ENERGY CONVERSION DEVICE

Abstract

An energy conversion device includes an inverter with multiple switching elements, wherein the inverter receives a DC voltage from a DC voltage source and converts the DC voltage into an AC voltage with variable voltage and variable frequency to output the AC voltage to a load, a control unit that controls an on / off control of the multiple switching elements via PWM control, and a series body consisting of a capacitor on the positive electrode side and a capacitor on the negative electrode side, which is connected on an input side of the inverter between a positive electrode and a negative electrode of the DC voltage source.The output potential of the inverter has at least one potential of the positive electrode and one potential of the negative electrode of the DC voltage source and one potential of a neutral point, which is a connection point between the capacitor on the positive electrode side and the capacitor on the negative electrode side.The control unit contains a modulation factor calculation unit, which calculates a modulation factor of the inverter based on the DC voltage and an output voltage command value; a gate signal generator, which generates a gate signal for on / off control of the switching elements to generate a pulse train based on a comparison between the calculated modulation factor and a carrier signal; and a gate signal assigner, which adjusts a gate signal assignment so that a voltage of the capacitor on the positive electrode side and a voltage of the capacitor on the negative electrode side are balanced.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a power conversion device and, more particularly, to a multilevel inverter. STATE OF THE ART

[0002] A multilevel inverter not only outputs a zero voltage, but also a voltage in at least one stage on both the positive and negative electrode sides. Pulses at each stage are generally generated using PWM (Pulse Width Modulation) control, known as triangular wave PWM.

[0003] For example, a multilevel inverter is configured with multiple power conversion units, each containing an inverter circuit capable of providing three-stage outputs, and an independent DC or DC power supply connected in series. Pulses at each stage are output by individual units, and a combination of pulse voltages output by the units is output by the multilevel inverter.

[0004] In this regard, in switching control of the multilevel inverter, when generating a pulse pattern for each stage to be output from each power conversion unit, a phase of a triangular wave to be compared with a command value is shifted for each stage so that a composite pulse has a staircase pattern, and further, a power conversion unit to which pulses are assigned is periodically changed so that losses between the power conversion units due to imbalance in switching by switching elements do not vary (see Patent Document 1).

[0005] Some multilevel inverters are neutral-clamped multilevel inverters. Such a multilevel inverter can output voltages in at least three levels. However, when the voltage or frequency varies, the switching time between the positive and negative electrode sides becomes unbalanced in a triangular wave comparator circuit. Depending on the power factor, the charging and discharging amounts of the capacitor voltage on the positive electrode side and the capacitor voltage on the negative electrode side become unbalanced in a DC bus circuit. A voltage difference between the positive and negative electrode sides may increase, resulting in an overvoltage of the capacitor voltage on either the positive or negative electrode sides, or an increased current fluctuation (torque ripple) during load operation.

[0006] In this regard, for example, in the case of a three-level inverter, a three-phase voltage command value to be compared with a triangular wave is appropriately corrected based on a capacitor voltage difference between the positive electrode side and the negative electrode side, a sign of a voltage command, or a current to change a pulse pattern after the comparison with the triangular wave, thereby preventing an increase in the capacitor voltage difference between the positive electrode side and the negative electrode side in the DC bus circuit during operation of the load (see Patent Document 2). BIBLIOGRAPHYPATENT LITERATURE Patent document 1: JP 4 352 787 B2 Patent document 2: JP 6 707 298 B2 SUMMARY OF THE INVENTION TECHNICAL TASK

[0007] In contrast, the switching control according to Patent Document 1 achieves uniform switching loss among the power conversion units. However, this document is silent regarding the purpose or effect of maintaining balanced capacitor voltages between the positive electrode side and the negative electrode side in a series circuit in each unit in a multilevel or at least 5-level inverter.

[0008] A method for eliminating the capacitor voltage difference between the positive electrode side and the negative electrode side of a DC bus circuit according to Patent Document 2 is applied to a neutral point clamped 3-level inverter.

[0009] Since an inverter that outputs voltages in steps (at least five steps) that exceed three steps has a different circuit configuration, the method is not applicable.

[0010] For example, for such a multilevel inverter, which includes an inverter circuit capable of outputting voltages in five steps, since it includes a circuit for each phase configured to connect a DC bus circuit and two three-step switching circuits, there are multiple switching combinations for outputting the same voltage, and the charging and discharging directions of capacitors on the positive electrode side and the negative electrode side vary depending on the combinations. Therefore, a method for changing a command value based on triangular wave comparison according to Patent Document 2 cannot adequately suppress the voltage difference between the two capacitors.

[0011] The present disclosure has been created to achieve the above object, and it is an object of the present disclosure to provide a power conversion device that suppresses an increase in an imbalance between a voltage of a capacitor on the positive electrode side and a voltage of a capacitor on the negative electrode side in a DC bus circuit depending on a change in voltage or frequency or a power factor, and prevents an occurrence of an overvoltage phenomenon of a capacitor voltage due to an increase in the imbalance or a phenomenon of an increase in torque ripple due to an asymmetry between a positive output voltage and a negative output voltage. SOLUTION TO THE TASK

