15-11-2012, 02:13 PM
Pulsewidth Modulation-A Survey
[attachment=39603]
Abstract-
This paper evaluates the state of the art in
pulsewidth modulation for ac drives fed from three-phase voltage
source inverters. Feedforward and feedback pulsewidth
modulation schemes having relevance for industrial application
are described and their respective merits and shortcomings are
explained. Secondary effects such as the influence of load current
dependent switching time delay and transients in synchronized
pulsewidth modulation schemes are discussed, and adequate
compensation methods are presented. Recorded oscillograms
illustrate the performance of the respective pulsewidth
modulation principles. The paper provides a guideline and quick
reference for the practicing engineer to decide which methods
should be considered for an application of a given power level,
switching frequency, and dynamic response.
INTRODUCTION
PU LSEWIDTH modulation (PWM) techniques have been the subject of intensive research during the last
few decades. A large variety of methods, different in
concept and performance, have been newly developed and
described. Their implementation in the design of ac drive
systems depends on the machine type, the power level,
and the semiconductor devices used in the power converter.
It is lastly performance and cost criteria which
determine the choice of a PWM method in a specific
application.
PERFORMANCCER ITERIA[/b]
Operation in the switched mode is a very efficient
means of controlling the power flow in the machine. The
switching harmonics are suppressed to a large extent by
the low-pass characteristic of the machine inductances
and by the inertia of the mechanical system. The remaining
distortions of the current waveforms and the electromagnetic
torque can be valued by performance criteria
[ 1-41.
Carrier-Based PWM
These are the classical and most widely used methods
of pulsewidth modulation. They have as common characteristic
subcycles of constant time duration, a subcycle
being defined as the total duration T, during which an
active inverter leg assumes two consecutive switching
states of opposite voltage polarity. Operation at subcycles
of constant duration is reflected in the harmonic spectrum
by two salient sidebands, centered around the carrier
frequency, and additional frequency bands around integral
multiples of the carrier.
1) Suboscillation Method: This method employs individual
modulators in each of the three phases (see Fig. l(a)).
Exemplified waveforms for phase a are shown in Fig. 1
(b), consisting of the sinusoidal reference voltage U: and
the triangular carrier signal U, of frequency f,. The
switched output waveform is U',. The maximum value of
the modulation index, mmaxl = 7i-/4 = 0.785, is reached
at a point where the amplitudes of the reference and the
carrier signal become equal (Fig. 1 (b)).
Polarity Consistency Rule
While the design of a single-phase PWM modulator is a
straightforward task, the voltage of one phase in a threephase
system depends on the switching states of all three
inverter legs. The result of such interaction can be seen
when looking at the individual phase voltages, or at the
a,P components of the switching state vector. These
waveforms are composed of up to five different voltage
levels, with the phase potentials having only the two
levels. The phase voltage at a given time instant should
not differ much from the value of the sinusoidal reference
wave for a well-designed modulation scheme. Generally
speaking, the switched voltage should assume at least the
same polarity as the reference voltage. The polarity consistency
rule is a means to identify ill-designed PWM
schemes.
Optimized Feedforward PWM
PWM inverters of higher power rating are operated at
low switching frequency to reduce the switching losses.
Here, only synchronized pulse schemes can be employed
in conjunction with feedforward schemes in order to avoid
intolerably high subharmonic components. The same applies
for drive systems operating at high fundamental
frequency while the switching frequency is in the lower
kilohertz range. As there are only a few switching instants
t, per fundamental period, small variations of the respective
switching angles uk = w1 . t, have considerable influence
on the harmonic distortion of the machine currents.
Feedback PWM with Optimization in Real-Time
1) Predictive Current Control: Pulsewidth modulation by
predictive current control has common elements with
look-up table methods discussed above. The time of
switching is determined by suitable error boundaries. As
an example, Fig. 19 shows a circular boundary, the location
of which is determined by the current reference
vector i*. When the current vector i touches the boundary
line, the next switching state vector is determined by
prediction and optimization.
