JP6293423B2 - Multi-level power converter and control method thereof - Google Patents

Description

  The present invention relates to a multilevel power conversion device that uses three single-phase inverter outputs as three-phase outputs.

  In recent years, multi-level inverters that Y-connect single-phase inverters are widely used in industrial application fields. This type of multi-level inverter has the advantage that it can be made relatively multi-level relatively easily and a single-phase inverter can be realized as a module with a simple configuration. However, in a multilevel inverter using a conventional method, control for reducing the maximum value of power of each single-phase inverter is not always performed. For this reason, not only the apparatus capacity | capacitance of each single phase inverter but the transformer and rectifier for producing | generating the power supply given to each single phase inverter become large.

  As one of conventional techniques related to such a problem, there is a method of effectively using a DC voltage of a single-phase inverter by using a third harmonic as a modulation factor (see Non-Patent Document 1).

Iwaji, Okuyama, Kaneko, Okamatsu, “PWM Control Method for High Voltage Direct Inverters”, The Institute of Electrical Engineers of Japan, Journal D, Vol. 121 (2001), No. 4, p. 476-483

  However, since the original purpose of the power conversion device described in Non-Patent Document 1 is to improve the voltage utilization factor, the load current is unbalanced, not a sine wave, or the load power factor varies. In such a case, the output power of each single-phase inverter cannot be reduced. As a result, there is a problem that the output power of the three single-phase inverters becomes uneven and the required rated capacity of the device increases.

  The present invention has been made to solve the above-described problems, and in a three-phase multilevel power conversion device including three single-phase inverters, the output power of the single-phase inverter is equalized, The purpose is to reduce the rated capacity.

  In order to solve the above problems, a multilevel power conversion device according to an aspect of the present invention includes three single-phase inverters, one output terminal of each single-phase inverter is connected to a load, and the three Compared to the case where the other output terminals of the single-phase inverter are connected to each other to form a neutral point, and a three-phase power converter, and the output power of each single-phase inverter does not change the potential of the neutral point And a control unit that controls the output voltage of each single-phase inverter so that the potential of the neutral point changes so as to decrease.

  Here, the “output voltage of the single-phase inverter” is the sum of the phase voltage that the single-phase inverter wants to apply to the load and the neutral point potential. This is also equal to the difference between the potential of one output terminal of the single-phase inverter and the potential of the other Y-connected output terminal.

  In general, three-phase load power is determined by line voltage, and there is a degree of freedom with respect to neutral point potential. In a so-called Y connection (star connection) multilevel power converter, the load sharing of the three single-phase inverters can be changed using this degree of freedom. Therefore, according to the above configuration, the control unit uses an appropriate optimization method so that the output power of each single-phase inverter is reduced as compared to the case where the neutral point potential is not changed. Since the output voltage of each single-phase inverter is controlled so that the potential changes, the output power of the single-phase inverter can be equalized and the rated capacity of the device can be reduced.

  The control unit may control the output voltage of each single-phase inverter so that the potential at the neutral point changes so that the maximum value of the output power of the three single-phase inverters is minimized.

  According to the above configuration, the neutral point potential can be optimized so that the maximum value of the output power of each single-phase inverter is reduced as compared with the case where the neutral point potential is not changed.

  Three phase current detectors that respectively detect phase currents that are output currents of one output terminal of the three single-phase inverters, and the control unit includes a phase voltage that the single-phase inverter wants to give to the load and Based on the phase current detected by each of the three phase current detectors, the maximum value of the output power of the three single-phase inverters is minimized, or the sum of squares of the DC section currents of the three single-phase inverters is minimized. Command value for controlling the output voltage of each single-phase inverter by calculating the neutral point potential which is the potential of the neutral point so as to be, and adding the calculated neutral point potential to the phase voltage May be generated.

  According to the above configuration, a configuration that suitably optimizes the neutral point potential can be specifically realized.

  The controller further satisfies the constraint that the absolute value of the sum of the phase voltage that the single-phase inverter wants to apply to the load and the neutral point potential is less than or equal to the input voltage of each single-phase inverter. A point potential may be calculated.

  According to the said structure, it can prevent that the output voltage of a single phase inverter is cut by making the input voltage of a single phase inverter into an upper limit or a minimum.

  When the neutral point potential does not satisfy the constraint condition, the control unit sets a close value between the maximum value and the minimum value satisfying the constraint condition, and the maximum value of the output power of the three single-phase inverters. It is good also as a value which becomes the minimum or the sum of squares of the DC part current of the three single phase inverters becomes the minimum.

