[Technical Field] [0001]
The present invention relates to a power converter and a power conversion method, and more specifically, it relates to a power converter and a power conversion method suitable for a main circuit configuration and for protection of a main circuit.
[Background Art] [0002]
Power converters having semiconductor switching elements have been widely used in the fields of industry, household appliances, transportation, automobiles, electric power/social infrastructure systems, and so on. Power converters for industrial applications requiring several hundred kilowatts or greater are often formed of assemblies containing a plurality of semiconductor switching elements. For such a power converter, even if a fault such as a short-circuit of a semiconductor switching element occurs, it is desirable that the fault phase be cut off quickly by using a protection means in the power converter to suppress the influence on the sound phases, thereby realizing a reduction in downtime until restoration. To do so, a protection

method in which a fuse is inserted inside the power converter at a location where a direct current (DC) circuit section is connected is often used.
[0003]
A configuration in which a fuse is inserted into a DC circuit section for protecting the power converter and to which a circuit having a diode and a capacitor connected in series is added so that the fuse can be melt-cut quickly if a short-circuit fault of a semiconductor switching element occurs is described in, for example, Japanese Unexamined Patent Application Publication No. 2014-96892. [Citation List]
[Patent Literature]
[0004]
[PTL 1] Japanese Unexamined Patent Application Publication No. 2014-96892 [Summary of Invention] [Technical Problem]
[0005]
The abovementioned conventional technology is a protection circuit for a configuration in which three phases collectively enclose a smoothing capacitor. Meanwhile, the voltage of power converters for industry or electric power/social infrastructure systems is increasing in order to reduce power loss, and in such a high-voltage power

converter, high-voltage/high-current components including a semiconductor switching element are used. When a three-phase collective type circuit is formed, the distance between a smoothing capacitor and a semiconductor switching element increases in order to secure the insulation distance and the creeping distance. As the distance increases, the parasitic inductance of the main circuit increases. If the parasitic inductance increases, the semiconductor switching element may be damaged by a spike voltage generated between the terminals of the semiconductor switching element when switching of the semiconductor switching element is performed, and therefore, an additional circuit, such as a snubber circuit, is required for controlling a spike voltage and thereby the size of the power converter is increased.
[0006]
As described above, in a power converter, especially in a high-voltage power converter, it is difficult to downsize the power converter while suppressing the influence of a fault phase on sound phases in a short-circuit fault.
[0007]
The object of the present invention is to provide a power converter and a power conversion method capable of downsizing the power converter and capable of being easily restored when a fault occurs. [Solution to Problem]

[0008]
To achieve the abovementioned object, the present invention is configured by having a plurality of phase circuits that form phases and that partially contain a common direct current voltage section in such a manner that the plurality of phase circuits each have a switching circuit for performing a power conversion operation composed of semiconductor switching elements, an alternating current (AC) terminal connected to the switching circuit, a plurality of capacitors connected in parallel between the switching circuit and the direct current voltage section, and a fuse connected between the switching circuit and the direct current voltage section.
[0009]
Alternatively, the present invention is configured by having an assembly, for each phase, composed of semiconductor switching elements, a first capacitor, a second capacitor, and wires, and common DC circuits in which DC circuits of the assembly are connected in parallel, in which the first capacitor and the second capacitor are connected in parallel between the DC circuits of the assembly, the first capacitor terminals are connected to points on the assembly DC circuit, the points being adjacent to the semiconductor switching element, the second capacitor terminals are connected to points on the assembly DC circuit,

the points being close to the common DC circuits compared with the first capacitor, and a fuse is inserted into a part of a wire connecting a first capacitor terminal connection point and a second capacitor terminal connection point. [Advantageous Effects of Invention] [0010]
According to the present invention, lower inductance of a DC circuit expected from a semiconductor switching element can be realized, and as a result, the downsizing of the device can be attained. In addition, since a fuse can be melt-cut by a discharge current of a capacitor inside an assembly, the influence of a fault phase on a sound phase in a short-circuit fault can be suppressed and restoration can be made easily by replacing a component of only the fault phase.
[Brief Description of Drawings] [0011]
[Fig. 1] Fig. 1 shows a configuration of a power converter of the first example of the present invention.
[Fig. 2] Fig. 2 is a diagram explaining a flow of current in the first example of the present invention.
[Fig. 3] Fig. 3 is a diagram explaining a circuit configuration and a current path in a reference example case.
[Fig. 4] Fig. 4 is a diagram explaining a circuit configuration and a current path in a reference example.

