Patent Application
20230231467
The present disclosure relates to a power conversion device.
In recent years, in power conversion devices used for high-voltage application such as a power grid, a multilevel converter in which a plurality of converter cells each including an energy storage element are connected in series in a multiplexed manner, has been put into practice. Such a converter is called a modular multilevel converter (MMC) type, a cascaded multilevel converter (CMC) type, or the like.
In a conventional power conversion device that converts three-phase AC power using a general inverter/converter as described in, for example. Patent Document 1, when imbalance of a power grid is detected, a sinusoidal base signal is generated for each phase, and a third-order harmonic is superimposed on each base signal, to generate a voltage command signal for each phase. Thus, the peak of the voltage command is reduced, whereby utilization of DC voltage can be improved.
Meanwhile, a conventional power conversion device described in Patent Document 2 is a power conversion device of the MMC type and outputs zero-phase-sequence voltage with a third-order harmonic of a fundamental wave superimposed thereon to an AC circuit as in the control described in Patent Document 1.
In the above conventional power conversion device of the MMC type, a different capacitor is provided for each arm, and power pulsation of each arm flows in/out to/from each capacitor. Therefore, depending on the capacitances of the capacitors or the operation condition, voltage pulsation might occur in the capacitors of each arm, thus hampering improvement in utilization of DC voltage.
The present disclosure has beer, made to solve the above problem, and an object of the present disclosure is to provide a power conversion device of the MMC type that can assuredly improve utilization of DC voltage.
A power conversion device according to the present disclosure includes: a power converter for performing power conversion between a DC circuit and a three-phase AC circuit and a control device for performing output control of the power converter. The power converter includes, for each phase of the three-phase AC circuit, two arms provided between common DC terminals and connected to each other, a connection portion therebetween being connected to the corresponding phase of the three-phase AC circuit, each of the arms being formed by connecting, in series, a plurality of converter cells each composed of a plurality of semiconductor switching elements and a DC capacitor. The control device includes a voltage command generation unit which generates, for each of the arms of the power converter, an output voltage command for the plurality of converter cells, and the control device generates, for each of the arms, a base command for output voltage of the plurality of converter cells, performs phase adjustment of a zero-phase-sequence voltage command having a frequency component that is three times a fundamental frequency of the three-phase AC circuit on the basis of voltage of the DC capacitor and the output voltage command, and superimposes the phase-adjusted zero-phase-sequence voltage command on the base command, thus generating the output voltage command.
The power conversion device according to the present disclosure makes it possible to provide a power conversion device of the me type that can assuredly improve utilization of DC voltage.
As shown in
The power converter 10 includes a plurality of leg circuits 100u, 100v, 100w (referred to as leg circuit(s) 100 when mentioning them collectively or any of them) connected in parallel to each other, between a positive DC terminal 5F and a negative DC terminal 5N which are common DC terminals.
The leg circuit 100 is provided for each of a plurality of phases (in this case, three phases U, V, W) forming AC. The leg circuits 100 are connected between the AC grid 2 and the DC circuit 4 and perform power conversion between AC and DC. AC input terminals 6 respectively provided to the leg circuits 100u, 100v, 100w are connected to the AC grid 2 via a transformer 3.
The positive DC terminal 5P and the negative DC terminal 5N connected in common to the leg circuits 100 are connected to the DC circuit 4. The DC circuit 4 is, for example, a DC power grid including a DC power transmission network and the like, or another power conversion device, in the former case, a high-voltage DC power transmission (HVDC) system is formed. In the latter case, two power conversion devices are connected, thus forming a back-to-back (ETB) system for connecting two AC grids different in rated frequency or the like.
The leg circuit 100u includes a U-phase upper arm Pu from the positive DC terminal 5P to the AC input terminal 6, and a D-phase lower arm Nu from the negative DC terminal 5N to the AC input terminal 6. The leg circuit 100v includes a V-phase upper arm Pv from the positive DC terminal 5P to the AC input terminal 6, and a V-phase lower arm Nv from the negative DC terminal 5N to the AC input terminal 6. The leg circuit 100w includes a W-phase upper arm Pw from the positive DC terminal 5P to the AC input terminal 6, and a w-phase lower arm Nw from the negative DC terminal 5N to the AC input terminal 6.
That is, the upper arms Pu, Pv, Pw for the respective phases and the lower arms Nu, Nv, Nw for the respective phases are connected in series to each other, and the connection points therebetween serve as the AC input terminals 6 for the respective phases.
The leg circuits 100u, 100v, 100w have the same configuration. Therefore, hereinafter, the leg circuit. 100u for U phase will be described as a representative.
The arm Pu is formed by connecting a plurality of (N) converter cells 11 and a reactor 12P in series. Similarly, the arm Nu is formed by connecting a plurality of (N) converter cells 11 and a reactor 12N in series.
The position where the reactor 12P is provided may foe any position in the arm Pu, and the position where the reactor 12N is provided may be any position in the arm Nu.
As each of the reactors 12P, 12N, a plurality of reactors may foe provided, and the inductance values of the reactors may be different from each other. Only one of the reactors 12P, 12N may be provided and another one may be omitted.
Although the case where the AC input terminals 6 are connected to the AC grid 2 via the transformer 3 is shown here, a configuration in which the AC input terminals 6 are connected to the AC grid 2 via an interconnection reactor instead of the transformer 3, may be adopted.
Instead of the AC input terminals 6 for the respective phases, primary windings may foe provided to the leg circuits 100u, 100v, 100w and the leg circuits 100u, 100v, 100w may foe electrically connected to the transformer 3 via secondary windings magnetically coupled with the primary windings. In this case, the reactors 12P, 12N may be used as the primary winding for each phase.
That is, the upper arras Pu, Pv, Pw for the respective phases and the lower arms Nu, Nv, Nw for the respective phases are connected to each other by connection portions such as the AC input terminals 6 or the primary windings, and are electrically connected to the AC grid 2 via the connection portions.
