VSC MTDC Systems With A Distributed DC Voltage Control - A Power Flow Approach
VSC MTDC Systems With A Distributed DC Voltage Control - A Power Flow Approach
VSC MTDC Systems With A Distributed DC Voltage Control - A Power Flow Approach
Abstract—In this paper, a power flow model is presented to both Line-Commutated Converters (LCC) and traditional AC
include a DC voltage droop control or distributed DC slack technology. Contrary to LCC, the VSC technology can support
bus in a Multi-terminal Voltage Source Converter High Voltage the AC grid due to a fast and independent control of active
Direct Current (VSC MTDC) grid. The available VSC MTDC
models are often based on the extension of existing point-to-point and reactive power. The inherent technical characteristics of
connections and use a single DC slack bus that adapts its active VSCs give the technology much better prospects for a multi-
power injection to control the DC voltage. A distributed DC terminal operation than its LCC counterpart. Furthermore, a
voltage control has significant advantages over its concentrated connection to remote wind farms would only be realistic with
slack bus counterpart, since a numbers of converters can jointly VSC technology. As European plans and studies to gradually
control the DC system voltage. After a fault, a voltage droop
controlled DC grid converges to a new working point, which construct such an overlaying DC grid are taken more concrete
impacts the power flows in both the DC grid and the underlying forms, the technical problems yet still unsolved are becoming
AC grids. Whereas current day research is focussing on the predominantly important in current day research on DC grids.
dynamic behaviour of such a system, this paper introduces An outstanding research issues is a distributed control of
a power flow model to study the steady-state change of the the system voltage in such a DC grid. A strong urge has risen
combined AC/DC system as a result of faults and transients
in the DC grid. The model allows to incorporate DC grids in in recent years to thoroughly address this issue and to provide
a N-1 contingency analysis, thereby including the effects of a robust control functions to increase the overall safety of the
distributed voltage control on the power flows in both the AC DC grid [1]. An extension of the control principles of point-
and DC systems. to-point connections gives rise to one converter controlling
the DC voltage at its terminals while the other converters
Index Terms—DC Grids, HVDC transmission, Load flow control their active power injections in the AC grid. The DC
analysis, Voltage control. voltage controlling converter is often referred to as the ‘DC
slack’ converter since the control adapts the output power
I. I NTRODUCTION automatically to compensate for the losses in the DC system.
Such a centralised DC slack converter has to react fast on DC
978-1-4244-8417-1/11/$26.00 ©2011
2
U
Pdc 6dc
Phase X
XX
XXX
Ss reactor S c
XX
XXX
Udc,0 XXX
X
k
Filter
Converter
Pdc
-
Fig. 1. VSC HVDC converter station connected to a DC grid. Pdc,0
grid, the system will start operating in a new operation point Data input
AC grids
determined by the voltage droop characteristics. The droop power flow
Idc =Ydc Udc , (2) The active power injected by the converters in voltage control
depends on the deviation of the bus voltage from its reference
with the DC current vector Idc given by value and is an unknown prior to the power flow. Using Fig.
Idc = [Idc1 , Idc2 . . . Idck , 0 . . . 0 ]T , (3) 2, the active power injection can be written as
1
working converters outage
Pdci =Pdc,0i − (Udci − Udc,0i ), (6)
ki
with n − k zero elements due to converter outages and DC
buses without a power injection. The DC voltage vector is with ki the voltage droop, defined as ΔUdc /ΔPdc .
given by Udc = [Udc1 , Udc2 . . . Udcn ]T and Ydc is the DC bus This system of non-linear equations can be solved with a
matrix. When line outages are taken into account, this involves Newton-Raphson (NR) method, as shown in the next section.
altering the DC bus matrix accordingly. For a monopolar,
symmetrically grounded DC grid, the power injections become IV. S EQUENTIAL POWER FLOW WITH DISTRIBUTED DC
VOLTAGE CONTROL
Pdci =2Udci Idci , ∀i ≤ k. (4)
In this section, the DC grid model with a distributed voltage
A similar expression holds for a bipolar configuration when control is introduced in the sequential AC/DC power flow
in steady-state operation. algorithm from [8]. Fig. 3 shows the flow chart of the power
The current injections Idc are unknown prior to the DC flow algorithm when a distributed voltage control scheme
grid power flow. Instead, the active power injections Pdc of is included. The modifications to the sequential power flow
the converters in P -control are calculated using the results algorithm from [8] are printed in bold. In [8], one converter
of the AC power flow. However, the DC slack bus active was assigned the task of the DC slack converter, adapting its
power injection Pdc is not known prior to the DC power flow. power to control the DC system voltage. When a DC voltage
When there are converters with a distributed voltage control, droop is applied by other converters, it is still possible, but not
the active power injection Pdc at these buses is not known necessary, to include a slack bus in the DC grid. From a control
prior to the DC power flow either. point of view, this stems with a converter that controls its DC
4
bus voltage to the reference value using a PI controller, whilst with Pc the active part of the complex power injected at the
the converters with a DC voltage droop use a proportional converter side, shown in Fig. 1.