  According to the above configuration, even when the neutral point potential does not satisfy the constraint condition, it is possible to suitably prevent the output voltage of the single-phase inverter from being cut with the input voltage of the single-phase inverter as the upper limit or the lower limit.

  The control method of the multilevel power conversion device according to another aspect of the present invention includes three single-phase inverters, one output terminal of each single-phase inverter is connected to a load, and the other of the three single-phase inverters The output terminals of the single-phase inverter are connected to each other to form a neutral point. The output voltage of each single-phase inverter is controlled so that the potential at the neutral point changes so as to decrease as compared with the case where the potential is not changed.

  According to the above configuration, the neutral point potential is reduced by the control unit using an appropriate optimization method so that the output power of each single-phase inverter is reduced as compared with the case where the neutral point potential is not changed. Since the output voltage of each single-phase inverter is controlled so as to change, the output power of the single-phase inverter can be equalized, and the required rated capacity of the device can be reduced.

  According to the present invention, in a three-phase output voltage converter having three single-phase inverters, it is possible to equalize the output power of the single-phase inverter and reduce the rated capacity of the device.

It is a block diagram which shows the whole structure of the power converter device which concerns on Embodiment 1 of this invention. It is a circuit diagram which shows the structure of the power converter of the power converter device of FIG. It is a block diagram which shows the structure of the control part of the power converter device of FIG. It is a vector diagram which shows the relationship between the phase voltage with respect to the DC voltage of the single phase inverter of FIG. 2, and a neutral point electric potential. It is a flowchart which shows an example of the flow of the arithmetic processing in the control part of FIG. It is the graph which showed the electric power of each phase with respect to the neutral point electric potential in the Y connection point of the power conversion part of FIG. It is a vector diagram which shows the relationship between the phase voltage and neutral point potential by the minimax method in the case of the modulation factor 1.0. It is the graph which showed the time change of the phase voltage, neutral point potential, and phase electric power by the minimax method in the case of the modulation factor 0.8. It is the graph which showed the time change of the phase voltage by the minimax method in the case of the modulation factor 1.0, a neutral point electric potential, and phase electric power. It is the graph which showed the time change of the phase voltage, neutral point potential, and phase electric power by the least square method in the case of the modulation factor 0.8. It is the graph which showed the time change of the phase voltage, neutral point potential, and phase electric power by the least square method in case of the modulation factor 1.0. It is the graph which showed the average value of the square sum of direct-current power by the minimax method and the least square method in the case of the modulation factor 1.0. It is the graph which compared the square sum of the electric power and peak electric power by the minimax method and the least square method. It is the graph which showed the phase voltage and neutral point potential by the minimax method in the case of the modulation factor 0.8 and the power factor 0.5.

  Embodiments of the present invention will be described with reference to the drawings. Below, the same code | symbol is attached | subjected to the element which is the same or it corresponds through all the drawings, and the overlapping description is abbreviate | omitted.

(Embodiment 1)
[Constitution]
FIG. 1 is a block diagram showing an overall configuration of the power conversion device according to the first embodiment. The use of power converter 100 of the first embodiment is not particularly limited. The power conversion apparatus 100 is mounted in, for example, an on-board power system, converts AC power supplied from an on-board electrical system into high voltage variable voltage variable frequency AC power, and drives a marine propulsion motor. As shown in FIG. 1, the power conversion device 100 is connected to a three-phase AC power source 1 via an input-side three-phase transformer 2 and connected to a three-phase AC motor (load) 4 via an output line 30. The power conversion unit 3 and the control unit 7 that performs PWM control on the power conversion unit 3 are provided.

  The power conversion unit 3 includes three single-phase inverters 3a to 3c of a phase to c phase, and one output terminal of each of the single phase inverters 3a to 3c is connected to the load 4 via the output line 30, and the other The output terminals have three-phase outputs that are Y-connected to each other to form a neutral point N. In the present embodiment, the output line 30 is provided with the current sensor 5. Current sensor 5 includes three phase current detectors that respectively detect a phase current that is an output current of one output terminal of three single-phase inverters 3a to 3c. The current sensor 5 outputs the detected phase current to the control unit 7.