[Fig. 5] Fig. 5 shows a configuration of a power converter of the third example of the present invention.
[Fig. 6] Fig. 6 shows a configuration of a power converter of the fourth example of the present invention.
[Fig. 7] Fig. 7 is a diagram explaining a circuit using semiconductor switching elements of the present invention.
[Fig. 8] Fig. 8 is a diagram explaining a circuit using semiconductor switching elements of the present invention.
[Fig. 9] Fig. 9 is an explanatory diagram of current flowing from a capacitor of the present invention. [Description of Embodiments] [0012]
The embodiments for carrying out the invention will be explained below with reference to the drawings. [Example 1] [0013]
Example 1 will be explained. Fig. 1 shows a circuit configuration of a power converter. In the present example, a three-phase two-level converter is explained as an example. A main circuit of the present power converter is constituted by circuits 102a, 102b, and 102c each forming one phase, and the DC terminals of the circuits 102a, 102b, and 102c are connected in parallel between a common DC circuit 100a (also referred to as DC voltage section (positive electrode)) and a common DC circuit 100b (also referred to as DC voltage

section (negative electrode)). The circuits 102a, 102b, and 102c are provided with disconnection configurations, which are not shown, between the circuits 102a, 102b, and 102c and the common DC circuits, thereby allowing each phase to be detached or replaced.
[0014]
The circuits (102a, 102b, 102c) each forming one phase form a U-phase, a V-phase, and a W-phase, for example, and a terminal (104a), a terminal (104b), and a terminal (104c) of the circuits (102a, 102b, 102c) each forming one phase are respectively connected to a U-phase terminal, a V-phase terminal, and a W-phase terminal of a three-phase AC power device. With this configuration, an AC power generator, which is not shown, is driven or regenerative power is converted into direct current.
[0015]
For example, when the circuit (102a) forming one phase is taken as an example, in a circuit (101) of the circuit (102a) forming one phase, as illustrated in Fig. 7, a pair of a positive switching element (401) and a positive backward diode (402) and a pair of a negative switching element (401) and a negative backward diode (402) are connected in series. A connection to the DC circuit (100a) is made via a wire (positive) (108a) connected to a circuit composed of semiconductor switching elements and a

connection to the DC circuit (100b) is made via a wire (negative) (109a) connected to the circuit composed of semiconductor switching elements. In addition, a connection point between the pair of the positive switching element (401) and the positive backward diode (402) and the pair of the negative switching element (401) and the negative backward diode (402) is an input/output terminal (104a). [0016]
By switching such as with PWM, the switching element (401) converts the DC power between the DC circuit (100a) and the DC circuit (100b) into AC power and outputs the AC power from the terminal (104a), or the switching element (401) converts the AC power from the terminal (104a) into DC power and outputs the DC power to the DC circuit (100a) and the DC circuit (100b). [0017]
A power conversion is performed by alternately switching between the upper and lower switching elements (401) by using a gate drive circuit and a gate control circuit, which are not shown. The switching elements (401) are controlled by a PWM modulation method in which a modulation wave and a carrier wave are compared using a PWM control circuit, which is not shown. [0018] The circuits (102b and 102c) each forming another phase

have the same configuration as the circuit (102a) forming one phase.
[0019]
In the circuit (102a) forming one phase, a first capacitor (103a) is installed adjacent to the circuit (101a) composed of semiconductor switching elements and so as to be parallel to the DC circuit (100a) and the DC circuit (100b). This circuit has the configuration illustrated in Fig. 7. In Fig. 7, an IGBT (401) is used as a semiconductor switching element, however, the IGBT (401) may be replaced with another semiconductor switching element (transistor, GTO, thyristor, etc.). As described above, since the length of a current path (C100) from the capacitor (103a) to the circuit (101a) composed of semiconductor switching elements decreases, the parasitic inductance of the current path can be reduced. In this case, the spike voltage applied between the terminals when the IGBT (401) performs switching is determined by the parasitic inductance and a current change (di/dt), and thus, by using the above configuration, the spike voltage can be controlled. Consequently, a snubber circuit for protecting a semiconductor switching element from a spike voltage is not required, and thereby the size of the power converter can be reduced.
[0020]
The configuration for achieving suppression of a fault

