Drive control device and drive control method for rotating electric machine

Abstract

In order to provide a PWM control technology that can output an accurate fundamental wave voltage while reducing harmonic current by adding an operation amount to modulation waves of three phases in accordance with the saturation states of the modulation waves of the three phases, this drive control apparatus for a rotating electrical machine drives the rotating electrical machine by controlling an electric power conversion device by means of a PWM signal that is generated by a PWM control unit from a carrier wave and three-phase modulation wave command values generated on the basis of a voltage command value. When the absolute value of a modulation wave command value of one phase, among the three-phase modulation wave command values, is greater than a prescribed value, a modulation wave operation unit, which is provided to the preceding stage to the PWM control unit, stops switching of the one phase by saturating the modulation wave, to the maximum value of the absolute value of the carrier wave, by adding an operation amount to the modulation wave command value of the one phase, and also adds the operation amount to the modulation wave command values of the remaining two phases. Moreover, when the absolute values of modulation wave command values of two phases, among the three-phase modulation wave command values, are greater than the prescribed value, the modulation wave operation unit stops switching of the two phases by saturating the modulation waves, to the maximum value of the absolute value of the carrier wave, by adding the operation amount to the modulation wave command values of the two phases.

Description

[Technical Field]

[0001] The present invention relates to a drive control device and a drive control method for a rotating electrical machine using PWM control. [Background technology]

[0002] Generally, an inverter that converts DC power into AC power is used as a power converter for driving a rotating electrical machine at a variable speed. In an inverter, semiconductor elements perform switching operations using PWM (Pulse Width Modulation) control, converting DC voltage into AC voltage of any frequency and amplitude.

[0003] When a rotating electrical machine is driven by an AC voltage controlled by PWM, the harmonic voltages contained in the voltage waveform of the PWM control cause harmonic currents to flow through the rotating electrical machine, which is known to lead to increased heat generation due to reduced efficiency and increased electromagnetic noise compared to when the machine is driven by a sinusoidal voltage waveform. Therefore, in PWM control, it is desirable to reduce harmonic currents by reducing the distortion of the voltage waveform.

[0004] On the other hand, PWM control also requires accurate voltage output. There are various factors that cause voltage errors in PWM control, such as the influence of dead time errors, but one of these is the influence of restrictions on the minimum pulse width.

[0005] Specifically, when the command value for the pulse width by PWM control becomes extremely narrow and falls below the minimum pulse width (set minimum pulse width) determined by the characteristics of the semiconductor elements and the switching delay characteristics, the pulse width that is actually output is limited to the set minimum pulse width, resulting in an error between the command value and the actual output voltage.

[0006] In particular, the higher the modulation rate and switching frequency of the PWM pulse mode, the greater the likelihood of narrow pulse widths, making this problem more likely to become apparent. Furthermore, if a voltage error occurs, it can lead to issues such as increased torque error and current ripple in the rotating electrical machine. As described above, a technical challenge in PWM control is the need to accurately output voltage while reducing harmonic currents.

[0007] To address the above issues, various improved techniques have been proposed. Patent Document 1 discloses a technology for controlling in accordance with a voltage command value obtained by adding half the intermediate value of the instantaneous voltage command value of each phase to the instantaneous voltage command value of each phase, with the aim of reducing harmonic currents and improving voltage utilization rate.

[0008] Patent Document 2 discloses a technique for the purpose of suppressing occurrence of voltage errors during minimum pulse width limitation and outputting accurate voltages by calculating the pulse width error caused by limiting the pulse width of a phase voltage signal using the error in the line voltage pulse width, and correcting the line voltage pulse width error using the pulse width of a phase voltage signal of a phase other than the phase whose pulse width is limited. At the same time, the technique discloses a technique for correcting the line voltage pulse width error caused by the pulse width correction of the phase voltage signal using the pulse width of a phase voltage signal of a phase other than the two phases in which the minimum pulse width limitation is performed and the pulse width correction is performed. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Patent No. 3233097 [Patent Document 2] Patent No. 6311402 Summary of the Invention [Problem to be solved by the invention]

[0010] The present inventors have conducted extensive research aimed at reducing harmonic currents caused by PWM control while outputting an accurate fundamental voltage, and as a result have come to the following findings. The technology described in Patent Document 1 discloses a technique for reducing harmonic currents by adding a control variable of a modulated wave containing a third-order harmonic component (a voltage waveform with a frequency component three times that of the fundamental voltage) to a three-phase sinusoidal modulated wave to distort the modulated wave. However, when the modulation rate in one-pulse mode (square wave drive) is set to 100%, under ideal conditions ignoring the minimum pulse width, there is a problem in that the output can only reach a maximum of approximately 91%.

[0011] In particular, in applications requiring a high voltage utilization rate, such as railway vehicles, if the output voltage is limited to 91%, the torque shortage caused by the magnetic flux (voltage) is compensated for by increasing the current, and the increase in fundamental current increases losses in the motor and inverter.In addition, in some cases, the maximum output of the rotating electrical machine decreases, making it impossible to obtain the desired torque characteristics.

[0012] To obtain an even higher voltage utilization rate in Patent Document 1, one possible measure would be to increase the modulation rate by eliminating some of the pulses in the PWM control and reducing the number of pulses, creating an overmodulation state in which the peak value of the modulating wave is higher than the carrier wave. However, simply eliminating pulses poses the problem of voltage errors occurring due to the manipulation of the modulating wave to eliminate these pulses.

[0013] Furthermore, while the technology described in Patent Document 2 can suppress voltage errors by ensuring a minimum pulse width, it does not disclose a means for suppressing voltage errors that occur when a pulse is lost. If a pulse is lost, the next pulse must be corrected half a cycle before, but the configuration is not capable of calculating voltage every half cycle. Furthermore, if two phases are subject to the minimum pulse constraint, voltage errors remain, making it impossible to output the correct fundamental wave voltage. As described above, there is room for improvement in reducing harmonic currents in PWM control and in outputting accurate voltages.

[0014] Therefore, the present invention has been made in consideration of the above points, and is based on the technology described in Patent Document 1. Even in the overmodulation region where the modulating wave is made higher than the carrier wave and some PWM pulses are eliminated, the present invention reduces harmonic currents while suppressing voltage errors due to the limitation of the minimum pulse width, thereby outputting an accurate voltage.

