TechRef Rectifier
TechRef Rectifier
PowerFactory 2018
I N T EG R AT E D P O W E R S Y S T EM A N A LY S I S S O F T WA R E F O R
T R A N S M I S S I O N / D I S T R I BU T I O N / I N D U S T RY / G EN E R AT I O N / I N T EG R AT I O N O F R EN E WA B L E S
Publisher:
DIgSILENT GmbH
Heinrich-Hertz-Straße 9
72810 Gomaringen / Germany
Tel.: +49 (0) 7072-9168-0
Fax: +49 (0) 7072-9168-88
[email protected]
December 4, 2017
PowerFactory 2018
Revision 1
Contents
Contents
1 General Description 1
1.1.2 Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Short-Circuit Calculations 16
4 Harmonics 17
5 Dynamic Simulation 19
A Parameter Definitions 22
B Signal Definitions 24
C References 25
List of Figures 26
List of Tables 27
1 General Description
11General
General Description
Description
Figure 2: Detailed Circuit with Commutation Reactance and DC Reactance (not part of the model)
Figure
• the2: Detailedline-commutated
three-phase Circuit with rectifier/inverter
Commutation Reactance and DC Reactance (not part of the model)
Figure 1.2: Detailed circuit with commutation reactance and DC reactance (not part of the
The diode rectifier is a full-bridge diode rectifier, which is rectifying the three-phase AC voltage to a 6-pulse DC
model) This converter
voltage. Due to the model
usage ofbasically represents
diodes, which twobedifferent
can neither turned-onthree-phase
nor turned-off converters:
externally, the DC voltage or
DC current of the rectifier can not be controlled.
• the three-phase diode rectifier
The model can be configured as:
The controlled converter model consists of six power thyristors, arranged as shown in Figure 2. These valves can
• the three-phase line-commutated rectifier/inverter
be turned-on by an external control signal (one dash), but only turns-off, when the current flowing through them
becomes negative. This converter can operate as rectifier or as inverter, depending on the control signals applied.
The diode rectifier is a full-bridge diode rectifier, which is rectifying the three-phase AC voltage to a 6-pulse DC
• Three-phase diode rectifier
voltage.
The Due frequency
fundamental to the usage of diodes,
representation which
of this canis neither
model be turned-on
used for load-flow nor turned-off
calculations and stabilityexternally,
analysis. the DC voltage or
DCis current
and describedof
in the rectifier
section can
1.1. The not be
detailed controlled.
modelling of all six thyristors is only necessary for EMT simulations,
• Three-phase line-commutated rectifier/inverter
where the converters are modelled as shown in Figure 2.
The controlled converter model consists of six power thyristors, arranged as shown in Figure 2. These valves can
The diode rectifier is abefull-bridge
turned-on by an external
diode control signal
rectifier, which(oneis
dash), but only turns-off,
rectifying when the current
the three-phase ACflowing
voltage through
to them
becomes negative. This converter can operate as rectifier or as inverter, depending on the control signals applied.
a 6-pulse DC voltage. Due to the usage of diodes, which can neither be turned-on -or
6-Pulse Bridge 4-
turned-off
externally, the DC voltage or DC current of the rectifier cannot be controlled.
The fundamental frequency representation of this model is used for load-flow calculations and stability analysis.
and is described in section 1.1. The detailed modelling of all six thyristors is only necessary for EMT simulations,
The controlled converter
where model consists
the converters of six
are modelled power
as shown thyristors,
in Figure 2. arranged as shown in Figure
1.2. These valves can be turned-on by an external control signal, but only turned-off when the
current flowing through them becomes negative. This converter can operate as rectifier or as
inverter, depending on the timing of the gate signal relative to the AC voltage wave.
6-Pulse Bridge -4-
A fundamental frequency model is used for load flow calculations and stability analysis, and is
described in section 1.1. The detailed modelling of all six thyristors is only necessary for EMT
simulations, where the converters are modelled as shown in Figure 1.2.
