula (2), R is an armature resistance.
Δvd = R · idd + Ld (d (idd) / dt) and Δvq = R · iqd + Ld (d (iqd) / dt) (2)

  The adders 23 and 24 calculate the corrected voltages vde and vqe by subtracting the current divided voltages Δvd and Δvq from the dq axis voltage commands vd and vq. The coordinate converter 25 calculates a component in the armature flux linkage direction from the corrected voltages vde and vqe using the armature flux linkage phase difference α. The adder 26 subtracts the armature flux linkage phase differences α and π / 2 from the phase difference Δθ to obtain a phase difference β between the armature flux linkage vector Φ and the current vector I (see FIG. 4 described later). The computing unit 27 calculates the reciprocal of the cosine value of the phase difference β (1 / cos β), and the multiplier 28 calculates the reciprocal of the cosine value of the phase difference β and the armature flux linkage component of the corrected voltages vde and vqe. Multiplication is performed to generate a voltage command error feature signal verrft which is an output of the voltage command error feature signal generator 13.

  Here, the state of the signal in detecting the voltage command error will be described with reference to FIGS. For example, when a current vector I as shown in FIG. 4 is given as a current command value and current control is ideally performed, the waveform of the U-phase current is expressed as shown in FIG. FIG. 3B shows the U-phase voltage command error generated by the PWM inverter short-circuit prevention time. The waveforms shown in FIGS. 3C and 3D are obtained by converting the U-phase voltage command error into the direction component of the current vector I and the orthogonal direction component thereof. The voltage command errors for the V phase and the W phase are obtained by shifting the phases of FIG. 3B by 120 ° and 240 °, respectively. (E) The signal has a waveform as shown in (f).

  That is, the voltage command error component in the current vector direction is generally a DC component, and the voltage command error component in the current vector orthogonal direction is a sawtooth wave signal having a period of 6 electrical angles, and the phase of the signal suddenly changes from positive to negative. However, it turns out that it corresponds with the phase in which each phase current becomes 0. FIG. 4 shows a voltage command error component in the current vector direction as a vector verr0, and a voltage command error component in the current vector orthogonal direction as a vector verr1. When current control is ideally performed, the voltage command includes an inverse signal that compensates for the voltage command error.

  On the other hand, in the permanent magnet motor, the magnetic flux generated by the permanent magnet includes harmonic components other than the fundamental wave. In addition, the armature reaction magnetic flux generated by the motor current may include a harmonic component due to the structure of the stator. For this reason, harmonic components are also included in the induced voltage due to these magnetic fluxes, and in many cases, components with a period of 6 times the electrical angle are included. This induced voltage is generated in a direction orthogonal to the armature linkage magnetic flux vector Φ, which is a combination of the permanent magnet magnetic flux Φm and the armature reaction magnetic flux Φa. The harmonic component contained in this induced voltage is shown in FIG. This is indicated by the vector vbemf.

  At this time, the d-axis and the armature flux linkage vector Φ have the phase difference α represented by the above-described equation (1). When current control is ideally performed, the voltage command also includes an inverse signal that compensates for this induced voltage harmonic component. Finally, the voltage command includes the voltage command error described above and the above-described voltage command error. An inverse signal that compensates for two of the induced voltage harmonic components will be included. However, as is apparent from FIG. 4, the voltage command in the direction orthogonal to the induced voltage (voltage vector vbemf) due to the magnetic flux, that is, in the same direction as the armature linkage magnetic flux vector Φ, does not include the component due to the induced voltage harmonics. Only the component due to the voltage command error is included. Further, in this case, a component obtained by observing the voltage command error component vector verr1 in the direction orthogonal to the current vector in the armature linkage magnetic flux vector direction is (Δθ−) between the voltage command error component vector verr1 and the armature linkage flux vector φ. Since there is a phase difference of [alpha]-[pi] / 2), it can be seen that the amplitude becomes a value multiplied by cos ([Delta] [theta]-[alpha]-[pi] / 2), and decreases accordingly.