[0015] The present invention aims to provide a PWM control technology that outputs an accurate fundamental voltage while reducing harmonic currents by adding a manipulated variable to the modulated wave of each of the three phases according to the saturation state of the modulated wave of each of the three phases. [Means for solving the problem]

[0016] In order to solve the above-mentioned problems, one representative drive control device for a rotating electric machine according to the present invention is a drive control device that drives a rotating electric machine by controlling a power conversion device with a PWM signal generated by a PWM control unit from a three-phase modulated wave command value generated based on a voltage command value and a carrier wave, and has a modulated wave operation unit provided upstream of the PWM control unit, and when the absolute value of the modulated wave command value for one phase of the three-phase modulated wave command values ​​is greater than a predetermined value, the modulated wave operation unit adds a manipulated variable to the modulated wave command value for that phase to saturate it up to the maximum absolute value of the carrier wave, thereby stopping switching of that one phase of the power conversion device, and also adds a manipulated variable to the modulated wave command values ​​for the remaining two phases, and when the absolute values ​​of the modulated wave command values ​​for two phases of the three-phase modulated wave command values ​​are greater than a predetermined value, adds a manipulated variable to the modulated wave command values ​​for those two phases to saturate it up to the maximum absolute value of the carrier wave, thereby stopping switching of the two phases of the power conversion device. [Effects of the Invention]

[0017] According to the present invention, for example, in the overmodulation region, it is possible to reduce harmonic currents and output accurate voltages, thereby improving the efficiency of the drive system and realizing stable control. Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiment of the invention. [Brief explanation of the drawings]

[0018] [Figure 1] 3A and 3B are diagrams illustrating an example of the relationship between a modulated wave and a carrier wave and a PWM voltage waveform according to the present invention. [Figure 2] 1 is a diagram illustrating an example of a driving device for a rotating electric machine to which a PWM control method according to a first embodiment is applied. [Figure 3] FIG. 2 is a diagram illustrating an example of a functional block of the control device according to the first embodiment. [Figure 4] 10 is a diagram showing the relationship between a pulse command value and an output voltage when the pulse width of the pulse command value is sufficiently wide. FIG. [Figure 5] 10 is a diagram showing the relationship between a pulse command value and an output voltage when a pulse command value having a pulse width equal to or less than a set minimum pulse width is output. FIG. [Figure 6] FIG. 10 is a diagram illustrating an example of a calculation method of a zero-phase modulated wave calculation unit. [Figure 7] FIG. 10 is a diagram illustrating a waveform when a zero-phase voltage is added. [Figure 8] FIG. 10 is a diagram showing a waveform obtained when a pulse is eliminated in advance by adding a manipulation amount to a modulated wave. [Figure 9] FIG. 10 is a diagram showing waveforms when a manipulated variable is added to a modulated wave of another phase when a pulse is eliminated in advance by adding a manipulated variable to the modulated wave. [Figure 10] 10 is a flowchart showing an example of a compensation amount calculation performed by a modulated wave error correction unit. FIG. [Figure 11] 10A and 10B are diagrams illustrating another example of the compensation amount calculation performed by the modulated wave error corrector. [Figure 12] This figure shows a waveform in which the carrier wave pulse base has 15 periods and the PWM voltage waveform has 7 synchronous pulses. [Figure 13] This is a diagram showing a waveform in which the carrier wave pulse base has 21 periods and the PWM voltage waveform has 21 synchronous pulses. [Figure 14] This is a diagram showing a waveform in which the carrier wave pulse base is 21 cycles and the PWM voltage waveform is changed from a state of 21 synchronous pulses to a state of 17 synchronous pulses. [Figure 15]FIG. 10 is a diagram showing a waveform in which the carrier wave pulse base is 21 cycles and the PWM voltage waveform changes from a state of 17 synchronous pulses to a state of 13 synchronous pulses (13 synchronous pulses A). [Figure 16] FIG. 16 is a diagram showing a waveform in the case of a modified example (synchronous 13 pulse B) of the synchronous 13 pulse shown in FIG. [Figure 17] FIG. 10 is a diagram showing a comparison of the harmonic current reduction effect of synchronous PWM on a synchronous 21-pulse basis. [Figure 18] FIG. 10 is a diagram showing a comparison of the harmonic current reduction effect of synchronous PWM on a synchronous 15-pulse basis. [Figure 19] This is a diagram showing a waveform in which the carrier wave pulse base is 15 cycles and the PWM voltage waveform is changed from a state of 17 synchronous pulses to a state of 13 synchronous pulses. [Figure 20] This is a diagram showing a waveform in which the carrier wave pulse base is 21 cycles and the PWM voltage waveform is changed from a state of 13 synchronous pulses to a state of 9 synchronous pulses. [Figure 21] This is a diagram showing a waveform in which the carrier wave pulse base is 21 cycles and the PWM voltage waveform is changed from a state of 9 synchronous pulses to a state of 7 synchronous pulses. [Figure 22] This is a diagram showing a waveform in which the carrier wave pulse base is 21 cycles and the PWM voltage waveform is changed from a state of 7 synchronous pulses to a state of 3 synchronous pulses. [Figure 23] FIG. 10 is a diagram illustrating an example of a functional block of a control device used in a second embodiment. [Figure 24] FIG. 10 is a diagram showing the relationship between the modulation rate (command value) of 11 synchronous pulses and the actual modulation rate when the carrier pulse base is 15 cycles. [Figure 25] FIG. 10 is a diagram illustrating an example of the configuration of a modulated wave amplitude correction unit added in the second embodiment. [Figure 26] 10 is a diagram showing the relationship between the modulation rate (command value) and the actual modulation rate in another PWM mode. FIG. [Figure 27] FIG. 10 is a diagram illustrating a concept of an output from an amplitude correction coefficient calculation unit when the modulation rate increases. [Figure 28] FIG. 10 is a diagram illustrating an example of a functional block of a control device used in a third embodiment. [Figure 29]FIG. 11 is a diagram illustrating an example of an outline of a function of a PWM mode switching unit added in the third embodiment. [Figure 30] 10A and 10B are diagrams illustrating modified examples of the functions of a PWM mode switching unit. [Figure 31] FIG. 10 is a diagram illustrating an example of the configuration of a railway vehicle according to a fourth embodiment. [Figure 32] FIG. 10 is a diagram showing an example of harmonic current distribution when driven with synchronous 11 pulses to synchronous 21 pulses according to Examples 1 to 3. DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, as modes for carrying out the present invention, Examples 1 to 4 will be described in detail with reference to the drawings. Here, components with the same reference numbers in each embodiment are shown as components with the same or similar functions. In addition, in later embodiments, explanations of components that are the same or similar to those in previous embodiments may be omitted. [Example]