[0012] A power conversion device according to an embodiment includes an inverter having a plurality of switching elements, the inverter receiving a DC voltage from a DC voltage source and converting the DC voltage into an AC voltage with a variable voltage and a variable frequency to output the AC voltage to a load, a control unit that controls on / off driving of the plurality of switching elements under PWM control, and a series body of a positive electrode side capacitor and a negative electrode side capacitor connected between a positive electrode and a negative electrode of the DC voltage source on an input side of the inverter.An output potential of the inverter has at least a positive electrode potential and a negative electrode potential of the DC power source and a potential of a neutral point which is a connection point between the capacitor on the positive electrode side and the capacitor on the negative electrode side.The control unit includes a modulation factor calculation unit that calculates a modulation factor of the inverter based on the DC voltage and an output voltage command value, a gate signal generator that generates a gate signal for driving on / off the switching elements to generate a pulse train based on a comparison between the calculated modulation factor and a carrier signal, and a gate signal allocator that adjusts an allocation of the gate signal so that a voltage of the capacitor on the positive electrode side and a voltage of the capacitor on the negative electrode side are balanced. ADVANTAGEOUS EFFECTS OF THE INVENTION

[0013] A power conversion device of the present disclosure can suppress an increase in imbalance between a voltage of a capacitor on the positive electrode side and a voltage of a capacitor on the negative electrode side in a DC bus circuit depending on a change in voltage or frequency or a power factor, and prevent an occurrence of a phenomenon of overvoltage of a capacitor voltage due to an increase in imbalance or a phenomenon of an increase in torque ripple due to asymmetry between a positive output voltage and a negative output voltage. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows a circuit diagram illustrating an overall configuration of a power conversion device 2 according to a first embodiment. Fig. 2 is a diagram illustrating a hardware configuration of the power conversion device 2 according to the first embodiment. Fig. 3 is a diagram showing a configuration of a portion corresponding to only one phase (U-phase) of an inverter 4 according to the first embodiment. Fig. 4 is a diagram illustrating a relationship between switching arms 8a and 8b and a step of an output voltage of a single-phase circuit of the inverter 4 according to the first embodiment. Fig. 5 is a diagram illustrating an operation of a carrier comparison PWM generator 12 as a comparison example. Fig. 6 shows a figure illustrating a value of a neutral point current Ic and an increase and decrease of the capacitor voltage difference V diff on one side of positive electrode in each pattern. Fig. 7 is a diagram illustrating gate pulse signals GA# and GB# according to the first embodiment. Fig. 8 is a circuit diagram showing an overall configuration of a power conversion device 2# according to a second embodiment. Fig. 9 is a diagram illustrating an operation of a neutral point potential switching control unit 15 according to the second embodiment. Fig. 10 is a flowchart showing the operation of the neutral point potential switching control unit 15 according to the second embodiment. DESCRIPTION OF EMBODIMENTS

[0014] An embodiment is described below with reference to the drawings. In the following description, the same elements are assigned the same reference numerals. Since their names and functions are also identical, their detailed description will not be repeated. First embodiment.

[0015] Fig. 1 shows a circuit diagram illustrating an overall configuration of a power conversion device 2 according to a first embodiment. The power conversion device 2 includes, as shown in Fig. 1, an inverter 4 and a control unit 10 which controls the inverter 4, and converts DC voltages from DC voltage sources 1a, 1b and 1c of the U, V and W phases into AC voltages of variable voltage and variable frequency to output the AC voltages to a motor 3 which is a load.

[0016] Inverter 4 implements a 5-level inverter in which two switching branches 8a and 8b, each forming a neutral-clamped 3-level inverter, are connected in parallel for each phase. The neutral-clamped 3-level inverter includes two series-connected capacitors, i.e., a capacitor 5a on the positive electrode side and a capacitor 5b on the negative electrode side, which divide a DC voltage from DC voltage source 1a, a plurality of switching elements 6 implemented by IGBTs or the like with respective antiparallel-connected diodes, and a clamping diode 7.

[0017] As described above, a circuit pattern or the like is described below in connection with a configuration of the 5-level inverter in which two switching arms 8a and 8b, each constituting the 3-level inverter, are connected in parallel for each phase. However, in view of the application of the present disclosure, it is not intended to be limited to the 5-level inverter, and a configuration for outputting voltages in at least nine levels in which inverter circuits in multiple levels are connected in series for each phase is equally applicable.

[0018] The inverter 4 converts DC voltages from the DC voltage sources 1a to 1c into AC voltages of any magnitude and frequency by controlling the switching elements 6 via PWM (Pulse Width Modulation) control to output the AC voltages. The inverter 4 further includes, in a connection section to the motor 3, a current sensor 18 as a load current detector that detects a current of the motor 3, which is the load currents iu, iv, and iw, and neutral point voltage sensors 20a, 20b, and 20c that detect, as a neutral point voltage, a voltage difference between the capacitor 5a on the positive electrode side and the capacitor 5b on the negative electrode side, which divide the voltages of the DC voltage sources 1a to 1c of the inverter 4.

[0019] The control unit 10 is composed of a U / f controller 19, a modulation factor calculation unit 11, a carrier comparison PWM generator 12 and a gate signal allocator 13, wherein the gate signal allocator 13 contains a pulse distributor 14.

[0020] Each of these components is described below.

[0021] The V / f controller 19 is a controller that outputs a voltage command value Vp by multiplying an input frequency command Fc by a prescribed coefficient. Generally, a coefficient obtained by dividing a rated voltage of the motor by a rated frequency is used.

[0022] The modulation factor calculation unit 11 calculates a modulation factor command m in accordance with expression (1) based on a DC voltage Vdc from the DC power sources 1a to 1c and an output voltage command value (corresponding to a line voltage) Vp of the inverter 4, and outputs the modulation factor command. [Expression 1] m=23∗VpVdc

[0023] The carrier comparison PWM generator 12 compares the modulation factor command m output from the modulation factor calculation unit 11 with a triangular wave (a carrier signal) in prescribed frequency cycles and generates and outputs two sets of 3-stage gate pulse signals output from two switching arms.