The models for load flow calculation and RMS-simulation are based on a fundamental frequency
approach. The equations of the thyristor converter and the diode rectifier are identical if the
diode rectifier is assumed as an uncontrolled thyristor converter (hence the firing angle α is set
to zero).
During steady-state operation the converter can be modelled as a load with constant active and
reactive power P and Q. The following equations describe the converter in a detailed way and
give hints for the layout of an HVDC system.
P d = Ud · Id (1)
The DC voltage of the ideal and uncontrolled converter, without load, is called the “ideal no-load
direct voltage” Ud0 , which is defined as follows:
√
s0 · q π 2
Ud0 = · sin · √ · ULL (2)
π q 3
where s0 defines the number of commutation groups, q is the number of branches in a commu-
tation group and ULL is the AC voltage supplied to the converter station. For a 6-pulse converter
there are two commutation groups (s0 = 2) and q is equal to 3, hence the Ud0 is
√
3· 2
Ud0 = ULL ≈ 1.35 · ULL (3)
π
This equation is valid for the uncontrolled thyristor converter (α = 0) as well as for the diode
bridge. The gate control of the thyristors can be used to delay the ignition of the valves. The
time delay due to the turn-on signal applied is defined to be ωt = α. Then the DC voltage
depends on the ignition angle α
The effect of the ignition angle α is shown in Figure 1.3, where the AC voltage, the phase
currents and the DC voltage can be seen for an idealized operation with the DC current Id
assumed to be constant. The ignition angle is also indicated in the figure. Here the time between
the transfer of the current from one valve to the next is assumed to be zero, i.e the leakage
reactance of the transformer is neglected and the commutation angle µ is zero.
To study more realistic converters the current commutation from one valve to the next must be
considered. The commutation leads to a drop in the DC voltage ∆Ud :
3 3
∆Ud = −Rc · Id = − · ω · Lc · Id = − · Xc · Id (6)
π π
General Description
DIgSILENT
1.50
1.00
0.50
-0.00
-0.50
-1.00
-1.50
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Rectifier: Phase Voltage A/Terminal AC in p.u.
Rectifier: Phase Voltage B/Terminal AC in p.u.
Rectifier: Phase Voltage C/Terminal AC in p.u.
1.50 α
1.00
0.50
0.00
-0.50
-1.00
-1.50
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Rectifier: Phase Current A/Terminal AC in p.u.
Rectifier: Phase Current B/Terminal AC in p.u.
Rectifier: Phase Current C/Terminal AC in p.u.
1.25
1.00
0.75
0.50
0.25
0.00
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Cub_1\Ud_R Measurement: Output Voltage, Real Part in p.u.
Figure 1.3:
Figure 3: Phase
Phasevoltages, phase
voltages, currents
phase and DC
currents voltage
and of a three-phase
DC voltage rectifier rectifier
of a three-phase operating with
operating
α = 30° and ◦ zero overlap angle μ
with α = 30 and zero commutation angle µ
Here the time between the transfer of the current from valve i to the next valve is assumed to be zero, i.e the
leakage reactance of the transformer is neglected and the commutation angle μ is zero.
3
Ud = Ud0 · cos(α) + · Xc · Id (7)
π
Figure 1.4 shows the equivalent circuit for the rectifier including the effects of commutation.
Note in the figure and in the equations above, that the DC current is negative for the rectifier
operation due to the representation with load-orientation.
Rcr
Id
Id
The reactance of the converter transformer is usually the biggest part of total reactance on the
AC side, hence it can be assumed that Xc is approximately that reactance:
Udo cos α Udr
2
ukr Ur,sec
Xc = Xr,sec = · (8)
100 Sr
where ukr , Ur,sec and Sr are the converter transformer short-circuit voltage (in %), rated voltage
on the secondary side and rated power respectively. With the DC current being equal to its
rated value Id , the DC voltage can be written in a different form as:
with
1 ukr
dxr = · (10)
2 100
The term dxr has been calculated from the following relationship:
3 |Id |
dxr = · Xc · (11)
π Ud0
assuming the converter transformer rated current Ir is equal to sqrt(2)/sqrt(3) · Id (see section
1.1.1) and expressing Udo according to equation 3.