[0020] FIG. 2 is a diagram illustrating an example of a driving device for a rotating electric machine to which the PWM control method according to the first embodiment is applied. This drive device is composed of a rotating electric machine 4 to be controlled, a drive unit 1 that drives the rotating electric machine 4, a main circuit unit 3 that applies a three-phase AC voltage to the rotating electric machine 4, a gate drive 6 that outputs a gate signal to drive the main circuit unit 3, a control unit 2 that outputs a PWM signal to the gate drive 6, and a current detector 5 that uses the control unit 2 to detect the current flowing through the rotating electric machine 4.

[0021] The control device 2 is equipped with a control program for driving and controlling the rotating electric machine 4 connected as a load. A voltage command value is calculated by vector control or the like so that the current detection value by the current detector 5 becomes the current value as commanded, and a PWM signal is output by pulse width modulation control (hereinafter referred to as "PWM control").

[0022] The main circuit unit 3 is composed of driving transistors such as IGBTs (Insulated Gate Bipolar Transistors) and power devices such as diodes. In Fig. 2, it is shown as six switching elements Sup to Swn, and converts the DC voltage of the smoothing capacitor 7 into AC voltage based on a switching command from the gate drive 6, and applies the three-phase AC voltage to the rotating electric machine 4. The gate drive 6 outputs a gate signal based on the PWM signal.

[0023] A drive current flows through the rotating electric machine 4 when a three-phase AC voltage output from the main circuit unit 3 is applied, and a rotational torque is generated. As the rotating electric machine 4, an induction motor, a synchronous motor, or the like is used.

[0024] The current detector 5 is composed of a Hall CT (Current Transformer) or the like, and detects three-phase current i of U phase, V phase, and W phase flowing through the rotating electric machine 4. u , i v and i w However, the current detector 5 does not necessarily need to detect the currents of all three phases, and may be configured to detect any two of the three phases and determine the waveform of the remaining phase by calculation, assuming that the three-phase currents are in a balanced state.

[0025] 3 shows an example of functional blocks of the control device according to the first embodiment. Here, only the minimum functional blocks necessary for the first embodiment are shown. The modulated wave command calculation unit 10 calculates the three-phase current i detected by the current detector 5. u , i v and i w Based on this, a voltage command value is calculated by vector control (not shown), and a three-phase modulated wave command value a u0 * , a v0 * and a w0 * For the sake of simplicity, the current control unit, frequency calculation unit, phase calculation unit, voltage vector calculation unit, and the like used in general vector control are collectively referred to as a modulated wave command calculation unit 10.

[0026] The PWM control unit 13 calculates the three-phase modulated wave command value a corrected by the zero-phase modulated wave calculation unit 11 and the modulated wave error correction unit 12. u * , a v * and a w * The pulse width modulation (PWM) signal is then output to the gate driver 6 (Fig. 2). Furthermore, the PWM control unit 13 has means for limiting the pulse width, such as a dead time or a minimum on time, in order to prevent breakdown of the switching elements that constitute the main circuit unit 3.

[0027] The pulse width limit will be explained below. FIG. 4 shows the relationship between the pulse command value and the output voltage when the pulse width of the pulse command value is sufficiently wide. Dead time T off A pulse command value is calculated taking this into consideration, and the final output voltage matches the pulse command value. However, during the period when both the upper and lower arms are off, the potential is determined by the polarity of the current, so in Figure 4, the current polarity during the dead time period is assumed to be positive, and the potential during the dead time period is set to zero (marked * next to "Output voltage" at the bottom of Figure 4).

[0028] FIG. 5 shows the relationship between the pulse command value and the output voltage when a pulse command value having a pulse width equal to or less than the set minimum pulse width is output. When the pulse command value is narrower than the minimum pulse width, the dead time T determined by the element characteristics and switching delay characteristics is off and minimum on-time T on The output voltage is limited by the minimum pulse width (=2T off +T on) In other words, the maximum modulation rate that can be output is determined by the minimum pulse width. Note that in Figure 5, the current polarity during the dead time period is assumed to be positive, and the potential during the dead time period is set to zero (marked with an asterisk next to "output voltage" at the bottom of Figure 5).

[0029] As described above, when the pulse width command value in PWM control is narrower than the minimum pulse width (set minimum pulse width), the output pulse width is limited to the set minimum pulse width, and the actual output voltage has an error with respect to the pulse command value. It is also clear that the above voltage error is likely to occur under conditions where the three-phase modulated wave crosses near the peaks and valleys of the carrier wave, that is, in a PWM pulse mode where the carrier wave frequency is high and a high modulation rate is used.

[0030] One of the objectives of the present invention is to reduce distortion in the overmodulation region, so that the switching frequency must be increased even at a high modulation rate, and it is understood that this condition makes the above-mentioned voltage error more likely to occur compared to a pulse mode with a low switching frequency. To address this issue, in the first embodiment, the zero-phase modulation wave calculation unit 11 and the modulation wave error correction unit 12 shown in FIG. 3 are used to reduce harmonic currents and limit the minimum pulse width to reduce distortion and suppress voltage errors.

[0031] The zero-phase modulation wave calculation unit 11 shown in FIG. 3 calculates the three-phase modulation wave command value a u0 * , a v0 * and a w0 * From the above, the zero-phase modulation wave correction amount a mod_0 * Calculate and output the following.

[0032] Fig. 6 is a diagram showing an example of a calculation method of the zero-phase-sequence modulated wave calculation unit 11. Fig. 7 shows an explanation of a waveform when a zero-phase-sequence voltage is added. In FIG. 6, as one of the preferred forms of zero-phase modulation wave calculation, as described in Patent Document 1, the average value of the maximum and minimum values ​​of the three-phase voltage command values ​​is calculated as the zero-phase modulation wave correction amount a mod_0 * A method of adding this to all three phase modulated waves is shown below.