The power conversion device 1 further includes DC voltage detectors 7P, 7N, an arm current detector 8 provided to each of the arms Pu, Nu, Pv, Nv, Pw, Nw, an AC voltage detector 9A, and an AC current detector 9B, as detectors for measuring electric quantities (current, voltage, etc.) used for control. Signals from these detectors are inputted to the control device 20 via signal lines.
The signal lines are formed by optical fibers, for example. In
Hereinafter, each detector will be specifically described.
The DC voltage detector 7P detects DC voltage VdcP of the positive DC terminal 5P. The DC voltage detector 7N detects DC voltage VdcN of the negative DC terminal 5N. A difference between the DC voltage VdcP and the DC voltage VdcN is defined as DC voltage Vac.
The arm current detectors 8 provided to the arms Pu, Nu, Pv, Nv, Pw, Nw respectively detect am currents IPu, INu, IPv, INv, XPw, INw flowing through the arms Pu, Nu, Pv, Nv, Pw, Nw. In the following description, the arm currents IPu, IPv, IPw are collectively referred to as upper am currents IAP, the arm currents INu, INv, INw are collectively referred to as lower arm currents IAN, and the upper arm current IAP and the lower arm current IAN are collectively referred to as arm currents IA.
The AC voltage detector 9A detects AC voltage Vacu for U phase, AC voltage Vacv for V phase, and AC voltage Vacw for w phase of the AC grid 2. In the following description, Vacu, Vacv, and Vacw are collectively referred to as Vac.
The AC current detector 9B detects AC current. Iacu for U phase, AC current Iacv for V phase, and AC current Iacw for W phase of the AC grid. In the following description, Iacu, Iacv, and Iacw are collectively referred to as lac.
The converter cell 11 shown in
This converter cell 11 includes a series unit formed by connecting, in series, two switching elements Q1, Q2 to which diodes D are connected in antiparallel, a DC capacitor 13, a voltage detector 14, and a bypass switch 15. The series unit of the switching elements Q1, Q2 and the DC capacitor 13 are connected in parallel.
In this case, both terminals of the switching element Q2 serve as input/output, terminals 11A, 11B of the converter cell 11. Through switching operations of the switching elements Q1, Q2, voltage Vc across the DC capacitor 13 or zero voltage is outputted. For example, when the switching element Q1 is OK and the switching element Q2 is OFF, the voltage Vc across the DC capacitor 13 is outputted. When the switching element Q1 is OFF and the switching element Q2 is ON, zero voltage is outputted.
The voltage detector 14 detects the voltage Vc across the DC capacitor 13. The bypass switch 15 is connected between the input/output terminals 11A, 11B. For example, by turning on the bypass switch 15 when the AC grid 2 is abnormal, the converter cell 11 is short-circuited, whereby the switching elements Q1, Q2 in the converter cell 11 are protected from overcurrent.
The converter cell 11 shown in
In this case, the middle point in the first series unit of the switching elements Q3, Q4 and the middle point in the second series unit of the switching elements Q5, Q6 serve as input/output terminals HA, 11B of the converter cell 11. Through switching operations of the switching elements Q3 to Q6, positive/negative voltage Vc across the DC capacitor 13 or zero voltage is outputted.
Similarly to the converter cell 11 shown in
As the switching elements Q1 to Q6 in the converter cell 11 shown in
In the following description, the converter cell 11 having the half-bridge configuration shown in
The converter cell 11 may have a configuration other than the above-described ones, and for example, a circuit configuration called a clamped double cell may be used. Also, the switching elements Q2 to Q6 and the DC capacitor 13 are not limited to the above-described ones.
Next, the configuration of the control device 20 will be described.
As described above, the control device 20 receives the DC voltages VdcP, VdcN, the arm currents IA of the arms Pu, Nu, Pv, Nv, Pw, Nw, the AC voltage Vac for each phase, the AC current lac for each phase, and the capacitor voltage Vc of each converter cell 11, which are detected values. Then, on the basis of the above received information, the control device 20 generates and outputs gate signals gs1, gs2 for driving the switching elements Q1, Q2 of each converter cell 11.
In the control device 20, from the above received information, the DC voltage Vac and DC current lac are calculated, and further, circulation currents Iccu, Iccv, Iccw for the respective phases are calculated. In the following description, Iccu, Iccv, and Iccw are collectively referred to as Icc.
The circulation current Icc for each phase is current circulating in the power converter 10 without flowing to the AC side and the DC side, and is represented as follows.
ICC=(IAP+IAN)/2−Idc/3
The control device 20 includes the main control unit 20A, and the arm control units 20B provided for the respective arms Pu, Nu, Pv, Nv, Pw, Nw. In
As shown in
The DC control unit 21 receives the DC voltage Vdc and the DC current Idc, and on the basis of these, generates a DC voltage command Vdc*. The generated DC voltage command Vdc* is inputted to the arm voltage command generation unit 24.
The circulation current, control unit 22 receives the circulation currents Ice for the respective phases, and on the basis of the circulation currents Icc, generates circulation voltage commands Vccu*, Vccv*, Vccw* for the respective phases. In the following description, Vccu*, Vccv*, and Vccw* are collectively referred to as Vcc*. The generated circulation voltage commands Vcc* for the respective phases are inputted to the arm voltage command generation unit 24.
The AC control unit 23 receives the AC voltages Vac for the respective phases and the AC currents lac for the respective phases, and on the basis of these, generates AC voltage commands Vacu*, Vacv*, Vacw* for the respective phases. In the following description, Vacu*, Vacv*, and Vacw* are collectively referred to as Vac*. The generated AC voltage commands Vac* for the respective phases are inputted to the arm voltage command generation unit 24.
The arm voltage command generation unit 24 receives the DC voltage command Vdc*, the AC voltage commands Vac* for the respective phases, and the circulation voltage commands Vcc* for the respective phases, and further receives a zero-phase-sequence voltage command Vo* generated by the zero-phase-sequence voltage command generation unit 25. Then, the arm voltage command generation unit 24 generates arm voltage commands VAPu*, VANu*, VAPv*, VANv*, VAPw*, VANw* (referred to as arm voltage commands VA* when mentioning them collectively) as output voltage commands for the respective arms Pu, Nu, Pv, Nv, Pw, Nw.