  In the voltage command error correction system according to the present embodiment, as shown in FIG. 2, the voltage calculators 21 and 22 and the adders 23 and 24 cause the current divided voltages Δvd and Δvq from the dq axis voltage commands vd and vq. Are respectively corrected to calculate corrected voltages vde and vqe. The corrected voltages vde and vqe are the sum of the voltage command error component and the induced voltage. On the other hand, the magnetic flux phase calculator 20 calculates the phase difference α between the armature linkage magnetic flux vector (Φ in FIG. 4) and the d axis, and the coordinate converter 25 corrects the armature linkage flux of the voltages vde and vqe. Find the component in the same direction as the vector Φ.

  As described above, the component does not include a component due to induced voltage harmonics, but is multiplied by cos (Δθ−α−π / 2) as compared with the voltage command error component vector verr1 in the direction orthogonal to the current vector I. Amplitude decreases by that amount. For this reason, after obtaining (Δθ−α−π / 2) by the adder 26, the arithmetic unit 27 and the multiplier 28 multiply the output of the coordinate converter 25 by the reciprocal of cos (Δθ−α−π / 2). Thus, a sawtooth wave signal having the same amplitude as the voltage command error component vector verr1 in the direction orthogonal to the current vector I and having a period of 6 times the electrical angle is obtained. As a result of the above processing, a voltage command error feature signal verrrft as shown in FIG. 4G is obtained. At this time, the DC voltage component of the voltage command error component in the induced voltage and current vector directions is added to the voltage command error feature signal verrrft.

  As can be seen from FIG. 3, the amplitude of the sawtooth wave included in the voltage command error feature signal verrrft in FIG. 3G and the magnitude of the voltage command error in FIG. For this purpose, the magnitude of the rectangular wave voltage to be corrected may be adjusted while observing the amplitude of the sawtooth wave component. Various methods can be considered for extracting the amplitude of the sawtooth wave of the voltage command error feature signal verrrft in FIG. 3G as the error signal verr. For example, as shown in FIG. If the voltage command error detection signal verrdt (see FIG. 3 (h)) which is an AC signal synchronized with a) the U-phase current is multiplied, an error signal verr corresponding to the amplitude of the sawtooth wave is obtained on average. . The error signal verr is a value proportional to the value in the direction orthogonal to the load induced voltage of the voltage command error, that is, the value in the armature flux linkage direction.

  Therefore, the current phase θi is obtained by the phase calculator 10 and the adder 11, and the voltage command error detection signal verrdt synchronized with this phase is generated from the signal generator 12. The voltage command error detection signal verrdt in this case may be a signal whose sign changes in synchronization with the voltage command error feature signal verrrft. For example, as shown in FIG. Using a sawtooth wave with a constant amplitude and having no direct current component is effective because the characteristic component of the voltage command error feature signal verrrft can be detected more accurately. As shown in FIG. 3 (i), an error signal having a correlation with an error component of a target phase is extracted by using the signal waveform of the opposite sign of FIG. 3 (d) which is an error signal for one phase. However, since the voltage command error is generally the same in each phase, an error signal verr obtained by multiplying the voltage command error detection signal verrdt in FIG. 3 (i) may be used.

Embodiment 2. FIG.
5 and 6 show a second embodiment of the present invention. FIG. 5 is a block diagram of the control device for the power converter, and FIG. 6 is a waveform diagram for explaining the operation. In the first embodiment, the voltage command error of each phase is corrected with the same correction value by the same correction system. However, in this second embodiment, independent correction values are obtained by individual correction systems. It is corrected by. In FIG. 5, the current phase θi is calculated by the phase calculator 10 and the adder 11. The signal generators 12a, 12b, and 12c generate voltage command error detection signals verrdtu, verrdtv, and verrdtw that are synchronized with the phase currents calculated from the current phase θi with respect to the calculated current phase θi. The relationship between each phase current iu, iv, iw and each phase voltage command error detection signal verrdtu, verrdtv, verrdtw is shown in FIGS. The voltage command error detection signals verrdtu, verrdtv, and verrdtw may be any signal whose sign changes in synchronization with the error signal for each phase as shown in FIG. Use of the inverse sign waveform of the error signal of the phase is advantageous because an error signal having a correlation with the error component of the target phase can be extracted particularly well.