[0033] However, the zero-phase modulation wave correction amount a mod_0 * The calculation method of may be a method of adding a sine wave with a frequency that is 3 times an integer multiple of the fundamental wave to the modulation wave command value of each phase, and a similar effect can be obtained by including a 3×n (n is an integer) harmonic. mod_0 * However, it is not limited to the above.

[0034] In FIG. 7, the zero-phase modulation wave correction amount a including the third harmonic is applied to the three-phase modulation waves (before correction) 200 to 202. mod_0 * By adding 210, the corrected three-phase modulated wave becomes waveforms 300 to 302 in which the peaks of the wave height values ​​of the modulated wave are suppressed compared to when sinusoidal wave modulation is performed, and it is possible to output with a higher voltage utilization rate.

[0035] If the modulation rate of one pulse mode (square wave drive) is set to 100% and the constraint of the minimum pulse width is ignored, the range of modulation rate that can be used without overmodulation is up to approximately 79% for three-phase modulated waves (before correction) 200 to 202, as shown in (Equation 1). On the other hand, as described in Patent Document 1, the zero-phase modulation wave correction amount a mod_0 * When the above is added, the three-phase modulated waves (after correction) 300 to 302 are multiplied by 2 / √3, and can be used up to approximately 91% as shown in (Equation 2).

number

number

[0036] As mentioned above, the zero-phase modulation wave correction amount a mod_0 *When the modulation rate is increased, the pulse width becomes narrower at the point indicated by 500 in the U-phase PWM signal in Fig. 7. As the modulation rate is increased, the pulse width falls below the minimum pulse width explained using Fig. 5, so in reality, the pulse width cannot be increased to 91%.

[0037] In the first embodiment, when a pulse narrower than the minimum pulse width is output in response to the constraint of this minimum pulse width, the occurrence of this voltage error is prevented by adding a manipulation amount to the modulated wave in advance to eliminate the pulse.

[0038] FIG. 8 shows a waveform obtained by adding a manipulation amount to the modulated wave and eliminating the pulse in advance, in comparison with the waveform shown in FIG. In the places where the pulse width becomes narrower than the minimum pulse width, a manipulated variable is added to the three-phase modulated wave to eliminate the pulse, so that the absolute value of the modulated wave becomes larger than the carrier wave only in those places.

[0039] The above operation prevents the output of a command value with a narrow pulse width, and avoids the occurrence of an error between the command value and the actual output voltage. However, by adding a control amount to the modulated wave and eliminating the pulse, the voltage that you want to output (a in Figure 3) u0 ** , a v0 ** and a w0 ** ) will have an error. This error is the UV modulation wave error 600 shown in FIG. 8, which is the difference between the modulation wave (a u0 ** , a v0 ** and a w0 ** ) occurs during the period when the pulse of each phase is lost.

[0040] The mode of eliminating the pulse by this modulation wave manipulation is to increase the modulation wave amplitude by adding a manipulation amount at the point where the pulse becomes narrower than the minimum pulse width, so that the original command value (a shown in Figure 3) u0 ** , a v0** and a w0 ** ), an error will occur in the direction of increasing the final output voltage.

[0041] Therefore, measures were taken to reduce the UV modulation wave error 600 shown in FIG. Figure 9 shows the waveform when a manipulated variable is added to a modulated wave to eliminate the pulse beforehand, and then a manipulated variable is added to the modulated wave of another phase. That is, a manipulated variable is added to a modulated wave to which a zero-phase modulation wave correction amount (Figure 7) has been added, and a pulse is eliminated by adding a manipulated variable (Figure 8), and then a manipulated variable is also added to the modulated wave of another phase. In other words, during the period when a manipulated variable is added to the modulated wave of only one phase to eliminate the pulse, the same manipulated variable is also added to the modulated waves of the remaining two phases.

[0042] For example, the U-phase modulation wave is mod_u * During the period when the pulse disappears after adding the control variable a mod_u * As a result, the phase voltage of the rotating electrical machine 4 changes as shown in (Equation 3), but as shown in (Equation 4) and (Equation 5), the manipulated variable a mod_u * The voltage between the lines (between U and V in this case) applied to the rotating electrical machine 4 before and after the application of the voltage is not affected.

number

number

number

[0043] Generally, the neutral point of the rotating electric machine 4 is not connected, so as long as the line voltage applied to the rotating electric machine 4 does not change, the voltage applied to the rotating electric machine 4 does not change. For the reasons above, during the period when switching is stopped by adding a manipulated variable to the modulated wave of one phase, the UV modulated wave error 600 can be canceled out by adding the same manipulated variable to the modulated waves of the remaining two phases. Here, the explanation has been given using UV, but the same cancellation can also be achieved for VW and WU.

[0044] Comparing the UV modulated wave error 600 shown in Figures 8 and 9, it can be seen that although the effect of reducing the modulated wave error is achieved, in the period when the manipulated variable is added to the modulated waves of certain two phases to eliminate the pulse, a slight error in the modulated wave remains, as shown by the UV modulated wave error 600 shown in Figure 9. This is because the error occurring in the line voltage cannot be offset by manipulating only the remaining phase. However, a method for solving this problem will be described later in Example 2. Here, the operation of the modulated wave error corrector 12 shown in FIG. 3 will be described. FIG. 10 is a flowchart showing an example of the compensation amount calculation performed by the modulated wave error corrector 12. As shown in Fig. 5, by comparing the command value of the pulse width of each phase with the set minimum pulse width, it is determined whether the modulated wave of one phase or two or more phases is saturated, and the manipulated variable a for each of the U phase to W phase is set. mod_u * From a mod_w * In addition, in FIG. 10, the compensation amount a of the U phase is calculated. mod_u * Although only the flowchart for calculating the compensation amount is shown and the illustration for the V-phase and W-phase is omitted, the compensation amount is calculated in the same manner for the V-phase and W-phase. Also, in FIG. 10, the configuration is such that the command value of the pulse width of each phase is successively compared with the set minimum pulse width, but the configuration is not necessarily limited to this as long as a similar compensation method is used. For example, in synchronous PWM, the phase relationship between the modulating wave and the carrier wave is synchronized, so when the modulation rate is increased, the phase at which the pulse width first narrows below the minimum pulse width is always the same (shown as 500 in Figure 7 for the U-phase PWM signal), and the number of phases that saturate at each phase (only one phase, or two or more phases, etc.) is also determined automatically. Therefore, as shown in FIG. 11, the modulated wave error correction unit 12 (distinguished as 12a in FIG. 11) in synchronous PWM may be configured to determine the compensation amount for each phase from the relationship between the voltage phase and modulation rate, without successively comparing the saturation state of each phase.