First, a base command based on the DC voltage command Vdc*, the AC voltage command Vac* for each phase, and the circulation voltage command Vcc* for each phase is generated, and the zero phase-sequence voltage command Vo* its superimposed on the base command, whereby the arm voltage command VA* is generated.
For example, the arm voltage commands VAPu*, VANu* for U phase are represented by the following Expressions (1) and (2). The same applies to V phase and W phase.
VAPu*=Vdc*−Vacu*+Vccu*−Vo* (1)
VANu*=Vdc*+Vacu*+Vccu*+Vo* (2)
The capacitor voltage summing unit 27 obtains the capacitor voltages Vc of the respective converter cells 11 and calculates capacitor voltage sums VcAPu, VcANu, VcAPv, VcANv, VcAPw, VcANw (referred to as capacitor voltage sums VcA when mentioning them collectively) for the respective arms Pu. Nu, Pv, Nv, Pw, Nw. The arm control unit 20B for each arm Pu, Nu, Pv, Nv, Pw, Nw collects N capacitor voltages Vc of the converter cells 11.
The PLL unit 28 extracts a phase θs from the AC voltage Vac for each phase. For example, the AC voltage Vacu for U phase is represented as Vacu=Vp·sin θs, using an amplitude Vp.
The zero-phase-sequence voltage command generation unit 25 receives the AC voltage commands Vac* for the respective phases, the respective arm voltage commands VA*, the capacitor voltage sums VcA for the respective arms, and the phase θs, and generates the zero-phase-sequence voltage command Vo*. The details of the zero-phase-sequence voltage command generation unit 25 will be described later.
The arm modulation command generation unit 26 receives the respective arm voltage commands VA* and the capacitor voltage sums VcA for the respective arms Pu, Nu, Pv, Nv, Pw, Nw. and generates arm modulation commands KAPu*, KANu*, KAPv*, KANv*, KAPw*, XAKw* (referred to as arm modulation commands KA* when mentioning them collectively) for the respective arms Pu, Nu, Pv, Nv, Pw, Nw.
As described above, the control device 20 includes similar arm control units 20E for the respective arms Pu, Nu, Pv, Nv, Pw, Nw. Hereinafter, the arm control unit 20P, for the arm Nu will be described.
As shown in
Each cell control unit 23 operates as follows.
In the converter cell 11 that is a control target of the cell control unit 29, the capacitor voltage Vc is detected by the voltage detector 14, and the capacitor voltage Vc is inputted to the cell control unit 23. In addition, the detected capacitor voltage Vc is also outputted from the cell control unit 29 to the capacitor voltage summing unit 27 of the main control unit 20A.
The cell control unit 29 controls the converter ceil 11 so that the capacitor voltage Vc of the converter cell 11 that is the control target approaches an average value (VCANu/N) of the capacitor voltages in the arm Nu.
More specifically, the ceil control unit 29 calculates a control quantity from the arm current INu and a deviation between the average value (VcANu/N) of the capacitor voltages and tires corresponding capacitor voltage Vc, and superimposes the control quantity on the arm modulation command KANu*, to correct the arm modulation command KANu*. Then, on the basis of the corrected arm modulation command KANu*, the cell control unit 29 generates and outputs gate signals gs1, gs2 for driving the switching elements Q1, Q2 of the corresponding converter cell 11.
In the above description of the control device 20, for simplification, the case where control for balancing the capacitor voltage sums VcA for the respective arms Pu, Mu, Pv, Nv, Pw, Nw is omitted has been shown. However, control for balancing the DC voltages between phases or between arms may be added.
As shown in
The zero-phase-sequence voltage command Vo* has a common value among U phase, V phase, and W phase, and theoretically, does not influence outputted AC voltage between phases. In addition, since the zero-phase-sequence voltage command Vo* has a frequency component that is three times the fundamental frequency of the AC voltage, when the AC voltage Vac is balanced and the numbers of normal converter cells 11 (the numbers of operating converter cells 11) included in the respective arms Pu, Nu, Pv, Nv, Pw, Nw are equal, instantaneous power for each phase is balanced.
The AC voltage command Vac* inputted to the zero-phase-sequence voltage command generation unit 25 is inputted to a zero-phase-sequence voltage generation unit 30 and a reference phase generation unit 31 in the zero-phase-sequence voltage command generation unit 25.
The reference phase generation unit 31 detects phase shift θv of AC voltage due to voltage drop by the transformer 3 and the reactor 12P (or 12N) (hereinafter, may be simply referred to as phase θv), and the phase shift θv is inputted to a phase adjustment unit 32. On the basis of the phase θv outputted from the reference phase generation unit 31, the reference phase of the AC voltage command Vac* becomes (θs+θv).
The phase adjustment unit 32 further receives the respective arm voltage commands VA*, the capacitor voltage sums VcA for the respective arms, and the phase θs, and determines an adjustment phase θo fox adjusting the phase of the zero-phase-sequence voltage command Vo* (hereinafter, may be simply referred to as phase θo).
The zero-phase-sequence voltage generation unit 30 generates the zero-phase-sequence voltage command Vo* on the basis of the AC voltage command Vac* and the phases θs, θv, θo.
When the circulation current Icc is 0, the arm voltage command VANu* for the U-phase lower arm Nu is represented by the following Expression (3), on the basis of the above Expression (2).
VANu*=Vdc*+Vacu*+Vo* (3)
Here, Vdc* and Vacu* are represented as Vdc*=Vdc/2 and Vacu*=Vp·sin(θs+θv), and the zero-phase-sequence voltage command Vo* is represented by the following Expression (4), using a constant a.
Vo*=a·Vp·sin 3(θs+θv+θo) (4)
Then, the above Expression (3) can be deformed into the following Expression (5).