[0024] The pulse distributor 14 of the gate signal allocator 13 determines to which switching elements in the two switching branches 8a and 8b the two sets of generated gate pulse signals are to be assigned, and in accordance with the determination, the pulse distributor 14 outputs a gate pulse signal 17 to the switching elements in each of the switching branches 8a and 8b.

[0025] Fig. Fig. 2 is a diagram illustrating a hardware configuration of the power conversion device 2 according to the first embodiment. As shown in Fig. 2, the inverter 4 converts DC voltages from the DC voltage sources 1a, 1b and 1c of phases U, V and W into AC voltages of variable voltage and variable frequency to output the AC voltages to the motor 3, which is the load.

[0026] The control unit 10 includes a processor 301 and a storage device 302. A program for the control unit 10 is stored in advance in the storage device 302. The processor 301 executes a functional program stored in the storage device 302. When the functional program is executed by the processor 301, various functions are performed.

[0027] The control unit 10 contains the V / f controller 19, the modulation factor calculation unit 11, the carrier comparison PWM generator 12, and the gate signal allocator 13. The gate signal allocator 13 contains the pulse distributor 14.

[0028] A detector 29 is composed of a sensor group that detects a voltage and a current from each component of the inverter 4. In the present example, for example, the detector 29 includes the current sensor 18 and the neutral point voltage sensors 20a to 20c.

[0029] The processor 301 generates a gate pulse signal 17 which drives the switching elements 6 in the inverter 4 to turn them on and off by processing information from the detector 29. The companies are described below.

[0030] The operations of the inverter 4 based on the on / off control of the switching elements 6 in the two switching branches 8a and 8b are described below.

[0031] Fig. 3 is a diagram illustrating a configuration of a portion corresponding to only one phase (U-phase) of the inverter 4 according to the first embodiment. As shown in Fig. 3, 3-stage output voltages V A and V Bfrom the respective switching branches 8a and 8b. The switching branch 8a is connected to a phase output terminal of the inverter, and the switching branch 8b is connected to a neutral point of the three phases. Therefore, a relationship between an output voltage V INV a single-phase inverter, an output voltage V A of the switching branch 8a and an output voltage V B of the switching branch 8b is shown in the following expression (2). [Expression 2] VINV=VA−VB

[0032] A capacitor voltage difference V diff is shown in a subsequent expression (3), where V PC represents a voltage of the capacitor 5a on the positive electrode side in a DC bus circuit and V CN represents a voltage across capacitor 5b on the negative electrode side of the DC bus circuit. [Expression 3] Vdiff=VPC−VCN

[0033] Fig. 4 is a diagram illustrating a relationship between the switching arms 8a and 8b and a step of an output voltage of a single-phase circuit of the inverter 4 according to the first embodiment.

[0034] In Fig. 4, “E” for an output voltage represents a unit of voltage per level output by the multilevel inverter.

[0035] An output stage or output stage of the output voltage V A of the switching branch 8a is reproduced unchanged at the output of the inverter 4, while the output stage of the output voltage V B of the switching branch 8b with reversed positive and negative values.

[0036] For example, when a voltage of +1E is output from the inverter 4 through the switching branch 8b, the switching elements 31 and 32 on one side of the positive electrode and one side of the negative electrode of the switching branch 8a are blocked (turned off) to obtain the output voltage V A to 0, and the switching element 33 on the positive electrode side of the switching branch 8b is blocked and the switching element 34 on the negative electrode side is turned on to produce -1E as the output voltage V B to spend. The following describes how the gate pulse signal is assigned.

[0037] The carrier comparison PWM generator 12 generates two sets of 3-stage gate pulse signals.

[0038] The gate signal allocator 13 with the pulse distributor 14 assigns the gate pulse signal 17 to one of the two switching branches 8a and 8b and outputs the gate pulse signal.

[0039] Fig. 5 is a diagram illustrating an operation of the carrier comparison PWM generator 12 as a comparison example. In (A) of Fig. 5, a waveform 21 represents the modulation factor command value Vp and shows a change in electrical degree in one cycle.

[0040] This example shows four carrier signals car1 to car4 (waveforms 22a to 22d). The carrier signals car1 and car2 are carrier signals for generating pulse signals on the positive electrode side of two 3-stage pulses. The carrier signals car3 and car4 are carrier signals for generating pulse signals on the negative electrode side of two 3-stage pulses.

[0041] In (B) of Fig. 5 is a 3-stage pulse signal generated based on a comparison with the carrier signal car1 and the carrier signal car4 in response to the modulation factor command value Vp in one cycle, a gate pulse signal GA (a waveform 23a).

[0042] A 3-level pulse signal generated based on a comparison with the carrier signal car2 and the carrier signal car3 in response to the modulation factor command value Vp in one cycle is a gate pulse signal GB (a waveform 23b).

[0043] The gate pulse signal GA is output to the switching branch 8a. The gate pulse signal GB is output to the switching branch 8b.

[0044] In (C) of Fig. 5, a single-phase output voltage Vcnv is shown as a difference (a waveform 25) between the gate pulse signal GA and the gate pulse signal GB.

[0045] In this case, a switching frequency of the switching branch 8b is higher than a switching frequency of the switching branch 8a, and the switching branch 8b has a longer duty cycle.

[0046] Referring to (B) of Fig. 5, based on the comparison of the gate pulse signal for the switching branch 8a with the gate pulse signal for the switching branch 8b, a section in which the switching branch 8a supplies 0 and the switching branch 8b supplies +1E or -1E is longer than other sections.