gain. In this section, the equations are written in their most During the iteration, this expression does not hold for the
general form, with the first converter being a regular DC slack DC slack converter and the m−1 converters under DC voltage
converter. droop for reasons explained above. However, the intermediate
The remainder part of the section only briefly discusses the AC grid states can be used for estimating the losses of these
sequential algorithm as such. Instead, emphasis is put on the converters to start the iteration that follows the DC power flow,
modifications introduced by to the distributed voltage control. shown in Fig. 3.
133.6 84.3 45+j15 40+j5 133.3 45+j15 40+j5 132.9 84.5 45+j15 40+j5
84.3
~ 1 3 4 ~ 1 3 4
~ 1 3 4
35.3 34.2 23.1 23.0 29.1 28.3 43.9 43.7 33.8 32.8 35.1 35.0
13.0 15.1 4.6 6.5 0.1 0.3 14.6 17.5 3.1 4.5 14.8 0.7 13.3 15.6 5.9 7.5 11.9 1.6
98.4 71.4 104.2 69.7 99.1 71.2
(a) Normal operation (b) Outage: P -control (c) Outage: DC voltage droop
Fig. 4. AC network power flow solution with a VSC MTDC system between buses 2, 3 and 5: (a) Normal operation, (b) Converter outage on bus 5 &
P -control, (c) Converter outage on bus 5 & DC voltage droop on converters 2 and 3. Legend: → Active power (MW) and Reactive power (MVAr).
After convergence, the voltages on all DC buses are known. 5 are under constant P and Q-control. The converter at bus 3 is
Thereafter, the power injections of the slack bus and the DC under constant U -control and is also designated the role of the
voltage droop buses are calculated using (5). DC slack converter. The DC voltages and DC power injections
from this power flow are used as the reference values for the
C. Additional converter loss iteration DC voltage droop control later on. Figs. 4b & 5b show the
results of the AC and DC test system with a converter outage
After having calculated all unknown DC grid voltages and on bus 5. As expected, the power of the slack converter at
powers, an additional iteration is needed to calculate the bus 3 adapts its power, whereas the power injected by the
AC grid power injection of the DC slack converter and the converter at bus 2 remains unaltered. In Figs. 4c & 5c, a
voltage droop converters since the AC powers injected by voltage droop control has been implemented on all converters.
these converters depend on the converter losses, which are The converters remaining in operation, namely those on buses
not known beforehand. 2 and 3 have a voltage droop that has been respectively set to
Due to the extension of the algorithm, the overall con- 5e-3 and 7e-3 p.u./MW. As shown by the results, the remaining
vergence criterium has to be based on the slack converter converter powers are both lowered, dictated by their droop
power, if present, as well as on the voltage droop buses’ power characteristics. As expected, the droop control also alters the
injections in the AC grid, as shown in Fig. 3. power flows in the AC grid, shown in Fig. 4.
When the droop control is implemented in a larger DC
V. S IMULATION RESULTS network, the contribution of each converter to the DC voltage
This section discusses the effect of a converter outage on control can be adapted by altering its droop characteristic.
the power flows in the AC and DC grid. The extension of the
sequential power flow algorithm allows to study the effect of VI. C ONCLUSION
the droop characteristics on the overall power flows after a In this paper, a model has been presented to include voltage
disturbance. droop characteristics in power flow calculations. The model
The DC voltage droop control has been included in a has been implemented and tested in a sequential AC/DC power
sequential AC/DC power flow algorithm, using MATPOWER flow algorithm. The implementation of the droop characteris-
[10], an open-source power flow toolbox for Matlab. The tics allows to extend contingency analyses to DC grids and
simulation results, shown in Figs. 4 – 5, are based on a 5-bus makes it possible to study the effects of the distributed voltage
test network from [11] with a 3-bus VSC MTDC network. control and the individual droop values of each converter on
Figs. 4a & 5a show the results of the AC and DC test system the post-disturbance power flows in both the AC and DC
under normal operation conditions. Converters on buses 2 and system. The droop characteristics have been included in a
6
MATPOWER based AC/DC power flow program and properly Jef Beerten (S’07) was born in Belgium in 1985.