[0045] FIG. 1 is a diagram showing the relationship between a modulated wave and a carrier wave and an example of a PWM voltage waveform according to the present invention. The zero-phase modulation wave correction amount a calculated by the zero-phase modulation wave calculation unit 11 is added to the three-phase modulation waves 200 to 202 (top row of FIG. 1) of the sine wave before correction. mod_0 * Add 210 (second row in Figure 1).

[0046] Furthermore, to avoid pulse width limitations due to the minimum pulse width setting, when the absolute value of the command value of the modulated wave from the control device 2 becomes larger than the modulated wave value determined from the minimum pulse width setting, manipulated variables 220-222 (the third to fifth rows in FIG. 1) are applied to the modulated wave to eliminate the pulse (periods indicated as A and B in the bottom row of FIG. 1). In this case, during the period in which a manipulated variable is applied to the modulated wave of only one phase to eliminate the pulse, the same manipulated variable is also applied to the modulated waves of the remaining two phases (period indicated as A in the bottom row of FIG. 1). Here, period A in the bottom row of FIG. 1 indicates the period in which the absolute value of the modulated wave of one phase becomes larger than a predetermined value, and period B in the bottom row of FIG. 1 indicates the period in which the absolute values ​​of the modulated waves of two or more phases become larger than a predetermined value. Periods A and B are similarly applied in FIGS. 10, 12-14, and 17-20. Through these operations, corrected three-phase modulated waves 300 to 302 are obtained.

[0047] 1 and 7 to 9, an example was given in which the carrier wave pulse base (peak-trough cycle) is 15 cycles (15PB) per cycle of the voltage waveform, and the PWM voltage waveform changes from a state of 15 synchronous pulses to 11 synchronous pulses by eliminating the pulses. This is just one example, and the same effect can be obtained by applying the same method even if the number of peak-trough cycles of the carrier wave is changed.

[0048] FIG. 12 shows a waveform in which the carrier wave pulse base (peak-valley period) is 15 periods (15PB) per period of the voltage waveform, and the PWM voltage waveform has 7 synchronous pulses. FIG. 13 shows the waveform before the disappearance of a PWM voltage waveform with 21 synchronous pulses, where the carrier pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform. FIG. 14 shows a waveform in which the carrier pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform, and the PWM voltage waveform changes from a synchronous 21 pulse state to a synchronous 17 pulse state by eliminating the pulse. Figure 15 shows a waveform in which the carrier wave pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform, and further pulses are eliminated from Figure 14, changing the PWM voltage waveform from a state of 17 synchronous pulses to a state of 13 synchronous pulses (13 synchronous pulses A).

[0049] FIG. 16 shows a waveform of a modified example (synchronous 13 pulse B) of the synchronous 13 pulse shown in FIG. In this modification, during a period in which pulses of two phases are eliminated, it is not possible to suppress the error in the modulated wave seen between lines by manipulating only the remaining phase, but it is possible to apply any manipulated variable. Therefore, as shown in period C, during a period in which the absolute values ​​of the modulated waves of certain two phases exceed a predetermined value, a manipulated variable is applied to the remaining phase. That is, period C is a period in which the absolute values ​​of the modulated waves of certain two or more phases exceed a predetermined value, and a manipulated variable is applied to the remaining phase. Meanwhile, period B in FIG. 14 is a period in which the absolute values ​​of the modulated waves of certain two or more phases exceed a predetermined value, but no manipulated variable is applied to the remaining phase.

[0050] For example, in the period A during which an manipulated variable is applied at the time of the synchronous 17 pulse shown in FIG. 14, the period during which an manipulated variable is applied to the remaining phase may be continued as in period C, taking into consideration the continuity of the calculation process for mode switching between the synchronous 17 pulse and the synchronous 13 pulse and the characteristics of harmonic currents. Using the above PWM mode as an example, the effects of reducing harmonic currents with and without correction by zero-phase modulation wave and modulation wave compensation to other phases when a pulse is lost will be compared.

[0051] Figure 17 shows a comparison of the harmonic current reduction effect of synchronous PWM on a synchronous 21-pulse basis. The vertical axis in Figure 17 shows the square root of the sum of the squares of the harmonic current of each frequency component in each PWM mode. The smaller the square root of the sum of the squares value, the smaller the total amount of harmonic current in the PWM mode. Similarly, FIG. 18 shows a comparison of the harmonic current reduction effect of synchronous PWM on a synchronous 15-pulse basis.

[0052] <Distortion reduction effect by adding zero-phase modulation wave> In comparison with the characteristics when operating with sinusoidal modulation of synchronous 21 pulses on a synchronous 21 pulse basis in Figure 17(a), the effect of reducing harmonic currents can be obtained by adding a zero-phase component to the sinusoidal modulation wave of synchronous 21 pulses on a synchronous 21 pulse basis in Figure 17(b). In comparison with the characteristics when operating with synchronous 15 pulse sinusoidal modulation on a synchronous 15 pulse base as shown in Figure 18(g), the same effect of reducing harmonic currents can be obtained when a zero-phase component is added to the synchronous 15 pulse sinusoidal modulation wave on a synchronous 15 pulse base as shown in Figure 18(h), and the effect can be obtained regardless of the pulse base, etc.

[0053] <Effect of reducing distortion by controlling the modulation wave amount> In Figure 17, (c) shows the characteristics when the modulated wave is manipulated on a synchronous 21 pulse basis to stop switching and no compensation is added to the modulated wave of other phases when the synchronous 17 pulse occurs, while (d) shows the characteristics when the modulated wave is manipulated on a synchronous 21 pulse basis to stop switching and compensation is added to the modulated wave of other phases when the synchronous 17 pulse occurs. Compared to (c), (d) achieves the effect of reducing harmonic currents at any modulation rate by correcting the modulated wave to suppress voltage distortion.