VANu*=(Vdc/2)+Vp·sin(θs+θv)+a·Vp·sin 3(θs+θv+θo) (5)
In a general 2-level three-phase power conversion device not having a MMC configuration, a common DC capacitor is connected for three phases collectively. Therefore, power pulsations occurring in the respective phases are canceled out and the voltage of the DC capacitor is constant while hardly pulsating. In such a case, phase adjustment for the zero-phase-sequence voltage command to be superimposed is not needed, that is, with the adjustment phase θo=0, the peak of the AC output voltage can be reduced. Here, power flows in/out to/from the DC capacitor due to switching, but is negligible because the switching frequency is a high frequency.
Also in a power conversion device having a MMC configuration, the same applies if voltage of the DC capacitor is almost constant without pulsating. This will be shown as a first comparative example below.
As shown in
In the first comparative example, it is found that the peak of the arm voltage command VAβ* on which the zero-phase-sequence voltage command with no phase adjustment has been superimposed is reduced. In this case, by setting the constant a to 1/6, a necessary output voltage range can be reduced by about 13%.
The power conversion device 1 having the MMC configuration according to embodiment 1 is used in a HVDC system or a BTB system, and in actuality, power pulsation occurring in each arm directly flows in/out to/from the DC capacitor 13 of each converter cell 11. For example, the arm current INu of the arm Nu which is the U-phase lower arm is represented by the following Expression (6). Here, Ip is an amplitude and φ is a power factor.
INu=Idc/3+Ip·sin(θs+φ) (6)
Instantaneous power pNu flowing in/out to/from the arm Nu is calculated as a product of the arm voltage command VANu* shown by the above Expression (5) and the arm current INu shown by the above Expression (6), and is represented by the following Expression (7). Here, the influence on the instantaneous power pNu due to superimposition of the zero-phase-sequence voltage command Vo* in Expression (5) is small and therefore is regarded as 0, for simplification.
pNu=(Vdc·Ip/2)·sin(θs+φ)−(Vp·Idc/3)·sin(θs+θv)−(Vp·Ip/2)·cos(2θs+θv+w) (7)
From the above Expression (0), it is found that power pulsation having a frequency that is equal to or two times the fundamental frequency of the AC grid 2 flows in/out to/from the arm Nu. Power pulsation ΔWNu at t=0 in the DC capacitor 13 in the arm Nu is represented by the following Expression (8). Here, θs=ωt is satisfied.
ΔWNu=−(Vdc·Ip/2Ω)·cos(θs+φ)+(Vp·Idc/3ω)·cos(θs+θv)−(Vp·Ip/4ω)·sin(2θs+θv+φ) (8)
With the power pulsation ΔWNu, the capacitor voltage sum VcANu for the arm Nu pulsates, and voltage that can be outputted by the arm Nu also varies with time. The pulsation of the capacitor voltage sum VcA increases as the capacitance of the DC capacitor 13 becomes smaller.
A case where, while the capacitor voltage sum VcA varies with time, the arm voltage command VAβ* on which the zero-phase-sequence voltage command with no phase adjustment has been superimposed is used as in the first comparative example, will be shown as a second comparative example below.
In the second comparative example, the arm voltage command VAβ* does not fall within an output voltage range between the value of the capacitor voltage sum VcA and 0, and as a result, there is a region where voltage cannot be outputted.
In addition, each converter cell 11 is individually controlled and the output voltage thereof might increase/decrease more or less. Further, pulsation due to switching is superimposed on the capacitor voltage Vc of each converter cell 11, so that the range of voltage that can be outputted might change more or less.
From the above, it is desirable that the arm voltage command falls within the output voltage range with a margin provided.
In the present embodiment, by performing phase adjustment of the zero-phase-sequence voltage command Vo*, the arm voltage command VA* is caused to fall within the output voltage range between the value of the capacitor voltage sum VcA and 0 with a margin provided. In this case, the control device 20 changes the adjustment phase θo for the zero-phase-sequence voltage command Vo* so that a margin of the arm voltage command VA* with respect to the output voltage range increases.
Phase adjustment of the zero-phase-sequence voltage command Vo* will be described in detail below.
As the margin of the arm voltage command VA* with respect to the output voltage range (0 to VcA) of the arm, there are an upper margin σU to an upper limit value VcA and a lower margin σL to a lower limit value 0. The upper margin σU is a control margin in the present disclosure. Both margins are instantaneous values. The upper margin σU is a value obtained by subtracting the arm voltage command VA* from the capacitor voltage sum VcA and the lower margin σL is equal to the arm voltage command VA*.
In order to increase the margin of the arm voltage command VA* with respect to the output voltage range (0 to VcA) of the arm, both of a minimum value σUmin. of the upper margin σU and a minimum value σLmin of the lower margin σL need to be made as great as possible in one cycle of the AC voltage. Therefore, the adjustment phase θo for the zero-phase-sequence voltage command Vo* is changed so that the smaller one of the two minimum values σUmin, σLmin in one cycle of the AC voltage increases.
Here, waveforms for the arm Nu are shown. The capacitor voltage sum VcA, the arm voltage command VA*, the upper margin σU, and the lower margin σL may re read as being replaced with a capacitor voltage sum VcANu, an arm voltage command VANu*, an upper margin σUNu, and a lower margin LNu.
In a case where positive power is transmitted from the AC grid 2 to the DC circuit 4, the capacitor voltage sum VcANu for the arm Nu pulsates as follows, in accordance with the above Expression (8). That is, on the capacitor voltage sum VcANu, negative-direction pulsation is superimposed in a range of 0≤θs−π/2, and positive-direction pulsation is superimposed in a range of n/2≤θs≤π.
If the adjustment phase θo for the zero-phase-sequence voltage command Vo* is 0, the arm voltage command VANu* has local maximum values at θs=π/3 and 2π/3. Therefore, around θs=π/3, the capacitor voltage sum VCANu is small and the range of voltage that can be outputted is small, so that the upper margin σUNu also becomes small. Around θs=2π/3, the capacitor voltage sum VcANu is great and the range of voltage that can be outputted is great, so that the upper margin σUNu also becomes great.
