Electrical transmission or interconnection systems – Switching systems – Condition responsive
Reexamination Certificate
2000-03-21
2002-04-23
Wong, Peter S. (Department: 2838)
Electrical transmission or interconnection systems
Switching systems
Condition responsive
Reexamination Certificate
active
06376935
ABSTRACT:
BACKGROUND OF THE INVENTION
The field of the invention is that of pulse modulation. More particularly, the invention relates to the control of the frequency and width of pulses for the control of switches, and particularly for power switches. Its field of use is in particular that of control systems with a variable structure (control by hysteresis, direct control of coupling (in the DTC angle: “Direct Torque Control”)) and modulations with width and/or frequency of pulses (MLI).
The modern electrical devices are comprised principally by four modules shown in FIG.
1
.
The source
11
can be continuous, monophase alternating or triphase. The power converter
12
ensures the function of adaptation of the source to the supply type necessary for the good operation of the load
13
:
continuous constant source converted into a monophase alternating source with variable frequency and amplitude;
triphase alternating source converted into a continuous source of variable amplitude;
etc.
The management of this conversion of type of source is ensured by the control module
14
. This module integrates in general two functions:
subjection (or regulation) of one or several sizes of load,
conversion of outputs of the regulators into control signals of the power switches.
The control systems with variable structure are more and more used, relative to conventional controls. This technique is also known as control by sliding mode or direct couple control DTC.
There can be distinguished two general types of control systems with variable structure, respectively in
FIGS. 2 and 3
.
FIG. 2
shows the case of a structural change by switching of a counter-reaction of condition. The electrical system
21
receives the voltage from an amplifier
22
controlled by a reference voltage delivered by one or another of the modules K
1
and K
2
(
23
1
and
23
2
), as a function of the switch
24
controlled by the commutation law S(x)
25
.
FIG. 3
shows the case of a change of structure by switching to the level of the amplifier (power converter). The commutation law S(x) then controls the switch
31
, which delivers the selected voltage E
max
or E
min
.
The commutation law S(x) is derived from a relationship between the reference and the variables of the condition of the system, so as to ensure the stability, the durability and the observance of the reference, namely:
u=E
max
siS
(
x
)>0
u=E
min
siS
(
x
)<0
This control has the advantage of a very rapid dynamic and a very good strength as to parametric variations. It is possible that the switching takes place at a very high frequency (theoretically infinitely high) such that the system works in the sliding mode.
However, this aspect turns out to be a major drawback in practice. Thus, there is no control over the frequency which requires this type of control for the switches of the power converter
12
which supplies the electrical supply. These frequencies can in certain cases be damaging to the static converter.
This problem is well known. Numerous methods seeking to solve it have already been proposed:
replacing the outlet relays of the regulator with hysteresis, with a band calculated as a function of the parameters of the system, associated with control systems of the width of the hysteresis band as a function of the operating point;
adding an auxiliary estimator and a regulator of the switching frequency. The parameters of this “frequency controller” are computed as a function of the parameters of the system to be controlled;
breaking down the regulator “with variable structure” into two sub-regulators: one is dedicated to the linearization of the model of the system to be controlled and the second is an image of the base regulator. It imposes the dynamic in closed loop and the durability for small variations of parameters.
Moreover, in the case of the regulation of the currents of the triphase motor with alternating current, there exist two typical solutions, control by MLI (modulation of the width of the pulses) and control by hysteresis.
The source is the continuous voltage type (which is to say a mean value which is not zero, in contrast to alternating voltage), the load is a synchronous or asynchronous triphase motor and the power converter is an inverter of triphase voltage (conversion from a constant continuous voltage into three alternate voltages of variable frequency and amplitude). The control means must be subject to the current from the motor with three alternative references, for example sinusoidal.
The principle of the control by MLI is given in FIG.
4
.
The error between the reference current
41
and the measured current
42
is processed by a corrector
43
. The control of the power switches
44
and
45
is obtained by the comparison
46
of the output of the regulator
43
and the triangular signal
47
of very high frequency relative to that of the reference currents (40 to 100 times, or even more).
This type of control ensures switching of the power switches at a constant frequency (frequency of the triangular signal of modulation), but the inversion of the current is not controlled, depending on the parameters of the load and of the point of operation. The synthesis of the regulator is in general based on the linear automatic control, which introduces an inherent dephasing into the linear transfer functions, at least of using a sophisticated corrector requiring a fairly powerful processor or a very complicated analog card.
In all cases, the quality of direct control is strictly connected to the fineness with which the parameters of the model of the system have been predetermined.
A second method utilized to render more sensitive, relative to the frequency of the current, the performances of the direct control, consists in carrying out a base change by means of a non-linear transformation matrix (so-called Park transformation), which transforms the alternative sizes into continuous components (constants) in the new base. The currents being continuous in this base, the corrections are determined to ensure good performance at zero frequency.
This method permits avoiding the problem of variable frequency of the references for the production of current regulators, but the sensitivity to the parameters of the model of the system remains unchanged relative to the previous method.
The control principle by hysteresis is shown in FIG.
5
. It consists in holding by means of three hysteresis comparators
51
(in the case of a triphase system) the real currents in the machine within a band of predetermined width, centered on the reference currents.
The inversion of the current is thus imposed but the frequency of switching is free and variable. It depends principally on the bandwidths imposed and the time constants of the system to be controlled (the motor in this case). This method has the following advantages:
a simple hysteresis comparator
51
permits directly controlling the currents and generating the controls for the switches of the power converter;
the direct control is not linear, which leaves the possibility of making almost zero the dephasing and minimizing the gain error between the reference currents and the real currents;
the performances of the direct control are less sensitive to the parameters of the model of the system.
Furthermore, they have several drawbacks, connected particularly to the poor control of the frequency of switching of the power switches piloted by the control signals
44
and
45
;
switching constraints are very important at the level of the power converter (heating, defects of switching, . . . );
the variation in frequency of switching can be a source of troublesome audible noises.
None of these known techniques is therefore satisfactory. Thus, they all depend on the parameters of the system to be controlled, which of course degrades the advantages reached by the control with variable structure, particularly in terms of universality.
Moreover, they are often costly. They require the use of complex electrical means, even of very rapid processors.
SUMMARY OF THE INVENTION
The inventio
Le Claire Jean-Claude
Le Doeuff René
Saillard Joseph
Siala Sami
Ecole Superieure Atlantique d'Ingenieurs en Genie Electriqu
Tibbits Pia
Wong Peter S.
Young & Thompson
LandOfFree
Method and device for controlling switches in a control... 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 and device for controlling switches in a control..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and device for controlling switches in a control... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2877081