The control unit 7 controls the single-phase inverters 3a to 3c so that the potential at the neutral point N changes so that the output power of the single-phase inverters 3a to 3c decreases as compared with the case where the potential at the neutral point N is not changed. The output voltage of 3c is controlled. In the present embodiment, the control unit 7 outputs the outputs of the single-phase inverters 3a to 3c so that the potential at the neutral point N changes so that the maximum value of the output power of the three single-phase inverters 3a to 3c is minimized. The voltages v ^* a, v ^* b, and v ^* c are controlled.

Here, output voltages of single-phase inverters 3a to 3c in the present embodiment will be described. FIG. 7 is a vector diagram showing the relationship of the output voltage by the minimax method of the present embodiment described later. FIG. 7A is a vector diagram of phase voltages va, vb, and vc that the single-phase inverter wants to apply to the load when the neutral point potential v0 is zero. FIG. 7B shows vectors of phase voltages va, vb, vc and output voltage v ^* a that the single-phase inverter wants to apply to the load when the neutral point potential v0 is zero or more. As shown in FIG. 7B, for example, the output voltage v ^* a of the single-phase inverter 3a is calculated by a phase voltage va that the single-phase inverter 3a wants to apply to the load 4 and a neutral point potential calculation unit 72 described later. It is expressed as the sum of neutral point potentials v0. The neutral point potential v0 is expressed by the difference between the potential at the neutral point N and a predetermined reference potential. Thus, the output voltages v ^* a, v ^* b, and v ^* c of the single-phase inverters 3a to 3c are the neutral voltages of the phase voltages va, vb, and vc that the single-phase inverters 3a to 3c want to apply to the load 4. The sum of potentials v0. Further, the predetermined reference potential here is the neutral point potential of the load 4, and therefore the neutral point potential v0 here is the neutral point potential of the load 4 and the single-phase inverters 3a to 3c. This is a difference from the sex point N. Further, the phase voltages va, vb, and vc are, for example, constant frequency signals, and are input to the control unit 7 from the outside in the present embodiment.

  Next, the configuration of single phase inverters 3a to 3c will be specifically described. Since the single-phase inverters 3a to 3c have the same configuration, only the single-phase inverter 3a will be described here. FIG. 2 is a circuit diagram showing a configuration of the power conversion unit 3 of FIG. As shown in FIG. 2, the single-phase inverter 3 a includes a forward converter 11 that converts an AC voltage supplied from the three-phase transformer 2 into a DC voltage, and a smoothing that smoothes the DC voltage converted by the forward converter 11. The capacitor | condenser 12 and the inverse converter 13 which outputs the electric power which carried out pulse width modulation (PWM) are provided. Note that the single-phase inverter 3a may be configured only by the inverse converter 13, and the forward converter 11 may configure a rectifier separate from this. In this case, the direct current of the single-phase inverter 3a (used in the least square method described later) is an input current to the single-phase inverter 3a.

  The forward converter 11 is composed of six rectifier diodes. The forward converter 11 may be a multi-pulse rectifier circuit or a PWM converter. The inverse converter 13 includes four switching elements Q1 to Q4 each having a diode connected in antiparallel. The inverse converter 13 is formed of a semiconductor element, and, for example, an IGBT is used for each of the switching elements Q1 to Q4. As another example, a single-phase multilevel inverter may be used as the single-phase inverter.

  In this embodiment, the current sensor 5 subtracts the phase currents ia and ib, the phase current detector that detects the phase current ia of the a phase, the phase current detector that detects the phase current ib of the b phase. The subtractor 51 calculates the c-phase phase current ic. In this way, the current sensor 5 outputs the detected phase currents ia, ib, ic to the control unit 7. Moreover, as another form, the current sensor 5 may be comprised from the phase current detector which detects all the phase currents of a phase-c phase.

Next, the configuration of the control unit 7 will be described with reference to the block diagram of FIG. As shown in FIG. 3, the control unit 7 includes three phase voltages va, vb, vc that the single-phase inverter wants to apply to the load 4 and phase currents ia, ib, ic detected by the current sensor 5, respectively. The neutral point potential v0, which is the potential at the neutral point N at which the maximum value of the output power of the single-phase inverters 3a to 3c is minimum, is calculated, and the calculated neutral point potential v0 is converted into the phase voltages va, vb, vc. A voltage command generator 70 that generates a command value for controlling the output voltages v ^* a, v ^* b, and v ^* c of the single-phase inverters 3a to 3c by addition, and a PWM signal based on the voltage command signal And a PWM signal generation unit 7 that generates and outputs a PWM signal to the power conversion unit 3.

