Method for controlling neutral point potential of inverter...

Details Method for controlling neutral point potential of inverter... Method for controlling neutral point potential of inverter...

2002-02-27

2002-12-03

Riley, Shawn (Department: 2838)

Electric power conversion systems

Current conversion

With condition responsive means to control the output...

C363S132000

Reexamination Certificate

active

06490185

ABSTRACT:

TECHNICAL FIELD
The present invention relates to a neutral point potential control method of a power conversion device, such as a three-phase neutral point clamp type inverter, for performing variable speed driving of a motor or system interconnection, and a power conversion device, such as an inverter servo drive, for performing variable speed driving of a motor, or a power conversion device for performing system interconnection.
BACKGROUND OF THE INVENTION
Conventionally, as a neutral point potential control method of a three-phase neutral point clamp type inverter, there generally exists a method for carrying out control by applying a zero-phase voltage to an instruction voltage as disclosed in Japanese Patent No. 2821168 “Inverter device and AC motor driving system”, and a system in which a neutral point potential is controlled by adjusting a time of an output vector using a space voltage vector system.
FIG. 1
shows a basic structure of a three-phase neutral point clamp type inverter, and in the drawing, reference numeral
1
designates a three-phase AC power source;
2
, a rectifying element;
3
,
4
, smoothing capacitors;
6
to
23
, diodes;
24
to
35
, IGBTs; and
36
, a motor.
In
FIG. 1
, when a potential difference between a neutral point voltage (voltage of a connection point O of serial-connected smoothing capacitors of the inverter) and a negative bus voltage is Vcn, in the neutral point potential control, Vcn must be controlled to be a half voltage of a bus voltage Vpn of the inverter.
In a case where the three-phase neutral point clamp type inverter as shown in
FIG. 1
selects an output vector shown in FIG.
2
and controls Vcn, as voltage vectors which can be used for control of Vcn, there are only twelve vector of xp(
1
), xn(
1
), xp(
2
), xn(
2
), xp(
3
), xn(
3
), yp(
1
), yn(
1
), yp(
2
), yn(
2
), yp(
3
), and yn(
3
).
FIG. 3
shows connection states of a load and the smoothing capacitors of the inverter in the case where the twelve vectors are outputted.
When a load current flows in a direction of an arrow shown in
FIG. 3
, for example, in the connection state of xp(
1
) and xn(
1
) of a region
1
, since directions of currents flowing to the neutral point become opposite to each other, if xp(
1
) and xn(
1
) are generated in such a minute time that a U-phase current does not change as shown in
FIG. 4
, Vcn rises at the time of the generation of xp(
1
), and drops at the time of the generation of xn(
1
), and if a generation time Tp of xp(
1
) is equal to a generation time Tn of xn(
1
), an average Vcn in a total generation time Tout (=Tp+Tn) of xp(
1
) and xn(
1
) becomes a constant voltage, and if Tp>Tn, then the average Vcn rises. If Tp<Tn, the average Vcn drops.
It is understood that in this way, by adjusting the generation time ratio of the vectors arranged side by side in
FIG. 3
(xp(
1
) and xn(
1
), yp(
1
) and yn(
1
), xp(
2
) and xn(
2
), yp(
2
) and yn(
2
), xp(
3
) and xn(
3
), yp(
3
) and yn(
3
)), Vcn can be controlled.
In the neutral point potential control in which the zero-phase voltage is applied to the instruction voltage, application of a positive zero-phase voltage is almost equivalent to lengthening of a generation time ratio of xp(
1
), xp(
2
), xp(
3
), yp(
1
), yp(
2
), or yp(
3
), and application of a negative zero-phase voltage is almost equivalent to lengthening of a generation time of xn(
1
), xn(
2
), xn(
3
), yn(
1
), yn(
2
), or yn(
3
).
In the system in which the space voltage vector is used, for example, when a certain voltage vector in a region i (i=1, 2, . . . , 6) is outputted, a total output time of xp(j) and xn(j) vectors is made Tx(i), a total output time of yp(k) and yn(k) vectors is made Ty(i), an output time of xp(j) is made Txp(j), an output time of xn(j) is made Txn(j), an output time of yp(k) is made Typ(k), and an output time of yn(k) is made Tyn(k), and when &agr; is defined to set such a relation as