tested. Numerical simulations show the effect of the distributed He received the M.Sc. degree in electrical engi-
neering from the Katholieke Universiteit Leuven
voltage control on the power flows in the system. (K.U.Leuven), Leuven, Belgium, in 2008, where he
is currently working towards the Ph.D. degree. He
is a Research Assistant with the division ESAT-
R EFERENCES ELECTA division of K.U.Leuven. His research in-
[1] D. Van Hertem and M. Ghandhari, “Multi-terminal VSC HVDC for terests include power system control, integration
the European supergrid: Obstacles,” Renewable and Sustainable Energy of wind energy, DC grids and multi-terminal VSC
Reviews, vol. 14, no. 9, pp. 3156–3163, Dec. 2010. HVDC in particular. Mr. Beerten is an active mem-
[2] C. Dierckxsens, “VSC MTDC systems with distributed slack bus,” ber of both the IEEE and Cigré.
Master’s thesis, K.U.Leuven, Jun. 2010. Mr. Beerten holds a Ph.D. fellowship from the Research Foundation
[3] R. Hendriks, G. Paap, and W. Kling, “Control of a multi-terminal VSC Flanders (FWO).
transmission scheme for connecting offshore wind farms,” in Proc.
European Wind Energy Conference & Exhibition, Milan, Italy, May,
7 – 10 2007.
[4] T. Haileselassie and K. Uhlen, “Primary frequency control of remote
grids connected by multi-terminal hvdc,” in Proc. IEEE PES GM ’10, Dirk Van Hertem (S’02, SM’09) graduated as a
Minneapolis, USA, Jul. 25–29, 2010. M.Eng. (industrieel ingenieur) in Electro-mechanical
[5] A. van der Meer, R. Hendriks, and W. Kling, “Combined stability and engineering in 2001 from the KHK, Geel, Belgium
electro-magnetic transients simulation of offshore wind power connected and as a M.Sc.(burgerlijk ingenieur) in Electrical
through multi-terminal VSC-HVDC,” in Proc. IEEE PES GM ’10, Engineering from the K.U.Leuven, Belgium in 2003.
Minneapolis, USA, Jul. 25–29, 2010, pp. 1 –7. In January 2009, he obtained his PhD, also from the
[6] T. Haileselassie, K. Uhlen, and T. Undeland, “Control of multiterminal K.U.Leuven. From 2010 to spring 2011, Dirk Van
HVDC transmission for offshore wind energy,” in Proc. Nordic Wind Hertem was a member of the Electric Power Sys-
Power Conference Presentation, Rønne, Denmark, Sep. 10–11, 2009. tems division at the Royal Institute of Technology,
[7] G. Asplund, B. Jacobson, B. Berggren, and K. Lindén, “Continen- Stockholm. Currently, he is working again as a post-
tal overlay HVDC-grid,” in CIGRE 2010 Session Proceedings, Paris, doctoral researcher in the ELECTA research group
France, 2010. at the K.U.Leuven, Belgium.
[8] J. Beerten, S. Cole, and R. Belmans, “Generalized steady-state VSC His special fields of interest are power system operation and control in
MTDC model for sequential AC/DC power flow algorithms,” IEEE systems with FACTS and HVDC and building the (super-)grid of the future
Trans. Power Syst., 2011, submitted for publication. with large amounts of renewable energy sources. Dr. Van Hertem is an active
[9] C. Barker and R. Whitehouse, “Autonomous converter control in a multi- member of both IEEE and Cigreé.
terminal HVDC system,” in Proc. IET ACDC ’10, London, UK, Oct.
20–21, 2010.
[10] R. D. Zimmerman, C. E. Murillo-Sanchez, and R. J. Thomas, “MAT-
POWER: Steady-state operations, planning, and analysis tools for power
systems research and education,” IEEE Trans. Power Syst., vol. 26, no. 1, Ronnie Belmans (S’77-M’84-SM’89-F’05) re-
pp. 12–19, Feb. 2011. ceived the M.Sc. degree in electrical engineering in
[11] G. W. Stagg and A. H. El-Abaid, Computer Methods in Power System 1979 and the Ph.D. degree in 1984, both from the
Analysis. Kogakusha, Japan: McGraw-Hill, 1968. K.U.Leuven, Belgium, the Special Doctorate in 1989
and the Habilitierung in 1993, both from the RWTH,
Aachen, Germany. Currently, he is a full profes-
sor with the K.U.Leuven, teaching electric power
and energy systems. His research interests include
techno-economic aspects of power systems, power
quality and distributed generation. He is also guest
professor at Imperial College of Science, Medicine
and Technology, London-UK.