In the present embodiment, the voltage command generation unit 70 uses the neutral point potential calculation unit 72 to calculate the three single-phase inverters 3a to 3c based on the phase voltages va, vb, vc and the phase currents ia, ib, ic. The neutral point potential v0, which is the voltage at the neutral point N at which the maximum value of the output power is minimum, is calculated, and the phase voltage va that the single-phase inverter wants to apply the neutral point potential v0 to the load by the three adders 73. , Vb, and vc to generate command values for controlling the output voltages v ^* a, v ^* b, and v ^* c of the single-phase inverters 3a to 3c.

The PWM signal generation unit 71 generates a PWM signal based on the command signals of the output voltages v ^* a, v ^* b, and v ^* c output from the three adders 73, and outputs the PWM signal to the power conversion unit 3. To do. Although the PWM signal may be generated by a known method, in this embodiment, the PWM signal generation unit 71 includes a sign inverter 74 that inverts the sign of the voltage command signal and a triangular wave generation circuit that generates a triangular wave that is a carrier wave. 78, a comparator 75 that compares two input signals and outputs a pulse according to the magnitude, a logic inverter 76, and a delay circuit 77.

With such a configuration, the control unit 7 outputs a control signal (PWM signal) to, for example, the single-phase inverter 3 a in the power conversion unit 3. The PWM signal is input to the control terminals (for example, the gate terminals of the IGBT) of the switching elements Q1 to Q4 of the single-phase cell inverter 3a, and the single-phase inverter 3a functions as an inverter by turning on / off the switching elements Q1 to Q4. Let The control unit 7 causes the other single-phase inverters 3b and 3c to function as inverters in the same manner. Thereby, the power conversion unit 3 is PWM-controlled so as to output voltages corresponding to the voltage command values v ^* a, v ^* b, and v ^* c calculated by the control unit 7.

  In the present embodiment, a part of the control unit 7 is configured by an arithmetic unit such as a field programmable gate array (FPGA), a programmable logic controller (PLC), a microcontroller, etc., and in particular, a neutral point potential arithmetic unit. Reference numeral 72 denotes a functional block realized by executing a program built in the arithmetic device. The control unit 7 is assumed to have a function of receiving a detection signal of the current sensor 5 in addition to the voltage command generation unit 70 and the PWM signal generation unit 71.

[Knowledge that became the basis of the present invention]
The inventors have described the output voltage of the power converter 3 in which the other output terminals of the three single-phase inverters 3a to 3c shown in FIGS. 1 and 2 are Y-connected to form a neutral point N. investigated. Assuming that phase voltages va, vb, vc (va + vb + vc = 0) to be applied to the load 4 of the power conversion unit 3 and the neutral point potential v0 of each single-phase inverter 3a-3c, the output voltage v of each single-phase inverter 3a-3c. ^{* A} , v ^* b, and v ^* c can be represented by Formula (1).

Here, the voltage of the DC voltage part of each of the single-phase inverters 3a to 3c is v _DC , and the DC voltages v _DC are equal to each other. The output voltage of each of the single-phase inverters 3a to 3c is limited as shown in Expression (2).

That equation (2), the absolute value of the output voltage ^{v * a, v * b,} v * c which is output to one end of the output terminal of each single-phase inverter, or less input voltage v _DC for each single-phase inverter Represents a constraint condition.

Figure 4 is a vector diagram showing the relationship between the phase voltage want single phase inverter 3 a to 3 c is applied to the load 4 va, vb, vc and neutral point potential v0 for the DC voltage _{v DC} single-phase inverter 3 a to 3 c. 4A and 4B show the case where the neutral point potential v0 = 0, that is, the neutral point potential seen from the inverters 3a to 3c and the neutral point potential seen from the load 4 match. , Vector diagrams when the absolute values of the phase voltages va, vb, vc are minimum and maximum are respectively shown.

With respect to the phase voltage shown in FIG. 4A, the absolute values of the phase voltages va, vb, and vc are allowed up to 2 / √3 × v _DC = 1.15 × v _DC as shown in FIG. 4B. It is known. That is, the maximum phase voltage is 1.15 times the DC voltage v _DC of the single-phase inverter.