The phase voltages and currents, as well as the DC voltage of a thyristor rectifier, including
effects of commutation can be seen in Figure 1.5. The ignition angle α and the commutation
angle µ are indicated in the figure.
General Description
DIgSILENT
1.50
1.00
0.50
-0.00
-0.50
-1.00
-1.50
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Rectifier: Phase Voltage A/Terminal AC in p.u.
Rectifier: Phase Voltage B/Terminal AC in p.u.
Rectifier: Phase Voltage C/Terminal AC in p.u.
1.50 α μ
1.00
0.50
0.00
-0.50
-1.00
-1.50
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Rectifier: Phase Current A/Terminal AC in p.u.
Rectifier: Phase Current B/Terminal AC in p.u.
Rectifier: Phase Current C/Terminal AC in p.u.
1.25
1.00
0.75
0.50
0.25
0.00
0.000 0.004 0.008 0.012 0.016 [s] 0.020
Cub_1\Ud_R Measurement: Output Voltage, Real Part in p.u.
Figure 1.5:
Figure 5: Phase voltages,
Phase phase
voltages, currents
phase and DC
currents andvoltage of a three-phase
DC voltage rectifierrectifier
of a three-phase operating with
operating
◦ ◦
α = 30° and an overlap angle of μ = 20°
with α = 30 and an overlap angle of µ = 20
Using these two angles two other angles can be defined, used in the HVDC theory. γ is called the extinction
angle, which is normally used to control the inverter side of the HVDC.
0 = π −α − μ −γ
Using these two angles two other angles can be defined. The extinction angle γ, which is
normally used in the control on the inverter side of the HVDC, is defined as:
γ =π−α−µ (12)
β =π−α (13)
β is often used in the HVDC controllers for both the rectifier and inverter side.
Using these different angles the DC voltage can be calculated differently for the rectifier and for
the inverter respectively:
cos(α) + cos(α + µ)
Udr = Ud0 · (14)
2
and
cos(β) + cos(γ)
Udi = Ud0 · (15)
2
The DC voltage for the inverter case is considered positive in equation 15. Using equations 14
and 9, the term dxr can be expressed as:
cos(α) − cos(α + µ)
dxr = (16)
2
The phase currents of the 6-pulse bridge are shown in Figure 1.3 and Figure 1.5. In the liter-
ature the AC current is often calculated approximately from the ideal rectifier current with the
commutation angle neglected. In PowerFactory the amplitude of the fundamental frequency
current IL1 is calculated using the Fourier analysis of the phase current waveform, so the effect
of the commutation is taken into account. This leads to the following relationship between the
RMS value of the fundamental frequency component and the direct current:
√
6
IL1 = k · · Id (17)
π
where k is equal to
p
[cos(2α) − cos 2(α + µ)]2 + [2µ + sin(2α) − cos 2(α + µ)]2
k= (18)
4 · [cos(α) − cos(α + µ)]
This factor is close to unity for small values of µ, but if the angle becomes larger, the error
increases up to 4% at µ = 60◦ . For unsymmetrical operation the phase currents have to be
calculated differently, which is described in section 1.1.3.
The power factor cos(ϕ) can then be calculated, given the power equivalence on the AC and
DC side and using equations 3, 14 and 17:
√
√ 3 2 cos(α) + cos(α + µ)
3 · ULL · IL1 · cos(ϕ) = Ud · Id = · ULL · · Id (19)
π 2
1
cos(ϕ) = · [cos(α) + cos(α + µ)] (20)
2k
• Built-in transformer
• External converter transformer
The built-in transformer features a tap-changer on the LV side to control the secondary voltage.