[0054] The above-mentioned effect can be seen from a comparison between (e) and (f) of Figure 17 (13 synchronous pulses), a comparison between (i) and (j) of Figure 18 (11 synchronous pulses), and a comparison between (k) and (l) of Figure 18 (7 synchronous pulses), that by adding a manipulation amount for the modulating wave, the effect of reducing harmonic current can be obtained regardless of the type of carrier wave or PWM mode.

[0055] For example, when using a sync 15 pulse base, the sync 15 pulse in Figure 18(h) is used at low modulation rates, and when the modulation rate increases and the minimum pulse width is reached, it switches to the sync 11 pulse in Figure 18(j). Then, when the sync 11 pulse reaches the minimum pulse width limit, it switches to the sync 7 pulse in Figure 18(l).

[0056] Furthermore, in order to further increase the modulation rate, Figure 19 shows a waveform in which the carrier pulse base (peak-valley period) is 15 periods (15PB) per period of the voltage waveform, and the PWM voltage waveform is changed from a state of 7 synchronous pulses to a state of 3 synchronous pulses.

[0057] In the synchronous three-pulse system, since the absolute value of the modulated wave of two or more phases is greater than a predetermined value in any period, the control amount is added to the modulated wave and only the process of stopping switching is performed, which prevents the occurrence of voltage errors due to the minimum pulse width.

[0058] As described above, when the carrier wave is based on 15 pulses, the pulse mode is switched in stages in the order of synchronous 15 pulses → synchronous 11 pulses → synchronous 7 pulses → synchronous 3 pulses → synchronous 1 pulse according to the desired voltage utilization rate. FIG. 20 shows a waveform in which the carrier pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform, and the PWM voltage waveform is changed from a state of 13 synchronous pulses to a state of 9 synchronous pulses. FIG. 21 shows a waveform in which the carrier pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform, and the PWM voltage waveform is changed from a synchronous 9 pulse state to a synchronous 7 pulse state. FIG. 22 shows a waveform in which the carrier wave pulse base (peak-valley period) is 21 periods (21PB) per period of the voltage waveform, and the PWM voltage waveform is changed from a state of 7 synchronous pulses to a state of 3 synchronous pulses.

[0059] Next, when the carrier wave is based on 21 pulses, the pulse mode is switched in stages in the order of synchronous 21 pulses → synchronous 17 pulses → synchronous 13 pulses → synchronous 9 pulses → synchronous 7 pulses → synchronous 3 pulses → synchronous 1 pulse according to the desired voltage utilization rate.

[0060] In the 21-pulse mode, in which the number of pulses is less than the synchronous 9 pulses shown in Figure 20, the absolute value of the modulated waves of two or more phases becomes larger than a predetermined value in any period, so only the process of adding an operating amount to the modulated waves to stop switching is performed.

[0061] Furthermore, the timing for eliminating the PWM mode pulse must be switched at least at the timing when it is restricted by the minimum pulse width, but it may be switched before it is restricted by the minimum pulse width in order to reduce the number of times the inverter is switched and suppress the temperature rise (loss) of the main circuit unit 3.

[0062] Furthermore, the PWM mode pattern does not necessarily have to be switched through all of the above modes in order; for example, it is possible to skip a pulse mode in between, such as transitioning from synchronous 21 pulses to synchronous 13 pulses. In the first embodiment, a method of adding a manipulated variable to a three-phase modulated wave has been described, but a similar process may be applied to a three-phase voltage waveform, a PWM pulse waveform, or a carrier wave.

[0063] In the first embodiment, the modulation wave correction by the modulation wave error corrector 12 has been described as a preferred mode in which a zero-phase modulation wave is added to a sinusoidal three-phase modulation wave by the zero-phase modulation wave calculator 11. However, even if the correction is performed on a sinusoidal three-phase modulation wave, it is of course possible to suppress distortion of the voltage waveform, reduce harmonic currents, and improve voltage output accuracy. As described above, the configuration of the first embodiment makes it possible to reduce the voltage error due to the minimum pulse width in the overmodulation region, while also achieving the effect of reducing harmonic currents. [Example]

[0064] The second embodiment differs from the first embodiment in that a modulated wave amplitude correction unit 20 is added to the control device 2. FIG. 23 is a diagram illustrating an example of functional blocks of the control device used in the second embodiment.

[0065] As explained above in the first embodiment using FIG. 9 , in a period in which a manipulated variable is applied to modulated waves of two phases to eliminate the pulses, the error occurring in the line voltage cannot be offset by manipulating only the remaining phase, and a slight error in the modulated wave remains, as shown by UV modulated wave error 600 in FIG. 9 .

[0066] Fig. 24 shows the relationship between the modulation factor (command value) and the actual modulation factor when the carrier wave pulse base is 15 cycles per cycle of the voltage waveform and the synchronous 11 pulses are used. (j) and (i) of Fig. 24 compare the modulation factor (fundamental wave voltage) with and without the modulation wave compensation for other phases as described in the first embodiment, and it can be seen that "with compensation" in (j) improves the voltage output accuracy (the effect described in the first embodiment).

[0067] However, in FIG. 24(j), due to the influence of the UV modulation wave error 600 shown in FIG. 9, the modulation rate (command value) Am * However, the actual modulation factor Am has an error. Therefore, in the second embodiment, the error occurring during the period in which a pulse is lost by adding a manipulated variable to a certain two-phase modulated wave is corrected so that the output is correct when viewed as a whole cycle of the voltage waveform.

[0068] FIG. 25 shows an example of the configuration of the modulated wave amplitude corrector 20 added in the second embodiment. The amplitude correction coefficient calculation unit 21 calculates the PWM mode and modulation rate (command value) Am * The input is the amplitude correction coefficient. *The relationship between the actual modulation factor Am and the actual modulation factor Am can be defined in advance for each PWM mode, as shown in Fig. 24. A correction coefficient is derived to equalize the relationship between the two. This correction coefficient can be calculated using a function that represents the characteristics shown in Fig. 24 or 26, or it can be calculated by directly using the table data shown in Fig. 24 or 26.

[0069] This is because the second embodiment is based on the premise of synchronous PWM in which the carrier wave and modulating wave are synchronized, and therefore the timing of eliminating a pulse and the effect of the eliminating operation on the voltage error are uniquely determined.