4 (c) and 4 (d) show vector diagrams when the neutral point potential v0 is varied when the maximum phase voltage is output. At the phase angles as shown in FIGS. 4C and 4D, the neutral point potential v0 when the maximum phase voltage is output is from (1 / √3−1) × v _DC to (1 −2 / √3) × v _DC . That is, when FIG. 4C and FIG. 4D are compared, there is still a degree of freedom in the neutral point potential v0 depending on the electrical angle even when the maximum voltage is being output. I understand. In general, three-phase load power is determined by line voltage, and there is a degree of freedom with respect to neutral point potential. Further, in the so-called Y connection (star connection) multilevel power conversion device 100, the load sharing of the three single-phase inverters can be changed using this degree of freedom. Therefore, the present inventors have focused on the degree of freedom of the neutral point potential v0 and have come up with a control method for leveling the instantaneous power of each single-phase inverter in the power conversion unit 3.

[Minimax method]
Hereinafter, a control method for leveling the peak power in the power conversion apparatus 100 according to the first embodiment will be described with reference to FIGS. 5 and 6. This control method calculates a neutral point potential v0 that minimizes the maximum value of the output power of the three single-phase inverters 3a to 3c based on the concept of linear programming (hereinafter referred to as a minimax method). Also called).

  FIG. 5 is a flowchart illustrating an example of the flow of the instantaneous power leveling process of the single-phase inverters 3 a to 3 c by the control unit 7. It should be noted that the processing executed in this flowchart is performed every fixed control cycle. As shown in FIG. 5, first, the neutral point potential calculation unit 72 includes three phase voltages va, vb, vc and phase currents ia, ib, ic that the single-phase inverters 3a-3c want to apply to the load. A neutral point potential v0 that is a voltage at the neutral point N at which the maximum value of the output power of the single-phase inverters 3a to 3c is minimized is calculated (step 1). Here, the output power of each phase is expressed by Equation (3). In the present embodiment, the neutral point potential calculation unit 72 calculates a neutral point potential v0 that minimizes Expression (4) representing the maximum value of the power of each phase represented by Expression (3).

Here, an example of the solution of the equation (4) executed in step 1 will be described with reference to FIG. FIG. 6 is a graph showing the power of each phase with respect to the neutral point potential at the Y contact of the power conversion unit 3. The vertical axis represents the power P of each phase, and the horizontal axis represents the neutral point potential v0. FIG. 6A shows power P combining all the phases a to c, FIG. 6B shows power P _ab combined with a phase and b phase, and FIG. 6C shows combination of b phase and c phase. The electric power P _bc and FIG. 6D show the electric power P _ca combining the c phase and the a phase, respectively, as an absolute value function. In FIG. 6A, three broken lines (solid line, one-dot chain line, and broken line) obtained from the absolute value function have a total of six intersections, and one of them is a solution. Now, consider the maximum absolute value of the power of the two phases given by the following equation (5) for each of the three phase combinations (a, b), (b, c), and (c, a).

The neutral point potential v0 that gives the minimum of P _xy is defined as v _xy . Here, the subscripts x and y represent one of a, b, and c, respectively.

It is relatively easy to minimize Equation (4) with respect to the neutral point potential v0, resulting in the problem of obtaining the intersection of four straight lines as shown in FIGS. 6 (b) to 6 (d). . When the intersection point gives the minimum value of P _xy is different sign of the slope of the two straight lines. If this property is used, the solution of equation (5) is as shown in equation (6) below. Note that because both the numerator and denominator have a first-order term of current, the current is determined only by its phase relationship.

Here, the relationship between the above three combinations of phases is examined. For example, considering phase combinations (a, b) and (b, c), P _ab > P _bc and the corresponding neutral point potential v 0 is v _ab and v _bc . The neutral point potential v _bc gives the minimum power in the case of the combination of the b phase and the c phase, but the value of the excluded a phase power | (va + vbc) × ia | is larger than P _{ab, and} thus is obtained. It is excluded from the solution candidates. Eventually, v _xy that gives the maximum P _xy among the three combinations is the solution of equation (3).

  Next, the neutral point potential calculation unit 72 determines whether or not the constraint condition of Expression (2) is satisfied (step 2). Here, when the constraint condition of Expression (2) is rewritten for v0, the following Expression (7) is obtained. In the present embodiment, it is determined whether or not the neutral point potential v0 is within the range of Expression (7).

  Next, when the neutral point potential v0 satisfies the condition of the expression (7), the neutral point potential calculation unit 72 sets the neutral point potential v0 calculated in step 1 as an optimum value (step 3).