[0047] Fig. 6 shows a figure illustrating a value of a neutral point current Ic and an increase and decrease of the capacitor voltage difference V diff on the positive electrode side in each pattern.

[0048] As in the Fig. 5 and Fig. 6, the neutral point current Ic flows in a section in which the switching branch 8a supplies 0 and the switching branch 8b supplies +1E or -1E, or in a section in which the switching branch 8b supplies 0 and the switching branch 8a supplies +1E or -1E.

[0049] In this case, when the output voltages of the switching branch 8a and the switching branch 8b fall below such a combination, the neutral point current Ic for charging and discharging the capacitor 5b is -iu (positive and negative of the phase current iu are reversed).

[0050] When a load is large and the phase current iu is large, the capacitor voltage difference V increases and decreases diff repeated strongly in units of 1 / 2 cycle.

[0051] Furthermore, in the case of asynchronous PWM, the carriers that generate pulses are not synchronized with a frequency of the electrical degree. Therefore, part of the neutral point current flowing into and out of capacitor 5b in each electrical degree cycle tends to be asymmetric between the positive electrode side and the negative electrode side, and consequently, the difference between two capacitor voltages of the DC bus circuit tends to become large.

[0052] To solve this problem, switching should be done so that the amount of charging of the capacitor and discharging of the capacitor is smaller by changing an orientation of the neutral point current Ic several times in half a cycle of the electrical degree.

[0053] In the first embodiment, the gate signal selector 13 adjusts the gate signal allocation so that the voltage of the positive-electrode-side capacitor and the voltage of the negative-electrode-side capacitor are balanced. Specifically, when switching to output the two three-stage gate pulse signals GA and GB based on the comparison of carriers, the pulse distributor 14 of the gate signal selector 13 adjusts so that the switching path 8a and the switching path 8b on which the switching elements are turned on alternate.

[0054] In particular, the tuning is performed in such a way that a section in which the neutral point current flows, that is, a section in which the output of one switching branch is 0 and the output of the other switching branch is either +1E or -1E, is made shorter and frequently changes a positive or negative direction.

[0055] Fig. Fig. 7 is a diagram illustrating gate pulse signals GA# and GB# according to the first embodiment. As shown in (A) of Fig. As shown in Figure 7, the gate pulse signal GA is a 3-stage pulse signal GA (waveform 23a) generated based on the comparison with the carrier signal car1 and the carrier signal car4 in response to the modulation factor command Vp. The gate pulse signal GB is a 3-stage pulse signal GB (waveform 23b) generated based on the comparison with the carrier signal car2 and the carrier signal car3 in response to the modulation factor command value Vp.

[0056] As in (B) of Fig. As shown in Figure 7, the gate pulse signals GA# (a waveform 24a) and GB# (a waveform 24b) are tuned such that the turn-on time of a switching pulse in the switching branch 8a and the turn-on time of a switching pulse in the switching branch 8b alternate.

[0057] In (C) of Fig. 7, a single-phase output voltage Vcnv is shown as a difference (waveform 25) between the gate pulse signal GA and the gate pulse signal GB.

[0058] Switching branches 8a and 8b are alternately switched on, the section where one of the two switching branches 8a and 8b supplies 0 E and the other +1 E or -1 E becomes shorter, and the orientation of the current flowing through the neutral point in this section changes from time to time. Therefore, the amount of current charged into and discharged from the capacitor decreases, and the fluctuation in the capacitor voltage difference can also be reduced.

[0059] Furthermore, the capacitor voltage difference V diff according to expression (3) based on the two capacitor voltages V PC and V CN the DC bus circuit of each phase obtained from the neutral point voltage sensors 20a to 20c.

[0060] If an absolute value exceeds a threshold, information about a switching branch to be switched on next is reset once.

[0061] Until the capacitor voltage difference V diff becomes smaller than the threshold value, switching occurs to switching branch 8a → switching branch 8b → switching branch 8a, which is switched on first in each cycle of the frequency command Fc.

[0062] For example, if the absolute value of the voltage difference V diffbetween the two capacitors 5a and 5b in the DC bus circuit, which is obtained by the neutral point voltage sensor 20a in the U-phase circuit, exceeds the threshold value, where the section (= corresponding to one cycle of the frequency command Fc) from a time when the phase of a next inverter output voltage becomes 0 degrees until the phase reaches 360° is defined as fnumber = 1, the switching arm 8a is assigned as the switching arm that is turned on first in this section. In a next cycle fnumber = 2, switching is then carried out such that the switching arm 8b is assigned as the switching arm that is turned on first. The accumulation of an imbalance between the positive and negative of the neutral point current flowing at the time of switching can be eliminated, and an increase in the capacitor voltage difference V diff can also be prevented.

[0063] As described above, the pulse distributor that allocates the gate signals to the switching elements in the switching branch 8a and the switching branch 8b so that the switching elements in the switching branch 8a and the switching elements in the switching branch 8b alternately conduct to provide the voltage +1E or -1E while the inverter 4 outputs the same voltage can suppress the increase of the imbalance between two capacitors 5a and 5b in the DC bus circuit even during the operation of the load and prevent overvoltage tripping (abnormal stop) of the capacitor voltage. Second embodiment.