The tap changer setting is specified with the actual winding ratio nntap parameter, which must
be between the maximum and minimum turns-ratio specified in the type. The actual winding
ratio nntap is given in per unit of the nominal turns-ratio tapnom of the converter transformer,
also specified in the type. The resulting voltage at the rectifier AC terminal, UAC,rec is calculated
as:
π
UAC,rec = Uterm · tapnom · nntap · ej·nt2ag 6 (21)
where nt2ag can be defined in the element load flow page and specifies the built-in transformer
phase shift and Uterm is the voltage at the grid terminal (i.e. on the primary side of the built-in
transformer).
The built-in transformer can be configured as Fixed Tap, to control the firing (alpha-control) or
extinction (gamma-control) angle of the converter. In the last two cases, the Relaxation Factor
(krelax) parameter can be entered to speed up (set a factor of greater than 1.0) or slow down
(set a factor of less than 1.0) the tap controller action.
If an external transformer is used for the converter model, the AC voltage drop over the trans-
former is estimated using the commutation reactance specified in the converter dialog window.
If the parameters specified in the external transformer and in the rectifier/inverter fit together,
the same results are obtained as with the built-in transformer.
The commutation reactance is always assumed to be specified on the secondary side of the
transformer. Hence Xc is calculated as follows:
2
ukr Ur,sec
Xc = · (22)
100 Sr
where ukr , Ur,sec and Sr are the converter transformer short-circuit voltage (in %), rated voltage
on the secondary side and rated power respectively. If a built-in transformer is used, the sec-
ondary transformer voltage can be calculated using the parameters Unom and tapnom defined
in the type (it is here assumed that the nominal rated AC voltage Unom is the same as the
built-in transformer rated voltage on the grid side):
If the rated DC voltage of the converter is known, the rated secondary voltage of the converter
transformer can be calculated using the following equation, which can derived from the equa-
tions above:
π Ud
Ur,sec = √ · (24)
3 2 cos(α) − dxr
The rated AC current on the converter side of the transformer equals the RMS value of the total
AC current which, neglecting commutation effects, consists of rectangular pulses with amplitude
equal to the rated DC current Id and duration of 120◦ . The rated AC current on the converter
side of the transformer is found as:
r
2
Ir,sec = · Id (25)
3
Note that this current is the RMS value of the total AC current, and not of only its fundamental
frequency component, which is calculated instead as in equation 17. Hence the rated power
can be calculated as
√ √ π Pd
Sr = 3 · Ir,sec · Ur,sec = 2 · Id · Ur,sec = · (26)
3 cos(α) − dxr
1.1.2 Losses
The model of the HVDC does not include the effects of losses so far. The losses in the converter
bridge are caused due to the different components, i.e. the resistances of valves, transformers,
smoothing reactances.
An exact representation of the losses associated with the converter station is very sophisticated,
so it is common practice to model the losses in the Load Flow analysis as an equivalent series
resistance on the DC side. Two more terms accounting respectively for the forward voltage
drop in the thyristors and no-load losses depending on the DC voltage are also considered. In
PowerFactory , rectifier/inverter losses are specified in fundamental frequency models as:
with:
To take into account losses, equation 7 must be modified, both for rectifier and inverter, as:
Rcr
3 Id
Ud = Ud0 · cos(α) + Xc · Id + resLossF actor · Id + Vdrop (28)
π
Udo cos
The representation of station losses Udr
α of a rectifier for load flow calculations is shown in Figure
1.6. Note in the figure and in the equation above, that the DC current is negative for the rectifier
operation due to the representation with load-orientation.
When the network voltages are unsymmetrical, the periods for natural conduction of the valves
are not the same and the DC voltage will in general be made of six pulses with different duration
and amplitude. The ideal no-load DC voltage can be calculated by taking the average of the
pulses over half a period. As an example, with reference to Figure 1.7, at ωt = θca + π the DC
current starts flowing in phase a and returns back through phase b until ωt = θbc .
