Electric power conversion systems – Current conversion – Having plural converters for single conversion
Utility Patent
1999-10-07
2001-01-02
Wong, Peter S. (Department: 2838)
Electric power conversion systems
Current conversion
Having plural converters for single conversion
C363S098000, C363S132000
Utility Patent
active
06169677
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a power converting system multiplexed with voltage dividing transformers, the voltage transformers themselves, and a controller for controlling the power converting system so that the system may output sine waves having reduced harmonics.
2. Description of the Prior Art
An inverter for driving a motor at variable speeds generates AC voltages including direct currents. To smoothly drive the motor at variable speeds, the voltages generated by the inverter must be of sine waves involving reduced harmonics.
To meet the requirements for outputting sine waves with reduced harmonics, NPC inverters (3-level inverters) are frequently employed for power converting systems.
FIG. 1
shows a main circuit of an NPC inverter according to a prior art. The NPC inverter is capable of providing three phase-voltages and five line-voltages, to greatly reduce harmonics. The NPC inverter is characterized in that a voltage applied to each switching element is theoretically half that of a conventional inverter, and therefore, is easy to increase the capacity and voltage thereof.
The NPC inverter of
FIG. 1
, however, employs capacitors C
1
and C
2
to divide a DC voltage to increase the number of output levels, and therefore, has limits to the constant voltage characteristics thereof. Namely, a neutral point NT of a DC power source is connected to a load through diodes and switching elements for a certain period during which a current flows to the neutral point NT. Then, in spite of a constant DC voltage, the potential of the neutral point NT varies at a frequency that is three times as large as an output frequency. If a DC component in an output voltage involves a bias, divided voltage levels will greatly be biased to apply large voltages to the switching elements.
Variations at the neutral point NT may be suppressed by controlling an output voltage of the inverter. To achieve this, a DC link voltage must be set higher than an output line voltage required by the load connected to the inverter. This limits a modulation factor of the NPC inverter. A peak output line voltage of the inverter is “(3/2)×M×Vdc,” where M is a modulation factor and Vdc is a DC voltage. If the NPC inverter has a limit modulation factor of about 0.8, a limit peak output line voltage will be about 0.69 times of the DC voltage. In other words, a required peak output line voltage is obtainable only from a DC voltage that is about 1.45 times as large as the peak voltage. To supply such a high DC voltage, many switching elements must be connected in series, thereby increasing the number of the switching elements and the cost of the power converting system.
PWM control for the NPC inverter carries out comparison with the use of a triangle wave having a fixed carrier frequency. This PWM control is low in voltage use ratio and involves unnecessary switching operations to be repeated in synchronization with the carrier frequency, thereby increasing a switching loss, decreasing the efficiency of the power converting system, enlarging a stack and cooling system, and heightening the total cost of the system.
If there is a load that requires accurate sine-wave voltages, inverters must be connected and operated in parallel and PWM control signals of the inverters must be multiplexed to reduce harmonics.
FIG. 2
shows a power converting system multiplexed with N sets of 3-phase bridge inverters connected in parallel with one another. The N inverters form an inverter group
100
to convert a direct current into an alternating current. U-phases of the inverters are connected to an AC reactor group
200
U, V-phases thereof to an AC reactor group
200
V, and W-phases thereof to an AC reactor group
200
W. The output of the reactor group
200
U is connected to a U-phase of a load
300
, the output of the reactor group
200
V to a V-phase of the load
300
, and the output of the reactor group
200
W to a W-phase of the load
300
. The inverters are PWM-controlled, PWM control signals for the inverters are multiplexed, and AC voltages from the inverters are synthesized through the reactor groups
200
U,
200
V, and
200
W into sine waves, which are supplied to the load
300
.
FIG. 3
shows the structure of one inverter in the inverter group
100
. The inverter consists of switching elements Q
1
to Q
6
and diodes D
1
to D
6
that are connected to the switching elements Q
1
to Q
6
, respectively, in a reversed parallel configuration.
FIG. 4
shows a structure of the AC reactor group
200
U connected to the U-phases of the inverter group
100
. Each reactor consists of an iron core
204
and a winding
205
wound around the iron core
204
. On the load side, the reactors are joined together. The reactors are magnetically independent of one another.
A current IU flowing to the load
300
is divided by N, and a current of 1/N of IU passes through each reactor and each inverter. The inductance of each reactor is designed to be sufficiently small so that it does not act as impedance on the frequency of a fundamental wave supplied to the load
300
. Namely, the inductance of each reactor is sufficiently small with respect to the inductance of the load
300
, and a voltage drop at each reactor due to the current flowing to the load
300
is sufficiently small.
The operation of the AC reactors will be explained. In
FIG. 4
, the AC reactor group
200
U passes load current components I
1
U to INU and cross-current components I
1
UC to INUC.
The cross-current components are not directly related to the load
300
. However, they are synthesized with the load current components and flow through the reactors and inverters. If the cross-current components are large, the current capacity of each switching element such as a GTO of each inverter must be high. Namely, the inverters must have current capacity that is greater than that matching the load
300
. This makes the power converting system uneconomic.
To suppress the cross currents, each reactor may be designed to produce large inductance. This, however, increases the size, weight, and cost of each reactor. This also increases an installation space of the power converting system, and the strength of the floor on which the system is installed must be reinforced, thereby increasing the total cost of the system.
FIG. 5
is a power converting system according to a prior art having N current-type inverters that are connected and operated in parallel to supply sine-wave currents to a load. The N inverters form an inverter group
600
. On the AC side of the inverter group
600
, U-, V-, and W-phases of the inverters are connected to U-, V-, and W-phases of the load
300
, respectively, through a smoothing capacitor
700
. DC terminals of the inverter group
600
are connected to a DC power source
900
through cross-current suppressing reactor groups
800
P and
800
N.
The DC power source
900
supplies a constant direct current ID. Since an average of DC terminal voltages E
1
D-P to END-P is equal to an average of DC terminal voltages E
1
D-N to END-N, each reactor of the reactor groups
800
P and
800
N passes a direct current of 1/N of ID. The reactor groups
800
P and
800
N generate substantially no impedance with respect to the direct current ID.
The current-type inverters are PWM-controlled and multiplexed. The DC terminal voltages E
1
D-P to END-P and E
1
D-N to END-N have an identical DC component and different momentary voltages. Due to the difference among the momentary voltages, cross currents flow to the DC terminals of the inverters.
The reactor groups
800
P and
800
N serve to suppress the cross currents and have the same structures as those of FIG.
4
. Namely, these reactor groups are large and heavy and increase the cost of the power converting system. In addition, they increase the installation space of the system, and the strength of the floor on which the system is installed must be reinforced, thereby further increasing the cost of the system.
SUMMARY OF THE INVENTION
A first object of the pres
Fujita Takashi
Kageyama Takahisa
Kanai Takeo
Kawaguchi Akira
Kitahata Takeshi
Foley & Lardner
Kabushiki Kaisha Toshiba
Vu Bao Q.
Wong Peter S.
LandOfFree
Power converting system multiplexed with voltage dividing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Power converting system multiplexed with voltage dividing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Power converting system multiplexed with voltage dividing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2442298