[attachment=39603]
Abstract-
This paper evaluates the state of the art in
pulsewidth modulation for ac drives fed from three-phase voltage
source inverters. Feedforward and feedback pulsewidth
modulation schemes having relevance for industrial application
are described and their respective merits and shortcomings are
explained. Secondary effects such as the influence of load current
dependent switching time delay and transients in synchronized
pulsewidth modulation schemes are discussed, and adequate
compensation methods are presented. Recorded oscillograms
illustrate the performance of the respective pulsewidth
modulation principles. The paper provides a guideline and quick
reference for the practicing engineer to decide which methods
should be considered for an application of a given power level,
switching frequency, and dynamic response.
INTRODUCTION
PU LSEWIDTH modulation (PWM) techniques have been the subject of intensive research during the last
few decades. A large variety of methods, different in
concept and performance, have been newly developed and
described. Their implementation in the design of ac drive
systems depends on the machine type, the power level,
and the semiconductor devices used in the power converter.
It is lastly performance and cost criteria which
determine the choice of a PWM method in a specific
application.
PERFORMANCCER ITERIA[/b]
Operation in the switched mode is a very efficient
means of controlling the power flow in the machine. The
switching harmonics are suppressed to a large extent by
the low-pass characteristic of the machine inductances
and by the inertia of the mechanical system. The remaining
distortions of the current waveforms and the electromagnetic
torque can be valued by performance criteria
[ 1-41.
Carrier-Based PWM
These are the classical and most widely used methods
of pulsewidth modulation. They have as common characteristic
subcycles of constant time duration, a subcycle
being defined as the total duration T, during which an
active inverter leg assumes two consecutive switching
states of opposite voltage polarity. Operation at subcycles
of constant duration is reflected in the harmonic spectrum
by two salient sidebands, centered around the carrier
frequency, and additional frequency bands around integral
multiples of the carrier.
1) Suboscillation Method: This method employs individual
modulators in each of the three phases (see Fig. l(a)).
Exemplified waveforms for phase a are shown in Fig. 1
(b), consisting of the sinusoidal reference voltage U: and
the triangular carrier signal U, of frequency f,. The
switched output waveform is U',. The maximum value of
the modulation index, mmaxl = 7i-/4 = 0.785, is reached
at a point where the amplitudes of the reference and the
carrier signal become equal (Fig. 1 (b)).
Polarity Consistency Rule
While the design of a single-phase PWM modulator is a
straightforward task, the voltage of one phase in a threephase
system depends on the switching states of all three
inverter legs. The result of such interaction can be seen
when looking at the individual phase voltages, or at the
a,P components of the switching state vector. These
waveforms are composed of up to five different voltage
levels, with the phase potentials having only the two
levels. The phase voltage at a given time instant should
not differ much from the value of the sinusoidal reference
wave for a well-designed modulation scheme. Generally
speaking, the switched voltage should assume at least the
same polarity as the reference voltage. The polarity consistency
rule is a means to identify ill-designed PWM
schemes.
Optimized Feedforward PWM
PWM inverters of higher power rating are operated at
low switching frequency to reduce the switching losses.
Here, only synchronized pulse schemes can be employed
in conjunction with feedforward schemes in order to avoid
intolerably high subharmonic components. The same applies
for drive systems operating at high fundamental
frequency while the switching frequency is in the lower
kilohertz range. As there are only a few switching instants
t, per fundamental period, small variations of the respective
switching angles uk = w1 . t, have considerable influence
on the harmonic distortion of the machine currents.
Feedback PWM with Optimization in Real-Time
1) Predictive Current Control: Pulsewidth modulation by
predictive current control has common elements with
look-up table methods discussed above. The time of
switching is determined by suitable error boundaries. As
an example, Fig. 19 shows a circular boundary, the location
of which is determined by the current reference
vector i*. When the current vector i touches the boundary
line, the next switching state vector is determined by
prediction and optimization.