[0064] Fig. Fig. 8 is a circuit diagram illustrating an overall configuration of a power conversion device 2# according to a second embodiment. As shown in Fig. 8, the energy conversion device 2# differs from that described with reference to Fig. 1 by replacing the control unit 10 with a control unit 10#. The control unit 10# differs from the control unit 10 by replacing the gate signal allocator 13 with a gate signal allocator 13#. The gate signal allocator 13# includes, in comparison to the gate signal allocator 13, a pulse distributor 14 and a neutral point potential switching control unit 15. Since the configuration is otherwise the same as with reference to Fig. 1, the detailed description is not repeated.

[0065] Fig. 9 is a diagram illustrating an operation of the neutral point potential switching control unit 15 according to the second embodiment.

[0066] Fig. 9 (A) is similar Fig. 7 (B), and the tuning of the gate pulse signals GA# (waveform 24a) and GB# (waveform 24b) is carried out such that the on-time of a switching pulse in the switching branch 8a and the on-time of a switching pulse in the switching branch 8b alternate due to the pulse distributor 14.

[0067] Fig. 9 (C) shows a pulse (26a) extracted from only a portion in which either the gate pulse signal GA# (24a) or the gate pulse signal GB# (24b) is 0 and the other is +1E or -1E.

[0068] At this time, a waveform 36 representing the neutral point current is the same as a U-phase motor current 35 while only the switching branch 8a is turned on (+1 E or -1 E). When only the switching branch 8b is turned on (+1 E or -1 E), the waveform 36 is inversely positive and negative to the U-phase motor current 35.

[0069] The neutral point current does not flow (0) when both switching branches 8a and 8b are switched off (0) or both switching branches 8a and 8b are switched on (+1 E or -1E).

[0070] The difference between the voltage of capacitor 5a and the voltage of capacitor 5b in the DC bus circuit decreases while the star or neutral point current 36 is positive and increases while the neutral point current is negative. As an example, Fig. 9 (D) the voltage difference V diff (one waveform 38).

[0071] As a result of the distribution processing by the pulse distributor 14, the increase in the voltage difference between the two capacitors 5a and 5b is suppressed. When the relationship between the frequency command Fc and a frequency of a carrier signal is such that the frequency command is synchronized with the carrier signal in a cycle that is an odd multiple of the carrier and an even multiple of the frequency command, a state may persist in which an odd number of pulses are present in two stages, the magnitude ratio between the on-time of the pulses on the positive electrode side and the on-time of the pulses on the negative electrode side may be fixed, and an asymmetry between the positive and negative may persist.

[0072] When the switching arms 8a and 8b are alternately turned on, the turn-on time on the positive electrode side of one of them is longer, the turn-on time on the negative electrode side of the other is longer, and a state persists in which either a discharge period or a charge period of the capacitor 5b is always slightly longer, which can rapidly increase the voltage difference between the two capacitors 5a and 5b.

[0073] If the absolute value of the voltage difference V diff between two capacitors a permissible threshold value V thand an orientation of the neutral point current is a direction that further increases the voltage difference at the next switching, in order to suppress such an increase, the switching of the switching branch 8a and the switching branch 8b is reversed to change the orientation of the neutral point current to a direction that reduces the capacitor voltage difference to suppress an increase in the voltage difference.

[0074] In particular, if it is expected that the capacitor voltage difference V diff becomes smaller than a threshold value -V th , as seen at a point 39a, and the neutral point current will flow in a direction to reduce the capacitor voltage difference V diffto further increase, such as a waveform 39b at the next switching (switching branch 8a off and switching branch 8b positive electrode on), a tuning is carried out such that the negative electrode of the switching branch 8a is switched on instead, while the switching branch 8b remains switched off to reverse the direction of the neutral point current, and the absolute value of the capacitor voltage difference V diff is within a range of the threshold value, as seen in a waveform 39c.

[0075] Fig. Figure 9 (B) shows gate pulses GAA# and GBB# obtained by further tuning the gate pulses GA# and GB#.

[0076] Fig. 10 is a flowchart illustrating the operation of the neutral point potential switching control unit 15 according to the second embodiment.

[0077] The following is a processing flow in a neutral point potential switching control unit with reference to Fig. 10 described.

[0078] First, it is determined whether or not the absolute value of a difference |V diff_u | between the voltage of the capacitor on the positive electrode side and the voltage of the capacitor on the negative electrode side in a U-phase DC bus circuit exceeds the permissible threshold value V th exceeded (step S401).

[0079] In step S401, when the absolute value of the difference |V diff_u | between the voltage of the capacitor on the positive electrode side and the voltage of the capacitor on the negative electrode side exceeds the permissible threshold value V th has not exceeded (NO in step S401), a state in step S401 is maintained. If the absolute value of the difference |V diff_u | between the voltage of the capacitor on the positive electrode side and the voltage of the capacitor on the negative electrode side exceeds the permissible threshold value V thexceeded (YES in step S401), it is checked whether or not a product of the capacitor voltage difference V diff_u and the phase current iu is greater than a product of the permissible threshold value V th (V th > 0) of the capacitor voltage difference and a phase current threshold I th (I th > 0) (step S402).

[0080] The determination in step S402 that the product of capacitor voltage difference V diff_u and phase current iu is greater than the product of permissible threshold V th (V th > 0) of the capacitor voltage difference and phase current threshold I th (I th > 0) (YES in step S402), means that the positive and negative signs of the capacitor voltage difference V diff_u and the phase current iu are equal.

[0081] As mentioned with reference to Fig.9 (C), which shows the waveforms 26a and 26b of the gate pulse signals that cause the neutral point current, the phase current iu (waveform 35) and the neutral point current Ic (waveform 36) to flow, the capacitor voltage difference V diff when the neutral point current Ic (waveform 36) is on the negative side. When the neutral point current Ic (waveform 36) is on the positive side, the capacitor voltage difference V diff away.