[0070] The influence of the UV modulation wave error 600 shown in Figure 9 cannot be corrected at that moment, but when viewed over one entire period of the voltage waveform, the modulation rate (command value) Am * can be corrected so that the actual modulation factor Am matches the

[0071] FIG. 26 shows the relationship between the modulation rate (command value) and the actual modulation rate in other PWM modes. Modulation rate (command value) Am * The relationship between the modulation factor Am and the correction coefficient varies depending on the PWM mode, so the correction coefficient is switched for each PWM mode.

[0072] FIG. 27 shows a conceptual diagram of the output from the amplitude correction coefficient calculation unit 21 when the modulation factor is increased. PWM mode and modulation rate (command value) Am * In response to this, an amplitude correction coefficient with a different slope is output for each PWM mode.

[0073] As described above, the driving device for a rotating electric machine to which the PWM control method according to the second embodiment is applied can correct the voltage error during the period in which the pulse is lost by adding a manipulated variable to a two-phase modulated wave, and output an accurate fundamental wave voltage, by correcting the modulated wave amplitude for each PWM mode using the modulated wave amplitude corrector 20. As a result, the second embodiment can reduce the occurrence of torque error and current ripple more than the first embodiment. [Example]

[0074] The third embodiment differs from the second embodiment in that a PWM mode switching unit 22 is further added to the control device 2. For example, in railway vehicles, overhead line voltage fluctuates greatly depending on the substation's sending voltage, the distance between the vehicle and the substation, and the operating conditions of other vehicles. As a result, the modulation rate also changes significantly when accelerating or decelerating. In addition to starting from a stop, the vehicle must also be restarted from coasting (free-running). Therefore, it is necessary to output voltage accurately at all frequencies and modulation rates, including when the overhead line voltage fluctuates.

[0075] Furthermore, as described in the first embodiment, the range of restrictions imposed by the minimum pulse width expands as the switching frequency increases with a higher modulation rate (for example, when driving at high speeds using multi-pulse synchronous PWM). Therefore, the range in which voltage output is possible with the minimum pulse width also changes depending on the frequency for each PWM mode.

[0076] FIG. 28 is a diagram illustrating an example of functional blocks of the control device used in the third embodiment. The PWM mode switching unit 22 receives the inverter frequency, modulation factor, and minimum pulse width as inputs, and outputs a PWM mode switching signal to the modulated wave amplitude corrector 20 and modulated wave error corrector 12 .

[0077] FIG. 29 shows an example of an outline of the function of the PWM mode switching unit 22 added in the third embodiment. In the third embodiment, the range of PWM modes to be used is defined according to the inverter frequency and modulation rate. For example, in FIG. 29, synchronous PWM-A to synchronous PWM-D are defined according to the inverter frequency and modulation rate. Also, the range of each PWM mode is divided by the "modulation rate output limit in synchronous PWM-X (X: any of A to C)" shown in FIG. 29, but the range of each PWM mode is set so as not to exceed the constraint of the minimum pulse width in each PWM mode. This makes it possible to output voltage accurately at any frequency and modulation rate without being restricted by the minimum pulse width in each PWM mode.

[0078] FIG. 30 shows a modified example of the function of the PWM mode switching unit 22. In the function of this modified example, as in Fig. 29, the range of each PWM mode is divided by "the modulation rate output limit in synchronous PWM-X (X: any of A to C)", but the range of each PWM mode is set so as not to exceed the constraint of the minimum pulse width in each PWM mode. Here, four patterns A to C are shown as an example, but the number of patterns is not limited.

[0079] Furthermore, in order to reduce harmonic losses and noise of the rotating electrical machine 4, the PWM mode may be configured to be driven by multi-pulse synchronous PWM. In this case, the modulation rate is controlled by field weakening control or the like so that it is not restricted by the minimum pulse width.

[0080] As described above, the driving device for a rotating electric machine to which the PWM control method according to the third embodiment is applied can output an accurate fundamental voltage at any frequency or modulation rate by switching the PWM mode according to the range defined by taking into consideration the constraints imposed by the minimum pulse width on the inverter frequency and modulation rate. This enables accurate voltage output even when the overhead line voltage fluctuates or when restarting from coasting (free run). [Example]

[0081] The fourth embodiment is a railway vehicle equipped with the drive device according to any one of the first to third embodiments. In railway vehicles, the resonance frequency of the car body is often around 400 to 600 Hz, and if the harmonic current of the 400 to 600 Hz component flowing through the rotating electric machine 4 is large, it can lead to car body resonance and increase noise, which is a problem.

[0082] In a driving device for a railway vehicle, although it depends on the number of poles of the rotating electric machine 4, it is common for the electrical angular frequency of the driving frequency of the rotating electric machine 4 to switch from asynchronous PWM to synchronous PWM when it is in the range of about 50 to 100 Hz.

[0083] In asynchronous PWM, a current with frequency components around 1 or 2 times the carrier frequency is generated. On the other hand, synchronous PWM uses a voltage waveform synchronized with the drive frequency, and when the current waveform contains large values ​​of the 5th and 7th low-order current harmonics, these become frequency components of approximately 250 to 500 Hz and 350 to 700 Hz, respectively. This creates a problem of increasing noise by matching with the vehicle body resonance frequency.

[0084] FIG. 31 is a diagram illustrating an example of a configuration of a railway vehicle according to the fourth embodiment. The railway vehicle 100 is equipped with a bogie 101, and the bogie 101 is provided with wheels 102. Rotating electric machines 4 are connected to the respective axles of the wheels 102 via gears and couplings (not shown).

[0085] 32 shows an example of the distribution of harmonic currents when driven with 11 to 21 synchronous pulses according to Examples 1 to 3. The horizontal axis represents the order of the harmonic, and the vertical axis represents the current amplitude. Figure 32(a) shows the case of a sync 21 pulse based on a sync 21 pulse, Figure 32(b) shows the case of a sync 17 pulse based on a sync 21 pulse, Figure 32(c) shows the case of a sync 15 pulse based on a sync 15 pulse, and Figure 32(d) shows the case of a sync 11 pulse based on a sync 15 pulse.