Further, when the condition of the equation (7) is not satisfied, the neutral point potential calculation unit 72 outputs the closest value among the maximum value and the minimum value that satisfy the condition of the equation (7) as the outputs of the three single-phase inverters. The maximum power value is set to a minimum value (step 4). That is, since the equation (4) is a downward convex function, when the neutral point potential that minimizes the equation (4) deviates from the range indicated by the equation (7), it is within the range of the equation (7). Select the closest value. In the present embodiment, the neutral point potential v0 is the maximum value v _DC -max (va, vb, vc) of the equation (7), and the minimum value -v _DC -max (va, vb, vc) of the equation (7). ) Is selected. Thereby, even when the neutral point potential v0 does not satisfy the constraint condition (7), it is possible to suitably prevent the output voltage of the single-phase inverter from being cut with the input voltage of the single-phase inverter as the upper limit or the lower limit.

Next, the voltage command generator 70 adds the neutral point potential v0 calculated by the neutral point potential calculator 72 and the phase voltages va, vb, vc, and outputs the output voltages v ^* a, v ^* b, v ^* c. Command signal is output (step 5).

Next, the PWM signal generation unit 71 generates a PWM signal based on the voltage command signal, and outputs the PWM signal to the power conversion unit 3 (step 6). As a result, the power conversion unit 3 is PWM-controlled so as to output the voltages v ^* a, v ^* b, and v ^* c corresponding to the voltage command signal calculated by the control unit 7.

According to the above configuration, the control unit 7 uses the optimization method called the minimax method, and the output power v ^* a, v ^* b, v ^* c of each of the single-phase inverters 3a to 3c is the potential at the neutral point N. Since the output voltage of each single-phase inverter is controlled so that the potential of the neutral point N changes so as to decrease compared with the case where the voltage is not changed, the output power of the single-phase inverter is equalized, and the rated capacity of the device Can be reduced.

  As described above, according to the present embodiment, the maximum instantaneous power for each phase is selected for the neutral point potential selection method that has been often discussed from the viewpoint of improving the voltage utilization rate and reducing the distortion rate. This has the effect of minimizing. In this respect, the peak power leveling control of the present embodiment is essentially different from the control for other purposes, and the power of the three-phase output having three Y-connected single-phase inverters. This is an original configuration unique to the converter.

(Embodiment 2)
[Least square method]
Next, a second embodiment of the present invention will be described. The description of the configuration common to the first embodiment is omitted, and only the configuration that is different will be described. The control method for leveling the peak power according to the second embodiment is based on the idea of the least square method, and the neutral point potential that minimizes the sum of squares of the DC currents of the three single-phase inverters 3a to 3c. v0 is calculated (hereinafter also referred to as a least square method).

  That is, in the power conversion device of the second embodiment, when compared with the first embodiment, the neutral point potential calculation unit 72 satisfies the constraint condition of the expression (5) in step 1 of FIG. 5 in the first embodiment. The difference is that the optimum value of the neutral point potential v0 to be satisfied is set to the optimum value of the neutral point potential v0 at which the sum of squares of the direct currents of the three single-phase inverters 3a to 3c is minimized.

For example, the current flowing in the DC voltage portion of the a-phase single-phase inverter 3a is (va + v0) ia / v _DC . Here, as in equation (8), a square sum W of direct currents of the three single-phase inverters 3a to 3c is considered.

  Here, since the equation (8) is a downward convex quadratic function with respect to v0, the neutral point potential calculator 72 sets the neutral point potential v0 that minimizes the equation (8) to the equation (9). ).

  Since the square sum of the direct currents of the three single-phase inverters corresponds to the power loss due to the wiring resistance in the wiring of the direct current part, the above configuration can minimize the power loss. In addition, when a transformer is connected to the inverter as in this embodiment, the current flowing through the secondary winding of the transformer is proportional to the direct current, so the heat generated by the winding resistance in the transformer winding. Can be minimized. For the same reason, the copper loss of the secondary winding of the transformer 2 and the loss of the smoothing capacitor 11 can be minimized.

  In addition, since the expression (8) is also a downward convex function like the expression (3) of the minimax method in the first embodiment, the neutral point potential that minimizes the expression (8) is expressed by the expression (5). When deviating from the range shown, the closest value within the range of the formula (5) may be selected as in the first embodiment.