[0082] An expression (4) shows the relationship between capacitor voltage difference V diff and neutral point current Ic. [Expression 4] Vdiff=VPC−VCN=−IcC1s

[0083] In this expression, C1 represents a capacitance of capacitors 5a and 5b. Since the ratio of the positive and negative signs of the capacitor voltage difference V diffand an integrated value of the neutral point current are opposite to each other, the product of capacitor voltage difference V diff and phase current iu, which is positive, that the signs of the integrated value of the neutral point current and the phase current iu are opposite to each other.

[0084] When switching, where the absolute value of the capacitor voltage difference V diff continues to rise, the switching branch 8a (also called leg or branch 8a) provides 0 (a C-potential) and the switching branch 8b (also called leg or branch 8b) provides a +1E or -1E output.

[0085] Therefore, following step S402, it is determined whether or not the next outputs of the branches 8a and 8b are branch 8a = 0 and branch 8b = -1E and the inverter 4 provides the +1E output (step S403).

[0086] If it is determined in step S403 that the next outputs of branches 8a and 8b are branch 8a = 0 and branch 8b = -1E and the inverter 4 provides the +1E output (YES in step S403), the absolute value of the capacitor voltage difference exceeds V diff the threshold value V th and continues to rise. Therefore, this switching should be swapped. When switching after swapping, branch 8a = +1E and branch 8b = 0. As described above, when the outputs of branch 8a and branch 8b are changed, if only one of the switching elements on the positive electrode side and the negative electrode side of switching branches 8a and 8b is switched (conductive or non-conductive), a condition for the combination of current outputs from the switching branches that can be swapped is naturally limited.

[0087] The combination of the current outputs of branches 8a and 8b, where the next switching outputs before the swap can be branch 8a = 0 and branch 8b = -1E or after the swap branch 8a = +1E and branch 8b = 0, includes two types of combinations, namely the combination of branch 8a = +1E and branch 8b = -1E, which results in a combination where an inverter output is +2E, and branch 8a = branch 8b = 0, which results in a combination where the output is 0.

[0088] In step S404, it is determined whether the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, which results in the combination where the output is 0 (step S404).

[0089] If it is determined in step S404 that the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (YES in step S404), the switching between branch 8a and branch 8b is swapped. In other words, the setting is branch 8a = +1E and branch 8b = 0, rather than the setting is branch 8a = 0 and branch 8b = -1E. Specifically, when the inverter output is +2E (branch 8a = +1E and branch 8b = -1E), the switching element 34 on the negative electrode side of branch 8b can be switched on → off to realize an output combination state achieved by swapping the switching of branch 8a and branch 8b.When the inverter output is 0 (branch 8a = branch 8b = 0), the switching element 31 on the positive electrode side of the branch 8a can be turned off → on to realize the output combination state achieved by interchanging the switching of branch 8a and branch 8b.

[0090] On the other hand, if it is determined in step S404 that the combination of current outputs of branch 8a and branch 8b is not the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (NO in step S404), the process returns to step S401.

[0091] On the other hand, if it is determined in step S403 that the next outputs of the branches 8a and 8b are not branch 8a = 0 and branch 8b = -1E and the inverter 4 does not provide the +1E output (NO in step S403), it is determined whether or not branch 8a = 0 and branch 8b = +1E and the inverter output is the -1E output (step S406).

[0092] If it is determined in step S406 that the next outputs of branches 8a and 8b are branch 8a = 0 and branch 8b = +1E and inverter 4 provides the -1E output (YES in step S406), the absolute value of the capacitor voltage difference exceeds V diff the threshold value V thand continues to rise. Therefore, this switching should be swapped. When switching after the swap, branch 8a = -1E and branch 8b = 0. As described above, when the outputs of branch 8a and branch 8b are swapped, if only one of the switching elements on the positive electrode side and the negative electrode side of switching branches 8a and 8b is switched (conductive or non-conductive), the condition for the combination of current outputs from the switching branches that can be swapped is naturally limited.

[0093] The combination of the current outputs of branches 8a and 8b, where the next switching outputs before the swap can be branch 8a = 0 and branch 8b = +1E or after the swap branch 8a = -1E and branch 8b = 0, includes two types of combinations, namely the combination of branch 8a = -1E and branch 8b = +1E, which results in a combination where the inverter output is -2E, and branch 8a = branch 8b = 0, which results in the combination where the output is 0.

[0094] In step S407, it is determined whether the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = -1E and branch 8b = +1E or branch 8a = branch 8b = 0, which results in the combination where the output is 0 (step S407).

[0095] If it is determined in step S407 that the combination of the current outputs of arm 8a and arm 8b is the combination of arm 8a = -1E and arm 8b = +1E or arm 8a = arm 8b = 0, resulting in the combination where the output is 0 (YES in step S407), the switching between arm 8a and arm 8b is swapped. In other words, the setting is arm 8a = -1E and arm 8b = 0, rather than the setting is arm 8a = 0 and arm 8b = +1E. Specifically, when the inverter output is -2E (arm 8a = -1E and arm 8b = +1E), the switching element 33 on the positive side of arm 8b can be turned on → off to realize the output combination state resulting from the swapping of the switching of arm 8a and arm 8b.When the inverter output is 0 (branch 8a = branch 8b = 0), the switching element 32 on the negative electrode side of the branch 8a can be switched off → on to realize the output combination state resulting from the interchange of the switching of branch 8a and branch 8b.