It is known that, in the arm voltage command VANu* shown by the above Expression (5), the necessary output voltage range can be effectively reduced in a case of a=1/6, and the arm voltage command VANu* can be represented by the following Expression (9).
VANu*=(Vdc/2)+Vp·sin(θs+θv)+(1/6)·Vp·sin 3(θs+θv+θo) (9)
On the basis of the above Expression (9), the upper margin σUNu and the lower margin σLNu for the arm Nu are calculated so as to be represented by the following Expression (10) and Expression (11).
σUNu=VcANu−((Vdc/2)+Vp·sin(θs+θv)+(1/6)·Vp·sin 3(θs+θv+θo)) (10)
σLNu=(Vdc/2)+Vp·sin(θs+θv)+(1/6)·Vp·sin 3(θs+θv+θo) (11)
The adjustment phase θo (=ωt) for the zero-phase-sequence voltage command Vo* changes, and if the above Expression (10) and Expression (11) are differentiated with respect to the adjustment phase θo, the following Expression (12) and Expression (13) are obtained. Here, VcANu in Expression (10) is hardly influenced by the zero-phase-sequence voltage command Vo* and therefore is neglected.
σUNu/dt=−(1/2)·Vp·cos 3(θs+θv+θo) (12)
σLNu/dt=(1/2)·Vp·cos 3(θs+θv+θo) (13)
From Expression (12) and Expression (13), as shown in
Conversely, when the adjustment phase θo is decreased, the upper margin σUNu around θs+θv=n/3 (region X1) decreases, and the upper margin σUNu around θs+V=2π/3 (region X2) increases. In addition, the lower margin σLNu around θs+θv=4π/3 (region X3) decreases, and the lower margin σLNu around θs+θv=5π/3 (region X4) increases.
As described above, through phase adjustment of the zero-phase-sequence voltage command Vo*, the minimum value σUmin of the upper margin σU obtained in the region X1 or the region X2 based on the reference phase (θs+θv) and the minimum value σLmin of the lower margin σL obtained in the region X3 or the region X4 are changed. At this time, by changing the adjustment phase θo so that the smaller one of the two minimum values σUmin, σLmin increases, the margin of the arm voltage command VA* with respect to the output voltage range is increased.
In the case shown in
In the region X1 and the region X3, the margin (upper margin σU or lower margin σL) increases with increase in the adjustment phase θo, and in the regions X2 and X4, the margin (upper margin σU or lower margin σL) decreases with increase in the adjustment phase θo. Therefore, a case where the margin of the arm voltage command VA* with respect to the output voltage range becomes greatest is when the smaller one of the minimum value of the upper margin σU in the region X1 and the minimum value of the lower margin σL in the region X3 coincides with the smaller one of the minimum value of the upper margin σU in the region X2 and the minimum value of the lower margin σL in the region X4.
Also in
In the case shown in
In the case shown in
In the case shown in
In the control device 20, in accordance with the capacitance of the DC capacitor 13, the AC voltage Vac, the DC voltage Vdc, or the like in the power converter 10, a circuit constant, and the operation state of the power converter 10, the above-described method is applied, thereby changing the adjustment phase θo so as to maximize the margin of the arm voltage command VA* with respect to the output voltage range.
Hereinafter, a specific adjustment method for the adjustment phase θo will be described with reference to FIG. 13.
As shown in
Here, the first local minimum value σUα is a value (magnitude: U1, around θs+θv=π/3 (region X1) in the first half (0≤θs+θv<π/2) of the first half cycle, and the first local minimum value σUβ is a value (magnitude: U2) around θs+θv=2π/3 (region X2) in the second half (π/2≤θs+θv<π) of the first half cycle. In addition, the second local minimum value σLα is a value (magnitude: L1) around θs+θv=4π/3 (region X3) in the first half (π≤θs+θv<3π/2) of the second half cycle, and the second local minimum value σLβ is a value (magnitude: L2) around θs+θv=5π/3 (region X4) in the second half (3π/2≤θs+θv<2π) of the second half cycle.
In the control device 20, first, the phase adjustment unit 32 of the zero-phase-sequence voltage command generation unit 25 monitors the upper margin σU and the lower margin σL, and detects the first local minimum values σUα, σUβ and the second local minimum values σLα, σLβ in one cycle of the AC voltage. As an example of the detection method, detection is performed when the upper margin σU or the lower margin σL h as become greater than the value in the previous cycle through consecutive three sampling cycles.
Next, from the detected first local minimum values σUα, σUβ and second local minimum values σLα, σLβ, a third local minimum value which is smallest in a range of the first half of the first half cycle and the first half of the second half cycle, is extracted. That is, the first local minimum value σUα and the second local minimum value σLα are compared with each other, and the smaller one is extracted as the third local minimum value.
Similarly, a fourth local minimum value which is smallest in a range of the second half of the first half cycle and the second half of the second half cycle, is extracted. That is, the first local minimum value σUβ and the second local minimum value CLS are compared with each other, and the smaller one is extracted as the fourth local minimum value.
In a case where one of the two local minimum values (first local minimum value σUα and second local minimum value σLα) to be used for extracting the third local minimum value is absent, the detected other local minimum value is used as the third local minimum value. Similarly, in a case where one of the two local minimum values (first local minimum value σUβ and second local minimum value σLβ) to be used for extracting the fourth local minimum value is absent, the detected other local minimum value is used as the fourth local minimum value.
Then, the third local minimum value is subtracted from the fourth local minimum value, to calculate a deviation. When the deviation is positive, the adjustment phase θo for the zero-phase-sequence voltage command Vo* is increased, and when the deviation is negative, the adjustment phase θo is decreased.
Here, in a case of U1<L2<U2<L1, the third local minimum value is U1 (first local minimum value σUα) and the fourth local minimum value is L2 (second local minimum value and the deviation is positive. Thus, the adjustment phase θo is increased.