For a thyristor converter, these angles can be delayed by αa and αc respectively. During this
period, the DC voltage is equal to the line-line voltage Uab . In PowerFactory , the thyristor
converter implements the Equidistant Pulse Control, igniting the valves at equal time intervals
[1].
The angles θab , θbc and θca , at which the corresponding line-line voltages cross zero and become
positive, are calculated internally in the model, given the terminal voltage phasors. This holds
only for load flow and RMS simulations.
Uab
Ubc
Uca
Ud
Ua
Ub
Uc
Ia
Ib
Ic
Figure 1.7: Currents and voltages for diode bridge rectifier with unbalanced network voltages
and smoothing reactor on the DC side
By taking the average of the resulting three pulses over half a period, the ideal no-load DC
voltage is calculated in load flow and RMS simulations as:
1
Ud = · Uab · [cos(θca + π + αa − θab ) − cos(θbc + αc − θab )]+
π
Ubc · [cos(θab + π + αb − θbc ) − cos(θca + αa − θbc )]+
Uca · [cos(θbc + π + αc − θca ) − cos(θab + αb − θca )] (29)
In unsymmetrical load flow and RMS simulations, a PLL will measure the angle θ1 of the positive
zero-crossing of the positive sequence line-line voltage. The firing angles of the valves are then
calculated in order to obtain an interval between the firing pulses of 60◦ . Therefore, the firing
angles αa , αb and αc will not be identical but differ according to the phase-shift of the phase
voltages. The firing angles of the three phases are calculated internally in the model as (the
negative sign is due to the fact that negative angles imply a lagging phase in equation 29):
In general it is not possible to obtain a 60◦ interval between the firing pulses, since this would
require leading firing angles. As a consequence, for very large unbalances the pulses will still
have different length even after applying different firing pulses for each phase.
The AC currents are affected by the different length of the pulses. Assuming a constant DC
current, the amplitude of the fundamental frequency component of the phase currents is in
general no longer equal because the conduction periods for each phase are different, as can
be seen in Figure 1.7. Let θIi represent the conducting time of phase i. The RMS value of the
fundamental frequency component of each phase current is easily calculated by Fourier series
expansion as:
4 Id θIi
IL1i = k · · √ · sin( ) (33)
π 2 2
The DC power is then calculated as the sum of the real power of all phases on the secondary
side of the transformer:
Pdc = Pa + Pb + Pc (34)
On the Basic Data page a name for the element has to be entered. A type has to be selected or
defined for the element. Furthermore, the orientation has to be specified to allow representation
of either a rectifier or an inverter.
In the basic data page of the type of the inverter/rectifier, the main parameters of the converter
layout have to be entered. You can choose between rated AC or DC voltage and between rated
DC power and DC current. Furthermore the kind of converter can be defined (uncontrolled
diode rectifier or thyristor converter).
If the built-in transformer is chosen (which is advisable for most types of converters), there is as
well the necessity to enter the turns-ratio tapnom of the converter transformer, which is given by
the ratio of secondary to primary voltage, and the nominal firing angle α.
Also the limits of the turns-ratio are given to specify the tap-changer ranges. The maximum
and minimum turns-ratio is given in per unit of the nominal turns-ratio tapnom of the converter
transformer.
In load flow analysis, it is common practice not to specify control variables directly but to define
the controlled variables instead. The control variable (the firing angle α) is then resulting from
the Load Flow calculation.
In the Load Flow command, several common control characteristics are supported by the HVDC
converter model. Meaning and typical application of the various control modes are the following:
• Vdc: The firing angle is adjusted to obtain a predefined value for the DC voltage of the
converter. This control mode is typically used at the inverter side of an HVDC transmission
system.
• Vac: Specifies the magnitude of the AC voltage at the converter terminals, when the DC
voltage is controlled externally. No typical application.
• P: The transmitted AC power is held constant. Typically used for rectifier side in HVDC
systems.
• Q: Specifies the amount of reactive power absorbed by the converter. No typical applica-
tion.