[simulation result]
In order to verify the effects of the above embodiments, the present inventors performed simulations using the minimax method and the least square method. In the simulation, it is assumed that the load currents ia, ib, and ic are balanced three-phase AC currents, and the phase voltages va, vb, vc, neutral point potentials v0, DC unit power that the single-phase inverters 3a to 3c want to apply to the load 4 are assumed. Of the squared sum W and instantaneous power P for each phase were calculated. 8 to 12 show graphs of calculation results when the load power factor is 1.0. Since the waveforms are complicated, only the two phases are shown. The modulation factor M in the figure is defined by a value obtained by dividing the phase voltage applied to the load by the DC voltage, and takes a value from zero to 2 / √3. In the figure, the range of the solution indicated by the condition of Expression (7) is also shown. When priority is given to improving the voltage utilization factor, the neutral point potential is often selected to be close to zero within the range of the equation (7). On the other hand, the calculation results show that the neutral point potential moves very greatly in order to minimize the phase power.

  First, the simulation result by the minimax method is shown. FIG. 8A and FIG. 8B are graphs showing temporal changes in the phase voltage and the neutral point potential when the modulation factor M according to the minimax method is 0.8. FIG. 9A and FIG. 9B are graphs showing temporal changes in the phase voltage and the neutral point potential when the modulation factor M by the minimax method is 1.0. As shown in the graphs of FIGS. 8 and 9, in the result of the minimax method, the neutral point potential has a triangular wave shape from the time when the modulation rate is small. When the modulation rate is large, it largely fluctuates within the limited range while maintaining the triangular wave shape. Looking at the instantaneous value of the phase power, in the range in which the modulation factor M does not exceed 1.0, the instantaneous powers of the two phases coincide with each other for approximately 1/6 period by the method of the minimax method.

  Next, the simulation result by the least square method is shown. FIGS. 10A and 10B are graphs showing temporal changes in the phase voltage and the neutral point potential when the modulation factor M by the least square method is 0.8. FIG. 11A and FIG. 11B are graphs showing temporal changes in the phase voltage and the neutral point potential when the modulation factor M according to the least square method is 1.0. As shown in the graphs of FIGS. 10 and 11, in the method using the square sum minimization, when the modulation rate is small, the neutral point potential is a sine wave having a frequency three times that of the fundamental wave. It moves in the same way as the third harmonic to improve the rate. However, as the modulation rate increases, the neutral point potential varies greatly within the limited range. Further, in the method using the square sum minimization, there is no coincidence section for each phase, and the peak power itself is slightly larger than the result by the minimax method.

  12 (a) and 12 (b) are graphs showing temporal changes in the average value of the square sum of the DC power when the modulation factor is 1.0 according to the minimax method and the least square method. As shown in FIG. 12, it can be seen that the average value of the square sum of the DC power is suppressed to the same extent by the fluctuation of the neutral point potential by the minimax method and the least square method.

  FIG. 13 summarizes the results of FIGS. FIG. 13 is a graph comparing the peak power with the sum of squares of power by the minimax method and the least squares method without correction. As shown in FIG. 13, when the modulation factor is 1.0, the maximum power is reduced by about 22% in the method using the square sum minimization as compared with the case of no correction in which the neutral point potential is not moved at all. . In addition, the method using the minimax method is improved by 25%.

  On the other hand, the average value of the sum of squares of the DC section current is negligibly small by the method of minimizing the sum of squares, but it can be considered that there is no significant difference between the two methods. The present inventors comprehensively evaluate that the minimax method is preferable to the least square method.

  FIGS. 14A and 14B are graphs showing the results of the phase voltage, neutral point potential, and phase power by the minimax method when the load power factor is 0.5 and the modulation factor is 0.8. . As shown in FIG. 14, the movement of the neutral point potential is greatly different from that in the case where the power factor is 1.0, and it can be seen that it works to minimize the instantaneous power according to the change in the power factor.

  From the simulation results of the above method of minimizing the sum of squares and the method using the minimax method, it was found that the method using the minimax method is slightly superior. When considering the computational complexity of both methods, the minimax solution requires at least six absolute value operations, three divisions, and a comparison of three variables. Therefore, the sum of squares minimization method is more advantageous in terms of calculation amount. However, considering the recent improvement in performance of control devices, the difference in time required for this calculation is considered to be slight. This method has the advantage of being able to handle unbalanced currents and harmonics associated with unbalanced loads and nonlinear loads.

  In addition, although the three-phase alternating current load 4 was connected to the output side of the power conversion unit 3 in each of the above embodiments via the output line 30, it is not limited to this and may be a two-phase load. . Even in such a case, the effects described in the above embodiments can be obtained.