Regarding increase/decrease of the adjustment phase θo, the adjustment phase θo may be changed on a certain value basis, or integral control or the like may be used. In addition, with a set value provided in advance, if the deviation obtained by subtracting the third local minimum value from the fourth local minimum value is smaller than the set value, the adjustment phase θo may be fixed without being changed. This can prevent the adjustment phase Oro from being frequently changed every time the arm voltage command VA* or the capacitor voltage sum VcA for the arm is slightly changed due to disturbance or the like. Therefore, a phenomenon in which new disturbance occurs due to frequent change of the adjustment phase θo so that the adjustment phase θo cannot converge, does not happen, and thus the adjustment phase θo can be reliably controlled.
In the present embodiment, as described above, by performing adjustment to increase/decrease the adjustment phase θo, phase adjustment of the zero-phase-sequence voltage command Vo* is performed. Thus, even if the capacitor voltage sum VcA pulsates, the margin of the arm voltage command VA* with respect to the output voltage range based on the capacitor voltage sum VcA can be increased, whereby utilization of DC voltage can be assuredly improved.
In the above phase adjustment, the case of adjusting the adjustment phase θo for the arm voltage command VA* for the U-phase lower arm Nu has been shown, and the zero-phase-sequence voltage command Vo* thus obtained is used in common for all the arms Pu, Nu, Pv, NV, Pw, NW. Under the assumption that the AC grid 2 is in a three-phase balanced state, calculation of the adjustment phase θo may be performed for at least one of the six arms Pu, Nu, Pv, Nv, Pw, Nw, whereby the margin of the arm voltage command VA* with respect to the output voltage range can be increased for all the arms and thus utilization of DC voltage can be assuredly improved.
In the above embodiment 1, the case where calculation of phase adjustment for the zero-phase-sequence voltage command Vo* is performed for one arm has been shown.
In embodiment 2, for all the six arms Pu, Nu, Pv, Nv, Pw, Nw included in the power converter 10, six values of each of the first local minimum values σUα, σUβ and the second local minimum values σLα, σLβ of the upper margin σU and the lower margin σL are detected per one cycle of the AC voltage. Then, among the six detected values of each local minimum value, the smallest first local minimum values σUαmin, σUβmin and second local minimum values σLαmin, σLβmin are extracted.
The subsequent method is the same as in the above embodiment 1. That is, among the first local minimum values σUαmin, σUβmin and the second local minimum values σLαmin, σLβmin, the first local minimum value σUαmin and the second local minimum value σLαmin are compared with each other, and the smaller one is extracted as the third local minimum value. Similarly, the first local minimum value σUβmin and the second local minimum value σLβmin are compared with each other, and the smaller one is extracted as the fourth local minimum value.
Then, the third local minimum value is subtracted from the fourth local minimum value, to calculate a deviation. When the deviation is positive, the adjustment phase θo for the zero-phase-sequence voltage command Vo* is increased, and when the deviation is negative, the adjustment phase θo is decreased.
As described above, by performing adjustment to increase/decrease the adjustment phase θo, phase adjustment of the zero-phase-sequence voltage command Vo* is performed. Thus, even if the capacitor voltage sum VcA pulsates, the margin of the arm voltage command VA* with respect to the output voltage range based on the capacitor voltage sum VcA can be increased, whereby utilization of DC voltage can be assuredly improved.
The AC grid 1 is basically in a three-phase balanced state, but in actuality, there is a case of not being in a three-phase balanced state. In addition, in a case where one of the converter cells 11 in any arm of the power converter 10 has failed, operation is performed by bypassing the failed converter cell 11 via the bypass switch 15. In this case, the number of the converter cells 11 that can perform output becomes smaller in the arm than in the other normal arms, so that the voltage that can be outputted decreases, and also, since the number of the DC capacitors 13 becomes small, pulsation of the capacitor voltage sum VcA also increases. Therefore, for the arm including the failed converter cell 11, the margin of the arm voltage command VA* with respect to the output voltage range tends to be smaller as compared to the other normal arms.
In the present embodiment, as described above, the upper margins σU and the lower margins σL for all the arms are monitored, and calculation of phase adjustment is performed on the basis of the first local minimum values, σUβ and the second local minimum values σLα, σLβ for all the arms. Thus, even in a case where the three-phase balance of the AC grid 2 is lost or a case where there are variations in the main circuit constants for the arms or there are variations in the numbers of normal converter cells 11 in the arms of the power converter 10, the margins of the arm voltage commands VA* with respect to the output voltage range can be increased in the entire power conversion device 1. That is, while the margin for one arm is improved, a problem such as deterioration in the margins for the other arms does not occur, and for all the arms, the arm voltage commands VA* can fall within the output voltage range with margins provided, whereby utilization of DC voltage can be assuredly improved with high reliability.
In the above embodiments 1 and 2, for phase adjustment of the zero-phase-sequence voltage command Vo*, the upper margin σU and the lower margin σL for the arm are monitored and the first local minimum values σUα, σUβ and the second local minimum values σLα, σLβ are detected in one cycle of the AC voltage.
In the present embodiment, extrema of the upper margin σU and the lower margin σL are not detected and the minimum value within a divided section is used. Matters other than phase adjustment of the zero-phase-sequence voltage command Vo* are the same as in the above embodiment 1.
In the control device 20, the phase adjustment unit 32 of the zero-phase-sequence voltage command generation unit 25 monitors the upper margin σU and the lower margin σL for the set arm in each PH section (PH=0 to 5), to detect the upper margin minimum value and the lower margin minimum value, and calculates and updates the adjustment phase θo at the timing T.