• I: The DC current of the converter is held constant. Typically used for rectifier control of an
HVDC transmission system.
• Gamma: The extinction angle γ is specified. Normally the inverter side of an HVDC
system is controlled to a minimum γ.
• EXT: The firing angle α is specified as an input to the model, provided by a Controller
which must be specified with the parameter pctrl. The Controller is a line-commutated
rectifier/inverter element (ElmRec or ElmRecmono) as well. The EXT control mode is
useful in a 12-pulse arrangement with two converters: one converter is the Controller
performing one of the other control modes, while the second converter is in EXT control
mode.
These control modes are enabled if the flag Automatic Firing Angle Control is selected. Oth-
erwise, the firing angle α is set equal to the Actual Firing Angle, specified with the parameter
alpha set.
Minimum and maximum firing angle limits can be specified. If the converter reaches one of
these limits, the firing angle will remain constant at the limit and the converter cannot perform
the chosen control function.
A minimum value of the extinction angle, gammamin, can also be entered. The angle gam-
mamin represents a safe value of γ in order to avoid commutation failure in normal operating
conditions; it does not represent the limiting value for commutation failure. If the inverter extinc-
tion angle γ reaches this limit, a warning is printed in the output window. For inverters in Vdc
(EXT) control mode, the option Consider minimum extinction angle (gammamin) for control is
available. If the option is selected, the inverter will no longer control the DC side voltage but will
switch to a gamma-control mode with gammamin as setpoint if gammamin is reached.
In load flow calculations, commutation failure is assumed to take place only when the sum of the
firing angle α and of the overlap angle µ would be higher than 180◦ (negative γ). A warning for
commutation failure is printed in the output window. In this case, the load flow is still calculated
assuming the overlap angle equal to zero. However, this may not represent a feasible operating
point.
During load flow calculation, if the thyristor converter current is very low, the converter current
is set to zero. The thyristor converter voltage is set to zero with the firing angle equal to 90◦ . A
message is sent to the output window warning about zero current flowing in the converter.
Furthermore the control of the tap-changers of the converter transformer can be chosen be-
tween:
• Fixed Tap: The position of the tap-changers is fixed to a given winding ratio.
• alpha-control: The secondary voltage is adjusted by the tap-changers to obtain a specified
setpoint of the firing angle. This is typically used at the rectifier station of the HVDC.
• gamma-control: The tap-changers are controlled to obtain a specified setpoint of the ex-
tinction angle. This is typically used at the inverter station of the HVDC.
Besides the firing angle control modes, the load flow page of the converter also comprises
additional information for the converter transformer. Here the commutation reactance Xc is
specified as the leakage reactance of the transformer, which is important for the calculation
of the commutation angle. Also the phase-shift of the converter transformer can be entered
here. This information is needed, when designing 12-pulse thyristor bridges with 30◦ phase-
shift between two converters in series to reduce harmonic currents fed into the network.
Attention: This information is also needed, when no built-in transformer is selected in the
converter type! The value of the commutation reactance is specified as the reactance of the
converter transformer, modelled externally. The value of the commutation reactance is used to
estimate the voltage on the transformer primary side, given the converter terminal voltage and
current. The estimated voltage is used to calculate the ideal no-load DC voltage. Specifying
the correct value of the commutation reactance is also important to get realistic values for the
commutation angle.
When the converter is in P control mode, usually the rectifier side in an HVDC system, the
active power setpoint can be modified by the following controllers if selected:
When the Angle-difference dependent P-droop option is selected, the active power setpoint is
modified. If the option Consider active power setpoint is selected, the new active power setpoint
for the rectifier and inverter case is given according to:
If the option Consider active power setpoint is not selected, the new active power setpoint for
the rectifier and inverter case is given according to:
When the Active power participation option is selected, the active power setpoint is modified.