Txp
(
j
)=&agr;
Tx
(
i
)
Txn
(
j
)=(1−&agr;)
Tx
(
i
)
Typ
(
k
)=&agr;
Ty
(
i
)
Tyn
(
k
)=(1−&agr;)
Ty
(
i
)
(j=1 when i=1,j=2 when k=1 and i=2,j=2 when k=1 and i=3, j=3 when k=2 and i=4,j=3 when k=2 and i=5,j=1 when k=3 and i=6,k=3),
in an electrical driving state,
when &agr; is made large, Vcn rises,
when &agr; is made small, Vcn drops,
besides, in a regenerative state,
when &agr; is made large, Vcn drops,
when &agr; is made small, Vcn rises,
whereby, the neutral point potential can be controlled by adjusting &agr;.
However, in the conventional neutral point potential control method in which the zero-phase voltage is added to the instruction, since the neutral point potential control becomes impossible at a load power factor of approximately zero, as a method of performing the neutral point potential control without receiving the influence of the load power factor to solve this, there is a method as disclosed in Japanese Patent Unexamined Publication No. Hei. 9-182455 in which an even component of modulation instructions is superimposed as the zero-phase voltage, however, there has been a problem that an effect is not remarkable although the control is complicated.
Besides, as disclosed in Japanese Patent No. 2888104, although there is a method in which a corresponding neutral point potential period is adjusted according to a direction of a current of a predetermined phase, there has been a problem that in a multi-phase inverter, the control of inter-phase output voltage can not be carried out well.
Then, a first problem to be solved by the present invention is to provide a neutral point potential control method of a three-phase neutral point clamp type inverter in which without degrading the quality of an inter-phase output voltage, neutral point potential control can be carried out irrespective of a power factor by simple measurement or prediction of a phase current, and neutral point potential fluctuation due to load current unbalance at the time of load ground fault can also be suppressed, whereby the quality, stability and safety of the inverter can be improved.
Further, as a conventional PWM pulse generating method of a three-phase neutral point clamp type PWM inverter, as disclosed in Japanese Patent Unexamined Publication No. Hei. 5-146160, there is a unipolar modulation/dipolar modulation for outputting a pulse by comparing an amplitude instruction with a carrier wave, or as disclosed in Japanese Patent Unexamined Publication No. Hei. 5-292754, there is a system in which a generation time of each vector is calculated by using an idea of a space vector and a PWM pulse is generated.
FIG. 5
is a vector diagram in which output voltage vectors of a three-phase neutral point clamp type inverter are shown on a plane. When a switch state in which a phase output terminal of the three-phase neutral point clamp type PWM inverter is connected to a positive bus is P, a switch state in which it is connected to a negative bus is N, and a switch state in which it is connected to a neutral line is O, and when they are arranged in the order of UVW of output phases, output voltage vectors which the three-phase neutral point clamp type inverter can take have 27 kinds of switch states as shown in FIG.
5
.
Here, for convenience of explanation, the 27 kinds of switch states shown in
FIG. 5
, which the three-phase neutral point clamp type PWM inverter can take, are classified into groups of
zero vector
PPP: Op
OOO: Oo
NNN: On
x vector
POO, OPO, OOP: xp
ONN, NON, NNO: xn
y vector
PPO, OPP, POP: yp
OON, NOO, ONO: yn
z vector
PON, OPN, NPO, NOP, ONP: z
a vector
PNN, NPN, NNP: a
b vector
PPN, NPP, PNP: b
and division is made such that
regions surrounded by the zero vector, the x vector and the y vector are
1
-
1
to
6
-
1
,
regions surrounded by the x vector, the a vector and the z vector are
1
-
2
to
6
-
2
,
regions surrounded by the x vector, the y vector and the z vector are
1
-
3
to
6
-
3
, and
regions surrounded b

Affiliated with

Also associated with

Rating

Method for controlling neutral point potential of inverter... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method for controlling neutral point potential of inverter..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for controlling neutral point potential of inverter... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2972204