effect on a sound phase by using a second capacitor and the mechanism thereof, which are features of the present invention, will be explained below.
[0021]
The power converter indicated in the present example is characterized in that an assembly provided in each phase has a first capacitor (103a) provided close to the semiconductor switching element (101a), fuses (105a, 106a) provided on the wires connected to the terminals of the capacitor (103a), and, in addition to the first capacitor (103a), a second capacitor (107a) installed closer to the common DC circuits (100a, 100b), compared with the fuses, and between the DC circuit (100a) and the DC circuit (100b).
[0022]
With this configuration, when a short-circuit fault of a semiconductor switching element occurs, only a fuse of the fault phase can be melt-cut, and thereby the influence on the assemblies (102b, 102c) of the sound phases can be reduced.
[0023]
The flows of current in melt-cutting a fuse will be explained with reference to Fig. 2.
[0024]
The fuse (105a) is inserted in such a manner that one terminal of the semiconductor switching element (101a) is

separated from the DC circuit 100a by melt-cut. The fuse (106a) is inserted in such a manner that the other terminal of the semiconductor switching element (101a) is separated from the DC circuit (100b) by melt-cut. [0025]
First, in a short-circuit of a semiconductor switching element, electric charge is discharged from the capacitor (103a) adjacent to the circuit (101a) composed of the semiconductor switching elements to the circuit (101a) composed of the semiconductor switching elements (current path (200)). The voltage between the terminals of the capacitor (103a) is reduced by the discharge of electric charge and thus current flows from the capacitor (107a) via the fuse (105a), and thereby the fuse is melt-cut by the current (current path (201))to separate the fault phase from the DC voltage sections (100a, 100b). Since the capacitor (107a) is provided in the same assembly as the circuit (101a) composed of the semiconductor switching elements, the parasitic inductance from the capacitor (107a) to the circuit (101a) composed of the semiconductor switching elements is small compared with the parasitic inductances from capacitors in the assemblies (102b, 102c) of the sound phases to the circuit (101a) composed of the semiconductor switching elements of the fault phase (102a). Consequently, the current necessary for melt-cutting the fuse of the fault

phase is supplied from the capacitor (107a). When the current for melt-cutting the fuse is supplied from the capacitor (107a), the terminal voltage of the capacitor (107a) is reduced, and thus current also enters from the sound phases. In order to reduce the influence on the sound phases, it is preferable that parasitic inductance from the sound phases to the capacitor (107a), the parasitic inductance being generated due to wires (L10, L20) for connecting phases, be sufficiently large compared with the parasitic inductance from the capacitor (107a) to the circuit (101a) composed of the semiconductor switching elements. Note that it is preferable that the parasitic inductance of the wire (L10) for connecting phases and the capacitance of each capacitor be selected such that the resonance frequency that is determined by the capacitance of the capacitor (107a) and the parasitic inductance of a current path (201) becomes four times or more the resonance frequency that is determined by the capacitor (103b) and the parasitic inductance of a current path (203). Consequently, as shown in Fig. 9 for example, before a current (801) entering from a sound phase reaches a peak value, the fuse can be melt-cut by a current (800) flowing from the capacitor (107a) of the fault phase. This is the mechanism for reducing the influence on the sound phases in the case where a short-circuit fault of a semiconductor switching

element occurs.
[0026]
The procedure of melt-cutting a fuse in a case where the capacitor (107a) is not located close to a fuse will be explained as a comparative example with reference to Fig. 3. Suppose that the leftmost phase (102a) is a fault phase and the other phases (102b, 102c) are sound phases. If a short-circuit of a semiconductor switching element occurs in this configuration, first, electric charge is discharged from a capacitor (103a) adjacent to a circuit (101a) composed of semiconductor switching elements (current path (200)) and then electric charge is discharged from capacitor (103b) of the sound phases (102b, 102c) (current path (202)), thereby melt-cutting a fuse, and therefore, fatigue develops not only in the fault phase (102a) but also in the fuses of the sound phases (102b, 102c). Consequently, when restoring from the fault, not only the fault phase but also the other phases need to be restored together.
[0027]
Next, a case where the capacitor (103a) is not located close to a circuit (101a) composed of semiconductor switching elements will be explained as a comparative example with reference to Fig. 4. In this configuration, when a short-circuit fault of a semiconductor switching element occurs, a fuse is melt-cut by discharge of electric

charge from a capacitor (107a) installed close to the fuse and, thus, as with the configuration shown in Fig. 1, current from another phase is prevented from entering and only the fuse of the fault phase is melt-cut. However, since the current path (C200) from the capacitor (107a) to the circuit (101a) composed of the semiconductor switching elements increases, the parasitic inductance of the current path increases. Therefore, there is a concern of a spike voltage in switching the semiconductor switching elements, and thus a protection circuit, such as a snubber circuit, is required, thereby increasing the size of the power converter.
[0028]
As described above, according to the present example, since lower inductance of a DC circuit expected from a semiconductor switching element can be realized, and consequently, addition of a snubber circuit is not required, downsizing of the power converter can be attained. In addition, since a fuse can be melt-cut by discharge current of the second capacitor inside the assembly, the influence of the fault phase on a sound phase in a short-circuit fault can be suppressed, and restoration can be made by replacing a component of only the fault phase. [Example 2]
[0029]
A method for selecting a capacitance of a capacitor