  On the other hand, the voltage command error feature signal generator 13 calculates the voltage command error feature signal verrrft in the same manner as in the first embodiment. The voltage command error feature signal verrrft is calculated by the multipliers 14a, 14b, and 14c. The error detection signals verrdtu, verrdtv, and verrdtw are respectively multiplied, and error signals verru, verrv, verrw proportional to the voltage command error of each phase are output. The error signals verru, verrv, and verrw are proportionally integrated by the calculators 15a, 15b, and 15c, respectively, to obtain voltage command error correction signals vcompu, vcompv, and vcompw.

  The sign detectors 16a, 16b, and 16c detect the signs of the current detection values iu, iv, and iw, and output the current signs. The current signs are multiplied by the voltage command error correction signals vcompu, vcompv, and vcompw. The voltage command correction values vucomp, vvcomp, vwcomp corresponding to each phase voltage command are calculated by the multipliers 17a, 17b, 17c. This voltage command correction value is added to the three-phase voltage command output from the coordinate converter 9 by the adder 18 to obtain final voltage commands vu, vv, vw. In the control device for the power converter according to the present embodiment, the voltage command error of each phase is individually corrected. Therefore, even when the voltage command error in each phase varies, the error can be accurately corrected. The present embodiment does not depend on the configuration of the voltage command error feature signal generator 13 described in the first embodiment, and can also be applied to a voltage command error correction method in a conventional power converter. .

  Furthermore, this method can be extended to adjust the voltage command correction value for each phase to a different value depending on the sign of the current. For example, by preparing six signal generators corresponding to the signs of the respective phase current values and using six multipliers and arithmetic units corresponding thereto, voltage command error correction signals respectively corresponding to the positive and negative of each phase can be obtained. If the voltage command correction value for each phase is calculated by the sign detector and the multiplier, the voltage command error for each phase can be adjusted to a different value depending on the sign of the current.

  In the control device for the power converter according to the present invention, the output voltage command means for giving the output voltage command value to the power converter, the current detection means for detecting the output current value of the power converter, the output voltage command value and the output A direction orthogonal to the induced voltage of the load is obtained based on at least one of the current values, and a value orthogonal to the induced voltage of the output voltage error of the power converter estimated from the output voltage command value and the output current value is obtained. Since the output voltage command value correcting means for correcting the output voltage command value based on the value in the orthogonal direction and the output current value is provided, even when driving a load including harmonics in the induced voltage, the voltage can be accurately detected. The command error can be corrected.

  And in the control apparatus of the power converter which concerns on this invention, each phase output voltage command means which gives an output voltage command value for every phase to a power converter, and the current detection means which detects the output current value of a power converter And each phase output voltage command value correcting means for correcting the output voltage command value for each phase based on the output voltage command value and the output current value. Even when the voltage command error is different, the output voltage command error can be appropriately corrected.

  Furthermore, the output voltage command value correcting means corrects the output voltage command value for each phase. Therefore, even when the output voltage command error for each phase is different, the output voltage command error is appropriately adjusted. Can be corrected.

It is a block diagram which shows the structure of the control apparatus of the power converter which is Embodiment 1 of this invention. It is a block diagram which shows the detailed structure of the voltage command error characteristic signal generator of FIG. It is a wave form diagram of a signal for explaining operation of a control device of a power converter of Drawing 1. It is a vector diagram for demonstrating operation | movement of the control apparatus of the power converter of FIG. It is a block diagram which shows the structure of the control apparatus of the power converter which is Embodiment 2 of this invention. It is a vector diagram for demonstrating operation | movement of the control apparatus of the power converter of FIG.

1 permanent magnet motor, 3 current detector, 7, 8 d-axis and q-axis current controller,
13 voltage command error feature signal generator, 14, 14a, 14b, 14c multiplier,
15, 15a, 15b, 15c calculator, 16, 16a, 16b, 16c code detector,
17 multiplier, 18 adder, 19 PWM inverter.

Claims (2)

An output voltage command means for giving an output voltage command value to the power converter, a current detection means for detecting an output current value of the power converter, and a load based on at least one of the output voltage command value and the output current value A direction orthogonal to the induced voltage of the power converter and a value in the direction orthogonal to the induced voltage of the output voltage error of the power converter estimated from the output voltage command value and the output current value are obtained. A control device for a power converter, comprising: output voltage command value correcting means for correcting the output voltage command value based on the output current value. 2. The control device for a power converter according to claim 1, wherein the output voltage command value correcting means corrects the output voltage command value for each phase .

Images