If the option Consider active power setpoint is selected, the new active power setpoint for the
rectifier and inverter case is given according to:
If the option Consider active power setpoint is not selected, the new active power setpoint for
the rectifier and inverter case is given according to:
Pr = +Kpart · Pmeas
Pi = −Kpart · Pmeas
where:
• Pmeas is the active power measured (assumed positive with load orientation) at a specified
cubicle/boundary.
The Angle-difference dependent P-droop and Active power participation options can be used
to adapt the active power of the converter depending on the active power flow on a parallel
AC line. When the Active power participation option is selected, the sign of the parameter
Kpart depends on the orientation of the power flow at the point where the parameter Pmeas
is measured. Figure 2.1 shows how to correctly define the sign of Kpart. In the example, the
converter INV is performing the active power control and has the Active power participation
option selected.
Figure 2.1: Active power participation example for converter INV. Active power setpoint is
P=10MW.
3 Short-Circuit Calculations
Typically the line-commutated converters are neglected during short-circuit calculations due to
the effect, that the thyristors are automatically blocking during very low voltages at the AC side.
This results in low short-circuit currents supplied by the converter. The calculation methods us-
ing the VDE, IEC or ANSI standards do neglect the contribution of the converters. If a complete
method short-circuit calculation is executed, the short-circuit current of the converter will not be
neglected but defined being the rated AC current of the converter.
Enabling the option Static converter-fed drive, the element can be used to represent in short-
circuit studies reversible static converter-fed drives, with the converter having then a different
layout than a six-pulse bridge. In this case, the contribution of the converter to the short-circuit
current is no longer neglected in the VDE 0102/0103 and IEC 60909 calculation method. Ac-
cording to these standards the converters are assumed to be asynchronous machines having
a short-circuit current ratio of Ishc /Irated = 3 and an R/X-ratio of R/X = 0.1. The short-circuit
current contribution is only considered in symmetrical short-circuits. In case of asymmetrical
short-circuits the current contribution of static converter drives is neglected. The contribution is
only used to calculate the initial and the peak short-circuit current (I” and ip ).
The ANSI and the complete calculation method are not affected by this option.
4 Harmonics
The currents of the 6-pulse thyristor-controlled converter, which are shown in Figure 1.3 with
the commutation effect neglected and in Figure 1.5 including commutation, are characteristic
waveforms. From these curves it can easily be seen, that the currents not only have a large
50-Hz-component but also cause a flow of harmonic currents of higher orders. Hence the most
accurate harmonic model of the HVDC converter is a harmonic current source. The order of the
harmonic currents is calculated as
h=6·n±1 (35)
IL1
Ih = (36)
h
The polarity of the harmonics (angle) is 180◦ for the 5th , 11th , etc. harmonics (represented in
the negative sequence) and 0◦ for the 7th , 13th , etc. harmonics (represented in the positive
sequence).
• Idealised 6-pulse-bridge
The typical spectrum of the converter is assumed up to a specified number (normally 31)
using the above two equations with the commutation reactance neglected. This assump-
tion usually causes the harmonic currents to be larger than in reality, but gives a good
approximation.
• Consider overlap angle
In this model the harmonic currents are obtained same as in the Idealised 6-pulse-bridge
model and then are multiplied with a reduction factor calculated using the overlap angle.
In this case, under typical loading conditions the 11th and 13th harmonics are 30% to 40%
of the harmonics calculated using the Idealised 6-pulse-bridge model [1].
• User defined
To represent the converter in a more realistic way, a harmonic current source can be de-
fined and the amplitude and angle of the harmonic currents can be defined as shown in
Figure 4.1. A balanced and unbalanced representation can be chosen. More information
can be derived from the Technical Reference of the type “Harmonic Sources” (TypHmc-
cur ).
5 Dynamic Simulation
The stability model uses the same equations as described in section 2 (Load Flow analysis).
The converter transformer data, commutation reactance and phase shift, are identical with the
values specified on the load flow page of the element.