[0096] On the other hand, if it is determined in step S406 that the next outputs of branches 8a and 8b are not branch 8a = 0 and branch 8b = +1E, and inverter 4 does not provide the -1E output (NO in step S406), the process returns to step S401. If it is determined in step S407 that the combination of the current outputs of branch 8a and branch 8b is not the combination of branch 8a = -1E and branch 8b = +1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (NO in step S407), the process returns to step S401.

[0097] In the switching control according to the second embodiment, in order to reduce the imbalance of the difference between two capacitor voltages in the DC bus circuit, to reduce the switching loss of the inverter 4, and to eliminate the imbalance of the element losses by the pulse distributor 14, only one of the elements 31 and 33 on the P side (positive electrode) and 32 and 34 on the N side (negative electrode) in the two switching arms 8a and 8b is controlled to be turned on or off at a time.

[0098] On the other hand, if it is determined in step S402 that the product of capacitor voltage difference V diff_u and phase current iu is less than the product of permissible threshold V th (V th > 0) of the capacitor voltage difference and phase current threshold I th (I th > 0) (NO in step S402), then it is determined whether or not the product of capacitor voltage difference V diff_uand phase current iu is less than -1 times the product of permissible threshold value V th (V th > 0) of the capacitor voltage difference and phase current threshold I th (I th > 0) (step S410).

[0099] If it is determined in step S410 that the product of capacitor voltage difference V diff_u and phase current iu is less than -1 times the product of permissible threshold V th (V th > 0) of the capacitor voltage difference and phase current threshold I th (I th > 0) (YES in step S410), the positive and negative signs of the capacitor voltage difference and the phase current are different because the sign of the product is negative.

[0100] Therefore, considering expression (4), the positive and negative signs of the integrated value of the neutral point current and the phase current are the same.

[0101] When switching, where the absolute value of the capacitor voltage difference V diff continues to rise, the switching branch 8a delivers +1E or -1E and the switching branch 8b delivers 0.

[0102] Therefore, following step S410, it is determined whether or not the next outputs of the branches 8a and 8b are branch 8a = +1E and branch 8b = 0 and the inverter 4 provides the +1E output (step S411).

[0103] If it is determined in step S411 that the next outputs of branches 8a and 8b are branch 8a = +1E and branch 8b = 0 and the inverter 4 provides the +1E output (YES in step S411), the absolute value of the capacitor voltage difference exceeds V diff the threshold value V thand continues to rise. Therefore, this switching should be reversed. When switching after the switchover, branch 8a = 0 and branch 8b = -1E. As described above, when the outputs of branch 8a and branch 8b are reversed, if only one of the switching elements on the positive electrode side and the negative electrode side of switching branches 8a and 8b is switched (conductive or non-conductive), the condition for the combination of current outputs from the switching branches that can be reversed is naturally limited.

[0104] The combination of current outputs from branches 8a and 8b, where the next switching outputs before the swap can be branch 8a = +1E and branch 8b = 0 or after the swap branch 8a = 0 and branch 8b = -1E, includes two types of combinations, namely the combination of branch 8a = +1E and branch 8b = -1E, which leads to the combination where the inverter output is +2E, and branch 8a = branch 8b = 0, which leads to the combination where the output is 0.

[0105] In step S412, it is determined whether the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, which results in the combination where the output is 0 (step S412).

[0106] If it is determined in step S412 that the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (YES in step S412), the switching between branch 8a and branch 8b is swapped. In other words, the setting is branch 8a = 0 and branch 8b = -1E, rather than the setting of branch 8a = +1E and branch 8b = 0. Specifically, when the inverter output is +2E (branch 8a = +1E and branch 8b = -1E), the switching element 31 on the positive electrode side of branch 8a can be switched on → off to realize the output combination state achieved by swapping the switching of branch 8a and branch 8b.When the inverter output is 0 (branch 8a = branch 8b = 0), the switching element 34 on the positive electrode side of the branch 8b can be switched off → on to realize the state of the output combination resulting from exchanging the switching of branch 8a and branch 8b.

[0107] On the other hand, if it is determined in step S412 that the combination of current outputs of branch 8a and branch 8b is not the combination of branch 8a = +1E and branch 8b = -1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (NO in step S412), the process returns to step S401.

[0108] If it is determined in step S411 that the next outputs of the branches 8a and 8b are not branch 8a = +1E and branch 8b = 0 and the inverter 4 does not provide the +1E output (NO in step S411), it is determined whether or not branch 8a = -1E and branch 8b = 0 and the inverter output is the -1E output (step S414).

[0109] If it is determined in step S414 that the next outputs of branches 8a and 8b are branch 8a = -1E and branch 8b = 0 and the inverter 4 provides the -1E output (YES in step S414), the absolute value of the capacitor voltage difference V diff the threshold value V th and continues to rise. Therefore, this switching should be swapped. When switching after swapping, branch 8a = 0 and branch 8b = +1 E. As described above, when changing the outputs of branch 8a and branch 8b, if only one of the switching elements on the positive electrode side and the negative electrode side of switching branches 8a and 8b is switched (conducting or non-conducting), the condition for the combination of current outputs from the switching branches that can be swapped is naturally limited.

[0110] The combination of current outputs of branches 8a and 8b, where the next switching outputs before the swap can be branch 8a = -1E and branch 8b = 0 or after the swap can be branch 8a = 0 and branch 8b = +1E, includes two types of combinations, namely the combination of branch 8a = -1E and branch 8b = +1E, which leads to the combination where the inverter output is -2E, and branch 8a = branch 8b = 0, which leads to the combination where the output is 0.