Hereinafter, an adjustment method for the adjustment phase θo according to the present embodiment will be described in detail with reference to
In
A timing θs+θv+θo=π/2 is a start point of the PH section at PH=0. For the arm Nu, a first instantaneous value σU and a second instantaneous value σU2 where the upper margin σU can become the minimum value are in the PH sections at PH=5 and PH=0, respectively, and a first instantaneous value σL1 and a second instantaneous value σL2 where the lower margin σL can become the minimum value are in the PH sections at PH=2 and PH=3, respectively. In the other PH sections at PH=1 and PH=4, the AC component of the arm voltage command VA* (lower margin σL) passes 0.
As described above, for each arm, the PH sections corresponding to the first instantaneous value σU1 and the second instantaneous value σU2 where the upper margin σU can become the minimum value, and the PH sections corresponding to the first instantaneous value σL1 and the second instantaneous value σL2 where the lower margin o-L can become the minimum value, are present.
From the above Expression (12) and Expression (13), when the adjustment phase θo is increased, the first instantaneous value σU1 of the upper margin σU increases and the second instantaneous value σU2 decreases. In addition, the first instantaneous value σL1 of the lower margin σL increases and the second instantaneous value σL2 decreases.
Conversely, when the adjustment phase θo is decreased, the first instantaneous value σU1 of the upper margin all decreases and the second instantaneous value σU2 increases. In addition, the first instantaneous value σL1 of the lower margin σL decreases and the second instantaneous value σL2 increases.
In each of the PH sections at PH=0 to 5, four instantaneous values, i.e., the first instantaneous value σU1 and the second instantaneous value σU2 of the upper margins σU and the first instantaneous value σL1 and the second instantaneous value σL2 of the lower margins σL axe each detected for different arms. With reference to
As described above, in each PH section, first to fourth arms which are arms for which the four instantaneous values, i.e., the first instantaneous value σU1 and the second instantaneous value σU2 of the upper margins σU and the first instantaneous value σL1 and the second instantaneous value σL2 of the lower margins σL are detected, are determined.
First, the phase adjustment unit 32 recognizes the PH section at PH=0, using the phase θs (=ωt) of the AC voltage Vac from the AC grid 2, phase shift θv between the AC voltage Vac and the AC component of the arm voltage command VA*, and the adjustment phase θo (step S00).
Next, by referring to the stored information in which the first to fourth arms are set, the instantaneous values (first instantaneous value σU1 and second instantaneous value σU2) of the upper margins σU for the first arm and the second arm and the instantaneous values (first instantaneous value σL1 and second instantaneous value σL2) of the lower margins σL for the third arm and the fourth arm are detected in the PH section at PH=0 (step S01).
Next, for θ*+=θs+θv+θo, whether PH=0 and θ*≥5π/6 are satisfied is determined, and in a case of No, the process returns to step S01 (step S02).
In a case of Yes in step S02, minimum values σU1min, σU2min, σL1min, σL2min of the instantaneous values σU1, σU2, σL1, σL2 in the PH section (PH=0) detected in step S01 are stored as a first value, a second value, a third value, and a fourth value (step S03). Then, the PH section is set to the section at PH=1 (step S04).
Next, by referring to the stored information in which the first to fourth arms are set, the instantaneous values (first instantaneous value σU1 and second instantaneous value σU2) of the upper margins σU for the first arm and the second arm and the instantaneous values (first instantaneous value σL1 and second instantaneous value σL2) of the lower margins σL for the third arm and the fourth arm are detected in the PH section at PH=1 (step SOS).
Next, for θ*=θs+v+θo, whether PH=1 and θ*≥7π/6 are satisfied is determined, and in a case of No, the process returns to step S05 (step S06).
In a case of Yes in step S06, minimum values σU1min, σU2min, σL1min, σL2min of the instantaneous values σU1, σU2, σL1, σL2 in the PH section (PH=1) detected in step SOS are stored as a first value, a second value, a third value, and a fourth value (step S07). Then, the PH section is set to the section at PH=2 (step S08).
Next, the above four steps from step 04 to step 08 are repeated while PH is increased by 1 for each time and the range of θ* is also increased by π/3 for each time, until the PH section is set to the section at PH=5 (step S20). Then, in the PH section at PH=5, similarly, the instantaneous values σU1, σU2, σL1, σL2 are detected (step S21).
Next, for θ*=θs+θv+θo, whether PH=5 and θ*≥π/2 are satisfied is determined, and in a case of No, the process returns to step S21 (step S22).
In a case of Yes in step S22, minimum values σU1min, σU2min, σL1min, σL2min of the instantaneous values σU1, σU2, σL1, σL2 in the PH section (PH=5) detected in step S21 are detected and stored as a first value, a second value, a third value, and a fourth value (step S23). Then, the adjustment phase θo is calculated and updated on the basis of the minimum values σU1min, σU2min, σL1min, σL2min stored in the respective PH sections (step S24), and the process returns to step S00 to set the PH section to the section at PH=0.
The calculation of the adjustment phase θo in step S24 is performed as follows. Each minimum value of six data of each of σU1min, σU2min, σL1min, σL2min in all the PH sections at PH=0 to 5 is defined as a first minimum value σU1m, a second minimum value σU2m, a third minimum value σL1m, and a fourth minimum value σL2m, respectively. Then, the first minimum value σU1m and the third minimum value σL1m are compared with each other and the smaller one is extracted as σUL1, and similarly, the second minimum value σU2m and the fourth minimum value σL2m are compared with each other and the smaller one is extracted as σUL2.
Then, the extracted σUL1 is subtracted from the extracted σUL2, to calculate a deviation. When the deviation is positive, the adjustment phase θo for the zero-phase-sequence voltage command Vo* is increased, and when the deviation is negative, the adjustment phase θo is decreased.
Regarding increase/decrease of the adjustment phase θo, the adjustment phase θo may be changed on a certain value basis, or integral control or the like may be used.
Here, with a set value provided in advance, if the deviation obtained by subtracting the extracted σUL1 from the extracted σUL2 is smaller than the set value, the adjustment phase θo may be fixed without being changed. This can prevent the adjustment phase θo from being frequently changed every time the arm voltage command VA* or the capacitor voltage sum VcA for the arm is slightly changed due to disturbance or the like, whereby the adjustment phase θo can be reliably controlled.