(107a) installed close to a fuse will be described with reference to Fig. 2. In Example 2 (and also in other examples) which is a modified example of Example 1, only different parts will be explained. Therefore, parts for which explanations are omitted are the same as those of Example 1. When a short-circuit fault of a semiconductor switching element (402) inside a circuit (101a) composed of semiconductor switching elements occurs, a fuse (105a) is melt-cut by discharge current of the capacitor (107a), as described in Example 1. In this case, the fuse is melt-cut by means of Joule integral value Is2t(A2s) that is determined by a discharge current Is(A) flowing for melt-cutting the fuse and a discharge time period ts(s). [0030]
Now, when the resistance of a current path (201) in a short-circuit fault is represented as Rs(fi) , the voltage of the capacitor (107a) before discharge is represented as V0(V), the voltage thereof after discharge is represented as VI(V), and the capacitance of the capacitor (107a) is represented as Cf(F), the time t(s) necessary for discharging the voltage of the capacitor from V0(V) to VI(V) is determined by Formula (1). That is, the time for melt-cutting a fuse in a short-circuit fault is determined by the characteristic (Joule integral value Is2t(A2s)) of the fuse, the resistance in the short-circuit fault, and the

capacitance of the capacitor (107a) for melt-cutting the fuse. Therefore, the capacitance of the capacitor (107a) for melt-cutting a fuse is selected according to the time at which the fuse is desired to be melt-cut in a case where a short-circuit fault of a semiconductor switching element (402) occurs. The selected component part is adopted. [0031]
t(s)=-CfxRsxln(V0/Vl) Formula 1 [Example 3] [0032]
In Fig. 5, an example of three-phase three-level converter is shown. In addition to a DC circuit (300a) (also referred to as DC voltage section (positive electrode)) and a DC circuit (300b) (also referred to as DC voltage section (negative electrode)), a DC circuit (300c) (also referred to as DC voltage section (common)) is employed. A capacitor (308a) is provided between the DC circuit (300a) and the DC circuit (300c), and a capacitor (308b) is provided between the DC circuit (300c) and the DC circuit (300b). [0033]
A capacitor (303a) is provided between a wire (positive) (309) connected to a circuit composed of semiconductor switching elements of a circuit (301) forming one phase and a wire (common) (311) connected to the circuit

composed of the semiconductor switching elements, and a capacitor (303b) is provided between the wire (311) and a wire (negative) (310) connected to the circuit composed of the semiconductor switching elements.
[0034]
A fuse (305a) is provided between the wire (309) of the circuit (301) forming one phase and the DC circuit (300a), a fuse (307a) is provided between the wire (311) and the DC circuit (300c), and a fuse (306a) is provided between the wire (310) and the DC circuit (300b).
As shown in Fig. 8, in the circuit (301) forming one phase, four pairs of a switching element (401) and a backward diode (402) are connected in series between the wire (309) and the wire (310), and a connection point between the upper two pairs of the switching element (401) and the backward diode (402) and a connection point between the lower two pairs of the switching element (401) and the backward diode (402) are connected with two diodes (501) connected in series. One terminal of the circuit formed of the four pairs of switching elements (401) and the backward diode (402) connected in series is connected to the DC circuit (300a) via the wire (309) and the other terminal thereof is connected to the DC circuit (300b) via the wire (310) . A connection point between the two diodes (501) is connected to the DC circuit (300c) via the wire (311) . A

connection point between the upper two pairs of the switching elements (401) and the backward diode (402) and a connection point between the lower two pairs of the switching elements (401) and the backward diode (402) are input/output terminals (304). Other circuits each forming one phase have the same configuration.
[0035]
The circuit shown in Fig. 8 operates at so-called three levels by means of a PWM modulation, and a positive output, a zero output, or a negative output is output to the input/output terminal 304, or AC power of the input/output terminal 304 is converted into DC power.
[0036]
As with the case of Example 1, the circuit is composed of capacitors (308a, 308b), fuses (305a, 306a, 307a), capacitors (303a, 303b), and a circuit (301) composed of semiconductor switching elements in this order. The effects and operations of this configuration is the same as those of Example 1. [Example 4]
[0037]
Fig. 6 shows another example of three-phase two-level converter. As with the case of Example 1, a first capacitor (103a) is installed adjacent to a circuit (101a) composed of semiconductor switching elements, and, with the circuit