A separate minimum extinction angle for commutation failure gammamindyn can be specified
in the RMS page. This angle specifies the minimum extinction angle below which commutation
failure is assumed to take place. When the extinction angle γ reaches the specified gammamin-
dyn, a message warning for commutation failure is printed in the output window, the DC side is
short-circuited and the converter current is assumed equal to zero. Notice that the gammamin-
dyn angle has a different meaning than the gammamin angle specified for control purposes in
the load flow page.
With the signal short dc is possible to short-circuit the DC side of the element, in order to bypass
the valves. The DC voltage and the AC current go to zero during active short dc. With the signal
block all all thyristors will be blocked. The AC and DC current are then both zero.
It is possible to select the way the extinction angle γ is handled when the converter is blocked
through the parameter gammaMode: The angle can either be set to zero or kept constant. An
additional output signal gamma state is available. The value of gamma state is 0 in normal
operation, 1 if the converter is blocked (AC side), 2 if the converter is shorted (DC side) or if in
commutation failure.
alpha
Simulati gamma
tap
on
fref RMS gamma_min
short_dc Simulation gamma_state
block_all
Figure 5.1: Input/Output definition of the HVDC converter model for stability analysis (RMS-
simulation)
5 Dynamic SimulationThe commutation reactance and its angle are identical with the values specified on the load-flow page
element.
With the signal short dc is possible to short-circuit the DC side of the element, through a valve
with resistance equal to Ron . With the signal block all all thyristors will be blocked. The AC and
Attention: Modelling the converter transformer externally (i.e. the “Built-In Transformer” is not used)
DC current are then both zero.
problems due to the exact value of commutation reactance specified on the load-flow pag
additional reactance
The converter transformer data, commutation inserted
reactance and by the transformer
phase element. with
shift, are identical Only the
small errors of this value a
commutation
values specified on the load flow page angle andThe
of the element. hence the EMT extinction
minimum simulation angle
can notgamma-
calculate the right initial condit
it is recommended to use the built-in transformer of the converter element instead to get
mindyn is not used in EMT simulations.
results in the EMT-simulation!
If the built in transformer is used the zero sequence current is always zero.
Attention: Modelling the converter transformer externally (i.e. the flag Built-In Transformer
in TypRec is not selected) can cause problems due to the exact value of commutation reac-
tance specified on the load flow page and the additional reactance inserted by the transformer
element. Only small errors of this value affect the commutation angle and hence the EMT sim-
ulation can not calculate the right initial conditions. Here it is recommended to use the built-in
transformer of the converter element instead to get correct results in the EMT-simulation!
6-Pulse Bridge
alpha
tap
gamma
fref RMS
Simulation gamma_min
short_dc
block_all
gamma
alpha gamma_min
tap Ip_A/B/C
Fmeas EMT Im_A/B/C
Simulation Upc_A/B/C
short_dc
block_all Umc_A/B/C
U0
Figure 5.3: Input/Output definition of the HVDC converter model for stability analysis (EMT-
simulation)
A Parameter Definitions
B Signal Definitions
C References
[1] P. Kundur, Power System Stability and Control. McGraw-Hill, Inc, 1994.
List of Figures
1.2 Detailed circuit with commutation reactance and DC reactance (not part of the
model) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Phase voltages, phase currents and DC voltage of a three-phase rectifier operat-
ing with α = 30◦ and zero commutation angle µ . . . . . . . . . . . . . . . . . . . 3
1.5 Phase voltages, phase currents and DC voltage of a three-phase rectifier operat-
ing with α = 30◦ and an overlap angle of µ = 20◦ . . . . . . . . . . . . . . . . . . 5
1.7 Currents and voltages for diode bridge rectifier with unbalanced network voltages
and smoothing reactor on the DC side . . . . . . . . . . . . . . . . . . . . . . . . 10
2.1 Active power participation example for converter INV. Active power setpoint is
P=10MW. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
5.1 Input/Output definition of the HVDC converter model for stability analysis (RMS-
simulation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5.3 Input/Output definition of the HVDC converter model for stability analysis (EMT-
simulation) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
List of Tables