In the present embodiment, as described above, by performing adjustment to increase/decrease the adjustment phase θo, phase adjustment of the zero-phase-sequence voltage command Vo* is performed. Thus, even if the capacitor voltage sum VcA pulsates, the margin of the arm voltage command VA* with respect to the output voltage range based on the capacitor voltage sum VcA can be increased, whereby utilization of DC voltage can be assuredly improved.
In addition, using instantaneous value information of the upper margins σU and the lower margins σL for all the arms, the adjustment phase θo is adjusted to be increased/decreased so that the smallest margin (σU, σL) increases. Thus, even in a case where the three-phase balance of the AC grid 2 is lost or a case where there are variations in the main circuit constants for the arms or there are variations in the numbers of normal converter cells 11 in the arms of the power converter 10, the margins of the arm voltage commands VA* with respect to the output voltage range can be increased in the entire power conversion device 1.
In addition, in the present embodiment, on the basis of information of the first to fourth arms set in advance for each PH section, the first minimum value σU1m and the second minimum vale σU2m of the upper margins σU and the third minimum value σL1m and the fourth minimum value σL2m of the lower margins σL are detected, and phase adjustment of the zero-phase-sequence voltage command Vo* is performed so as to increase the margin of the arm voltage command VA* with respect to the output voltage range.
Therefore, it is not necessary to detect the local minimum values σUα, σUβ, σLα, σLβ of the upper margin σU and the lower margin σL as shown in the above embodiments 1 and 2. Presence/absence or the number of the local minimum values of each of the upper margin σU and the lower margin CL in a range of interest is not constant, and detection for the local minimum values is performed by recognizing the crests and the troughs of the waveforms of the upper margin σU and the lower margin σL and therefore is complicated.
In the present embodiment, without using the local minimum values of the upper margin σU and the lower margin σL, phase adjustment of the zero-phase-sequence voltage command Vo* can be performed easily and reliably.
In the present embodiment, as in the above embodiment 3, on the basis of information of the first to fourth arms set in advance for each PH section, the first minimum value σU1m and the second minimum value σU2m of the upper margins σU and the third minimum value σLim and the fourth minimum value σL2m of the lower margins σL are detected to perform adjustment to increase/decrease the adjustment phase θo.
In the above embodiment 3, the adjustment phase θo is updated at the timing T for every cycle of the AC grid 2, whereas in the present embodiment, a period in which detection for the instantaneous values σU1, σU2, σL1, σL2 of the upper margins σU and the lower margins σL is stopped is provided and the update interval for the adjustment phase θo is expanded.
Hereinafter, update of the adjustment phase θo in embodiment 4 will be described on the basis of comparison with the above embodiment 3.
The first update timing T is θs+θv+θo=π/2, and in the graphs, θs+θv+θo (θo=θ1) before update of the adjustment phase θo and θs+θv+θo (θo=θ1+Δθo) after the update are shown.
In the above embodiment 3 shown in
On the other hand, in embodiment 4 shown in
First, a determination signal Ca1 which is 0 or 1 is set to an initial value 1 (step S100).
Next, whether the determination signal Ca1 is 0 or 1 is determined (step S101). If the determination signal Ca1 is 1, the processing from step S00 to step S24 shown in the above embodiment 3 is performed to calculate and update the adjustment phase θo on the basis of the minimum values σU1min, σU2min, σL1min, σL2min stored in the respective PH sections (step S102), and the determination signal Ca1 is set to 0 (step S103), to return to step S101.
In step S101, if the determination signal Ca1 is 0, the PH section is set to the section at PH=0 (step S104). Then, for θ*=θs+θv+θo, whether PH=0 and θ*≥5π/6 are satisfied is determined, and this determination processing is repeated until the determination indicates Yes (step S105). In a case of Yes, the PH section is set to the section at PH=1 (step S106).
Then, for θ*=θ*+θv+θo, whether PH=1 and θ*≥7π/6 are satisfied is determined, and this determination processing is repeated until the determination indicates Yes (step S107). In a case of Yes, the PH section is set to the section at PH=2 (step S108).
The above two steps of step 107 and step 108 are repeated while PH is increased by 1 for each time and the range of θ* is also increased by π/3 for each time, until the PH section is set to the section at PH=5 (step S120). Then, for θ*=θs+θv+θo, whether PH=5 and θ*≥π/2 is satisfied is determined, and this determination process is repeated until the determination indicates Yes (step S121). In a case of Yes, the determination signal Ca1 is set to 1 (step S122), to return to step S101.
As described above, using the determination signal Ca1, if the determination signal Ca1 is 1, the continuous detection period 41 is set, and if the determination signal Ca1 is 0, the detection stop period 42 is set. Thus, the continuous detection period 41 for the instantaneous values σU1, σU2, σL1, σL2 can be a period that continues over one cycle, whereby the first minimum value σU1m and the second minimum value σU2m of the upper margins σU and the third minimum value σL1m and the fourth minimum value σL2m of the lower margins σL are reliably acquired and reliability of phase adjustment is improved.
In the above description, the continuous detection period 41 is set as one cycle of the AC voltage. However, the continuous detection period 41 may be a cycle that is an integer multiple of the AC voltage cycle.
If the continuous detection period 41 is present at least over one cycle of the AC voltage, the detection stop period 42 may not necessarily be provided.
Although the disclosure is described above in terms of various exemplary embodiments and implementations, it should be understood that the various features, aspects, and functionality described in one or more of the individual embodiments are not limited in their applicability to the particular embodiment with which they are described, but instead can be applied, alone or in various combinations to one or more of the embodiments of the disclosure.
It is therefore understood that numerous modifications which have not been exemplified can be devised without departing from the scope of the present disclosure. For example, at least one of the constituent components may be modified, added, or eliminated. At least one of the constituent components mentioned in at least one of the preferred embodiments may be selected and combined with the constituent components mentioned in another preferred embodiment.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2020/028887 | 7/28/2020 | WO |