(101a) composed of semiconductor switching elements and the first capacitor (103a) as one circuit group (700), circuit groups are connected in parallel with respect to fuses (105a, 106a). Note that, in the present example, a circuit configuration having two parallel circuit groups is described; however, the circuit configuration may have a different number of parallel circuit groups. In addition, the present configuration is also applied to the configuration described in Example 3. [Reference Signs List] [0038]
100a, 300a: DC voltage section (positive) 100b, 300b: DC voltage section (negative) 101, 101a, 101b, 301: circuit composed of semiconductor switching elements
104a, 104b, 304: terminal (AC output terminal) 105a, 106a, 105b, 106b, 305a, 306a, 307a: fuse 102a, 102b, 102c, 302a, 302b, 302c: circuit forming a phase
103a, 103b, 107a, 303a, 303b, 308a, 308b: capacitor
108, 309: wire (positive) connected to a circuit composed of semiconductor switching elements
109, 310: wire (negative) connected to a circuit composed of semiconductor switching elements
200, 201, 202, 203: current path

300c: DC voltage section (common)
311: wire (common) connected to a circuit composed semiconductor switching elements
401: IGBT
402: diode
700: circuit group composed of a circuit composed o semiconductor switching elements and a capacitor
800, 801: current for melt-cutting a fuse
C100, C200: current path
L10, L20: wire for connecting phases

CLAIMS
[Claim 1]
A power converter having a plurality of phase circuits that form phases and that partially contain a common direct current voltage section, the plurality of phase circuits each having: a switching circuit for performing a power conversion operation composed of semiconductor switching elements; an alternating current terminal connected to the switching circuit; a plurality of capacitors connected in parallel between the switching circuit and the direct current voltage section; and a fuse connected between the switching circuit and the direct current voltage section.
[Claim 2]
The power converter according to Claim 1, wherein the plurality of capacitors connected in parallel are constituted by a first capacitor connected adjacent to the switching circuit and a second capacitor connected parallel to the first capacitor, wherein the fuse is inserted into a wire that connects the first capacitor and the second capacitor, and wherein the second capacitor has a capacitance necessary for melt-cutting the fuse in a desired time period.
[Claim 3]
The power converter according to Claim 1 or 2, wherein a parasitic inductance of a wire connecting a capacitor in

each phase circuit and a semiconductor switching element of the phase is smaller than a parasitic inductance of a wire connecting phases.
[Claim 4]
The power converter according to Claim 2, wherein a resonance frequency that is determined by an inductance between the second capacitor and the circuit composed of semiconductor switching elements in one phase and a capacitance of the second capacitor becomes four times or more a resonance frequency that is determined by an inductance between the switching circuit in the one phase and the first capacitor of another phase different from the one phase and a capacitance of the first capacitor in the other phase.
[Claim 5]
The power converter according to any one of Claims 1 to 4, wherein one phase composed of the plurality of semiconductor switching elements, the capacitors, wires, and the fuse is configured as one assembly so that the assembly can be integrally detached when a component is replaced.
[Claim 6]
The power converter according to Claim 1, wherein a capacitance of the capacitor for melt-cutting the fuse is selected according to a fuse melt-cutting time period in a short-circuit fault of the semiconductor switching element.

[Claim 7]
The power converter according to Claim 1, wherein the direct current voltage section is a positive direct current voltage section, a negative direct current voltage section, and a common direct current voltage section, wherein a capacitor is arranged between the direct current voltage section and the common direct current voltage section, a capacitor is arranged between the common direct current voltage section and the negative direct current voltage section, and wherein fuses are arranged so as to correspond to respective ones of the positive direct current voltage section, the negative direct current voltage section, and the common direct current voltage section. [Claim 8]
A power conversion method for controlling a power converter having a plurality of phase circuits that form phases and that partially contain a common direct current voltage section, wherein each of the plurality of phase circuits performs a power conversion by a switching operation of a semiconductor switching element, a spike voltage of the semiconductor switching element is absorbed by using a plurality of capacitors connected in parallel between the switching circuit and the direct current voltage section, and, when a fault occurs, a cutoff operation is performed by a fuse connected between the switching circuit

and the direct current voltage section.