  From the foregoing description, many modifications and other embodiments of the present invention are obvious to one skilled in the art. Accordingly, the foregoing description should be construed as illustrative only and is provided for the purpose of teaching those skilled in the art the best mode of carrying out the invention. The details of the structure and / or function may be substantially changed without departing from the spirit of the invention.

  The present invention can be used for a three-phase output power converter.

DESCRIPTION OF SYMBOLS 1 Three-phase alternating current power supply 2 Three-phase transformer 3 Power conversion part 3a-3c Single phase inverter (a phase-c phase)
4 Motor (load)
5 Current Sensor 7 Control Unit 11 Forward Converter 12 Smoothing Capacitor 13 Inverse Converter 30 Output Line 70 Voltage Command Generation Unit 71 PWM Signal Generation Unit 72 Neutral Point Potential Calculation Unit 73 Adder 74 Sign Inverter 75 Comparator 76 Logic Inversion 77 Delay circuit 78 Triangle wave generation circuit 100 Power converter

Claims (8)

Three-phase including three single-phase inverters, one output terminal of each single-phase inverter connected to a load, and the other output terminal of the three single-phase inverters connected to each other to form a neutral point The power converter of
A control unit for controlling the output voltage of each single-phase inverter so that the potential at the neutral point changes so that the output power of each single-phase inverter is reduced as compared with the case where the output power at the neutral point is not changed; With
The control unit is configured to reduce the potential of the neutral point so that the maximum value of the output power of the three single-phase inverters is minimized or the sum of squares of the DC unit currents of the three single-phase inverters is minimized. A multi-level power conversion device that controls the output voltage of each single-phase inverter so as to change . Wherein the control unit controls the output voltage of each of the single-phase inverters so that the maximum value of the output power of the three said single-phase inverter changes the potential of the neutral point so as to minimize, according to claim 1 Multi-level power converter. Further comprising three phase current detectors respectively detecting phase currents that are output currents of one output terminal of the three single-phase inverters;
The control unit is configured to minimize the maximum output power of the three single-phase inverters based on the phase voltage that the single-phase inverter wants to apply to the load and the phase currents detected by the three phase current detectors, respectively. The neutral point potential, which is the potential of the neutral point, is calculated, and the command value for controlling the output voltage of each single-phase inverter is generated by adding the calculated neutral point potential to the phase voltage. The multilevel power conversion device according to claim 2 . Wherein the control unit controls the output voltage of each of the single-phase inverters so that the square sum of the DC portion current three said single-phase inverter changes the potential of the neutral point so as to minimize, to claim 1 The multilevel power conversion device described. Further comprising three phase current detectors respectively detecting phase currents that are output currents of one output terminal of the three single-phase inverters;
The control unit calculates the sum of squares of the DC unit currents of the three single-phase inverters based on the phase voltage that the single-phase inverter wants to apply to the load and the phase currents detected by the three phase current detectors, respectively. A neutral point potential, which is the potential of the neutral point that is minimized, is calculated, and a command value for controlling the output voltage of each single-phase inverter by adding the calculated neutral point potential to the phase voltage. The multilevel power conversion device according to claim 4 , wherein the multilevel power conversion device is generated. The control unit, the neutral point further satisfying the constraint that the absolute value of the sum of the phase voltage that the single-phase inverter wants to apply to the load and the neutral point potential is less than or equal to the input voltage of each single-phase inverter The multilevel power conversion device according to claim 3 or 5 , which calculates a potential. When the neutral point potential does not satisfy the constraint condition, the control unit sets a close value between the maximum value and the minimum value satisfying the constraint condition, and the maximum value of the output power of the three single-phase inverters. The multilevel power conversion device according to claim 6 , wherein the multilevel power conversion device has a minimum value or a value that minimizes a sum of squares of input currents of the three single-phase inverters. Three-phase including three single-phase inverters, one output terminal of each single-phase inverter connected to a load, and the other output terminal of the three single-phase inverters connected to each other to form a neutral point A method for controlling a multi-level power conversion device including the power conversion unit of
It comprises controlling the output voltage of each of the single-phase inverters so that the output power of each of said single-phase inverter changes the potential of the neutral point so reduced as compared with the case that does not change the potential of the neutral point ,
Each of the neutral points changes so that the maximum value of the output power of the three single-phase inverters is minimized or the sum of squares of the DC currents of the three single-phase inverters is minimized. The control method of a multilevel power converter further comprising controlling the output voltage of the single phase inverter .

Images