[0086] These are examples of PWM modes intended for use when passing through at frequencies of around 50 to 100 Hz. By obtaining a low-distortion PWM voltage waveform according to the present invention, fifth- and seventh-order harmonic currents can be reduced, enabling quieter railcars.

[0087] As described above, in a railway vehicle equipped with a driving device for a rotating electric machine to which the PWM control method according to any one of the first to third embodiments is applied, the effect of reducing noise can be achieved by reducing low-order current harmonics.

[0088] Furthermore, although the above describes Examples 1 to 4 as forms for implementing the present invention, the present invention is not limited to the above-described Examples, and various modifications are possible within the scope of the gist of the present invention. [Explanation of symbols]

[0089] 1... drive device, 2... control device, 3... main circuit section, 4... rotating electric machine, 5... current detector, 6... gate drive, 7... smoothing capacitor, 10... modulated wave command calculation section, 11... zero-phase modulated wave calculation section, 12... modulated wave error correction section, 13... PWM control section, 20... modulated wave amplitude correction section, 21... amplitude correction coefficient calculation section, 22... PWM mode switching section, 100... railway vehicle, 101... bogie, 102... wheel, 200... U-phase modulated wave (before correction) ), 201...V-phase modulated wave (before correction), 202...W-phase modulated wave (before correction), 210...Zero-phase modulated wave correction amount, 220...U-phase modulated wave control amount, 221...V-phase modulated wave control amount, 222...W-phase modulated wave control amount, 230...Carrier wave, 300...U-phase modulated wave (after correction), 301...V-phase modulated wave (after correction), 302...W-phase modulated wave (after correction), 400...U-phase PWM signal, 500...Point where pulse width narrows, 600...UV modulation wave error

Claims

1. A drive control device for a rotating electric machine that drives a rotating electric machine by controlling a power conversion device using a PWM signal generated by a PWM control unit from a three-phase modulated wave command value generated based on a voltage command value and a carrier wave, a modulated wave operation unit is provided in front of the PWM control unit; The modulated wave manipulation unit is When the absolute value of the modulated wave command value of one phase among the three-phase modulated wave command values ​​is larger than a predetermined value, a manipulated variable is added to the modulated wave command value of the one phase to saturate the carrier wave to a maximum value of its absolute value, thereby stopping switching of the one phase of the power conversion device, and the manipulated variable is also added to the modulated wave command values ​​of the remaining two phases, When the absolute values ​​of the modulated wave command values ​​of two phases among the three-phase modulated wave command values ​​are larger than the predetermined value, the manipulated variable is added to the modulated wave command values ​​of the two phases to saturate the absolute value of the carrier wave to a maximum value, thereby stopping switching of the two phases of the power conversion device. A drive control device for a rotating electric machine characterized by:

2. 2. A drive control device for a rotating electric machine according to claim 1, The predetermined value is a value of a modulated wave determined based on a minimum pulse width corresponding to a state in which the power conversion device can output. A drive control device for a rotating electric machine characterized by:

3. 3. A drive control device for a rotating electric machine according to claim 1, The modulated wave operation unit adds, to each of the three-phase modulated wave command values, a zero-phase component including a frequency component that is an integer multiple of three of the fundamental wave frequency of the three-phase modulated wave command value. A drive control device for a rotating electric machine characterized by:

4. 4. A drive control device for a rotating electric machine according to claim 1, a modulated wave amplitude correction unit is further provided in the modulated wave operation unit, The modulated wave amplitude correction unit corrects the modulated wave command value according to a PWM mode in which the power conversion device is operating and a fundamental wave of the three-phase modulated wave command value, and matches the fundamental wave amplitude of the voltage command value with the fundamental wave amplitude of the output voltage of the power conversion device. A drive control device for a rotating electric machine characterized by:

5. 5. A drive control device for a rotating electric machine according to claim 2, The modulated wave operation unit switches to the PWM mode according to the frequency and the modulation rate, using a PWM mode usage range defined based on the frequency for driving the power conversion device, the modulation rate in the PWM control unit, and the minimum pulse width. A drive control device for a rotating electric machine characterized by:

6. A drive control method for a rotating electric machine, which drives a rotating electric machine by PWM-controlling a power conversion device using a PWM signal generated from a three-phase modulated wave command value generated based on a voltage command value and a carrier wave, comprising: When the absolute value of the modulated wave command value of one phase among the three-phase modulated wave command values ​​is larger than a predetermined value, a manipulated variable is added to the modulated wave command value of the one phase to saturate the carrier wave to a maximum value of its absolute value, thereby stopping switching of the one phase of the power conversion device, and the manipulated variable is also added to the modulated wave command values ​​of the remaining two phases, When the absolute values ​​of the modulated wave command values ​​of two phases among the three-phase modulated wave command values ​​are larger than the predetermined value, the manipulated variable is added to the modulated wave command values ​​of the two phases to saturate the absolute value of the carrier wave to a maximum value, thereby stopping switching of the two phases of the power conversion device. A method for controlling the drive of a rotating electric machine, comprising:

7. 7. A method for controlling a rotary electric machine according to claim 6, comprising: The predetermined value is a modulated wave command value determined based on the minimum pulse width corresponding to a state in which the power conversion device can output. A method for controlling the drive of a rotating electric machine, comprising:

8. 8. A drive control method for a rotating electric machine according to claim 6 or 7, comprising: A zero-phase component including a frequency component that is an integer multiple of three of the fundamental wave frequency of each of the three-phase modulated wave command values ​​is added to each of the three-phase modulated wave command values. A method for controlling the drive of a rotating electric machine, comprising:

9. 9. A drive control method for a rotating electric machine according to claim 6, comprising: The modulation wave command value is corrected according to the PWM mode in which the power conversion device is being driven and the fundamental wave of the three-phase modulation wave command value, and the fundamental wave amplitude of the voltage command value and the fundamental wave amplitude of the output voltage of the power conversion device are made to match. A method for controlling the drive of a rotating electric machine, comprising:

10. 10. A drive control method for a rotating electric machine according to claim 7, comprising: Using a PWM mode usage range defined based on a frequency for driving the power conversion device, a modulation rate in the PWM control, and the minimum pulse width, the PWM mode is switched to the PWM mode according to the frequency and the modulation rate. A method for controlling the drive of a rotating electric machine, comprising:

Images