[0111] In step S415, it is determined whether the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = -1E and branch 8b = +1E or branch 8a = branch 8b = 0, which results in the combination where the output is 0 (step S415).

[0112] If it is determined in step S415 that the combination of the current outputs of branch 8a and branch 8b is the combination of branch 8a = -1E and branch 8b = +1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (YES in step S415), the switching between branch 8a and branch 8b is swapped. In other words, the setting is branch 8a = 0 and branch 8b = +1E, rather than the setting of branch 8a = -1E and branch 8b = 0. Specifically, when the inverter output is -2E (branch 8a = -1E and branch 8b = +1E), the switching element 32 on the negative electrode side of branch 8a can be switched on → off to realize the output combination state achieved by swapping the switching of branch 8a and branch 8b.When the inverter output is 0 (branch 8a = branch 8b = 0), the switching element 33 on the positive electrode side of the branch 8b can be switched off → on to realize the state of the output combination resulting from exchanging the switching of branch 8a and branch 8b.

[0113] On the other hand, if it is determined in step S414 that the next outputs of branches 8a and 8b are not branch 8a = -1E and branch 8b = 0, and inverter 4 does not provide the -1E output (NO in step S414), the process returns to step S401. If it is determined in step S415 that the combination of the current outputs of branch 8a and branch 8b is not the combination of branch 8a = -1E and branch 8b = +1E or branch 8a = branch 8b = 0, resulting in the combination where the output is 0 (NO in step S415), the process returns to step S401.

[0114] The neutral point potential switching control unit 15 performs the above processing not only to reduce the switching imbalance by the switching elements, minimizing the switching by each element, but also to suppress a sudden increase in the voltage difference between two capacitors 5a and 5b in the DC bus circuit due to the influence of the load and the power factor by switching under a prescribed frequency condition, but also to suppress the generation of an overvoltage of the capacitor voltage and the increase in the torque ripple due to asymmetry between the positive output voltage and the negative output voltage and the resulting unstable control.

[0115] The present disclosure is capable of freely combining part or all of each embodiment or modifying or omitting each embodiment as applicable within the scope of the disclosure.

[0116] The configurations exemplified as the above-described embodiments are exemplary configurations of the present disclosure, and can be combined with other known techniques or configured to be modified, for example, partially omitted, without departing from the essence of the present disclosure. In the above-described embodiments, the processing and configuration described in other embodiments can be adopted and implemented as appropriate.

[0117] It should be understood that the embodiments disclosed herein are in all respects illustrative and not restrictive. The scope of the present disclosure is defined by the terms of the claims rather than by the above description and is intended to include all modifications within the scope and meaning that correspond to the terms of the claims. LIST OF REFERENCE SYMBOLS

[0118] 2 Energy conversion device; 3 Motor; 4 Inverter; 5a Capacitor on the positive electrode side; 5b Capacitor on the negative electrode side; 6, 31, 32, 33, 34 Switching element; 7 Clamping diode; 8a, 8b Switching branch; 10 Control unit; 11 Modulation factor calculation unit; 12 Carrier comparison PWM generator; 13 Gate signal sorter; 14 Pulse distributor; 15 Neutral point potential switching control unit; 18 Current sensor; 19 V / f controller; 20a, 20b, 20c Neutral point voltage sensor. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] JP 4 352 787 B2

[0006] JP 6 707 298 B2

[0006]

Claims

[1] Energy conversion device, comprising: an inverter having a plurality of switching elements, the inverter receiving a DC voltage from a DC voltage source and converting the DC voltage into an AC voltage with variable voltage and variable frequency to output the AC voltage to a load; a control unit that controls an on / off control of the plurality of switching elements by pulse width modulation or PWM control; and a series body comprising a capacitor on one side of the positive electrode and a capacitor on one side of the negative electrode, which is connected on an input side of the inverter between a positive electrode and a negative electrode of the DC voltage source, wherein an output potential of the inverter has at least a positive electrode potential and a negative electrode potential of the DC voltage source and a neutral point potential which is a connection point between the capacitor on the positive electrode side and the capacitor on the negative electrode side, and the control unit has: a modulation factor calculation unit that calculates a modulation factor of the inverter based on the DC voltage and an output voltage command value, a gate signal generator that generates a gate signal for driving the switching elements on / off to generate a pulse train based on a comparison between the calculated modulation factor and a carrier signal, and a gate signal allocator that adjusts an allocation of the gate signal so that a voltage of the capacitor on the positive electrode side and a voltage of the capacitor on the negative electrode side are balanced. [2] The power conversion device according to claim 1, wherein the inverter includes a plurality of switching circuits, and the gate signal allocator includes a pulse distributor that sequentially allocates the gate signal for on / off driving switching elements in the plurality of switching circuits and allocates pulse generation to the plurality of switching circuits. [3] The power conversion device according to claim 2, wherein the pulse distributor assigns an order such that the plurality of circuits are alternately turned on in conjunction with the gate signal required for the on / off control of the switching elements in the plurality of circuits. [4] The power conversion device according to claim 2 or 3, wherein the gate signal allocator further includes a neutral point potential switching control unit that, when an absolute value of a difference between the voltage of the capacitor on the positive electrode side and the voltage of the capacitor on the negative electrode side exceeds a threshold value, determines whether the difference will further increase based on an allocation of the gate signal, and changes the allocation of the gate signal to the switching elements in the plurality of switching circuits so that the difference will decrease when the gate signal allocator determines that the difference will increase.

Images