21, rue d’Artois, F-75008 PARIS C1-324 CIGRE 2016

http: //www.cigre.org

Improvement of mode controllability and short-circuit currents limitation

in metropolises power grid by means of electromechanical AC links as an
alternative to DC links

Y. DEMENTYEV, N. PINCHUK, V. DYACHKOV, A. MAYOROV,

Y. SHAKARIAN, V. NOVOZHILOV, Y. KUCHEROV, A. SHABASH
P. SOKUR V. TRETYAKОV D. YAROSH Energocomplex
Scientific and Power Machines System Operator of Russia
Research Center of Russia the UPS
Federal Grid Russia
Company of Unified
Energy System
Russia

SUMMARY

The concentrated power grid of almost any megapolis is characterized by considerable

power consumption, large installed generating capacity, and relatively short power lines rated
110–500 kV. Russian and international experience shows that megapolises’ electricity
consumption is steadily on the increase. As a result, the rated level of short-circuit currents in
the main electricity grid of megapolises’ power systems often exceeds the maximum rated
interrupting current for commercially available circuit breakers.
For limiting short-circuit currents to an acceptable level, limiting reactors and special
devices for deep current limitation can be used. Nevertheless, the most common (forced)
measures to restore the rated conditions are to partition the electricity grid, up to splitting the
power grid into isolated fragments. Partitioning the electricity grid makes it possible to reduce
the levels of short-circuit currents but decreases the possibility of the redundant power supply
to individual substations and power districts, resulting in reduced electricity supply reliability.
There are proposals to use voltage converter–based DC links for joining grid sections
that are disconnected due to the limitation of short-circuit currents. This solution makes it
possible to limit short-circuit currents and provide predetermined active power flows and
reactive power control, the need for which grows as cable grids develop. The main
disadvantages of this technical solution are the high price and significant dimensions, which
causes difficulties in land allocation for substations to accommodate a DC link in urban areas.
An almost complete functional equivalent of a DC link as to its technical
specifications is an electromechanical AC link based on double-feed electrical machines. The
electromechanical link is a unit that has two electrical machines on a common shaft. Each of
the machines can be operated as a generator and a motor. Vector excitation control of each
machine provides predetermined active power flows and reactive power control at two points
of their connection. In addition, the electromechanical AC link prevents the transmission of
reverse- and zero-sequence currents and higher-harmonic currents and voltages from one grid
to another, actually providing a galvanic isolation of connected grid sections.
We have carried out preliminary research and engineering studies for an
electromechanical AC link with a rated capacity of 100 MW and 200 MW. The
[email protected]
electromechanical link is a vertical (or horizontal) unit consisting of the two asynchronous
motor generators. The vertical design makes it possible to minimize the area. The rated rotor
speed is 750 rpm. The unit is completely air cooled. The excitation system of each motor
generator supplies power to the three-phase rotor with a current of a variable frequency from
0 to 0.25 Hz. For starting the unit from the idle position a static frequency converter is used.
The time for active power increase from zero to the rated value is not more than 0.3 second.
The project to use this device in the Moscow electricity supply system is currently under
consideration.

KEYWORDS

Short-circuit current, AC link, DC link, reactive power, reliability, double-feed

machine, vector control.

1
 The development of Moscow’s power grid, characterized by the increased installed
capacity of power plants and substations and the increased number of high-voltage grids, has
created the problem of elevated levels of short-circuit currents in them. For instance, the
growth of short-circuit currents in 110 kV and 220 kV high-voltage grids became stable in the
mid-20th century. There was a need for special measures to coordinate the levels of short-
circuit currents by the 1970s, i.e. to harmonize the specifications of applied electrical
equipment and circuit designs. At present, this problem is particularly acute for the 220 kV
grid of the capital, which is determined by a high rate of its development, the widespread use
of cable products, the commissioning of new generation facilities, and an increase in
autotransformer capacity.
The rising levels of short-circuit currents in the development of a power grid set high
requirements to the electrodynamic and thermal resistance of components of power grids’
electrical devices and to the switching capability of electrical equipment. Additionally, there
are pressing issues related to the impact of short-circuit currents not only on rigid busbars,
cables, and electrical equipment, but also on generators, power transformers, flexible
conductors of electrical installations, and grounding systems of electrical installations.
For 50 years and until recently, the only measure to limit the short-circuit currents in
the Moscow power system was grid splitting. At the present time, the Moscow power system
has about 100 and more than 40 sections in the 110 kV and 220 kV grids respectively. The
500 kV grid remains of the closed type, with the levels of short-circuit currents nearly
equalling the interrupting capacity of 500-kV circuit breakers at some substations.
Grid splitting should be regarded as a forced measure to limit short-circuit currents. It
is a mistake to believe that grid-splitting measures are cost free because the disconnection of
power grid equipment (open busbars at substations and power plants, power line
disconnection) impacts badly the power supply reliability and the capacity delivery reliability
of power plants, reducing as well the power system control flexibility, and breaking the
optimal (economical) flow distribution in the grid. It is evident that the most important current
goal in developing the Moscow power grid is to formulate the scenario that will substantially
limit the levels of short-circuit currents while maintaining the required reliability,
controllability, and cost effectiveness characteristics of a high voltage grid.
Effective measures to limit short-circuit currents are DC links. Furthermore, they
allow predetermined active power flows through a link. Modern voltage converter–based DC
links also make it possible to control reactive power at both ends.
The disadvantages of the DC link are as follows:
 generation into the harmonic grid;
 a large area of installation;
 limited overload capacity;
 high costs.
However, the full functional equivalent of the DC link in terms of voltage conversion
(based on IGBT transistors) is an electromechanical AC link based on double-feed machines
(“AC link”).
The AC link is a unit that has two double-feed machines on a common shaft. Fig. 1
shows a schematic diagram of the AC link.

2
 Fig. 1. Electromechanical AC link.

While the first electrical machine operates as a generator, the second one operates as a
motor. When the direction of the capacity flow changes, due to the reversibility of the
electrical machines, the first machine will switch to the motor mode and the second one will
switch to the generator mode.
For example, if the current frequency is f=48 Hz in one power grid and is f=50 Hz in
the other, the rotor shaft rotates at a frequency equal to half the sum of the frequencies (49
Hz). This is achieved by rotating the excitation field in the first machine against the direction
of rotor rotation with a frequency of 1 Hz and in the direction of rotation with a frequency of
1 Hz in the second machine.
If one power grid experiences any dynamic disturbances (e.g. short circuits), then they
are contained in this power grid and are not transferred to the other one because there is no
galvanic connection between the joined grids.
Furthermore, each of the electrical machines makes it possible to independently
control the voltage in each power grid by operating in modes from the delivery to the deep
consumption of reactive power.
If two power districts having the same frequency are interconnected, the rotor will
rotate at a synchronous rotation speed (or with a minimal slip to ensure that excitation
windings are heated uniformly).
Structurally, the electrical machine is an asynchronous machine with a three-phase
rotor. Excitation is controlled by the vector principle [1]. The vector of the excitation current
is generated in the coordinate axes “x, y” rotating with the power frequency s (Fig. 2).

3
 Fig. 2. Vector diagram.

It is known that the projections of the excitation current vector to the coordinate axes
x, y are proportional to the active power P and reactive power Q of the machine stator:
P  x
i fy  
U S  x ad (1)
2
Q  x  U C
i fx 
U S  x ad
where xad is the mutual resistance of stator and rotor windings; x is the resistance of
the self-induction of the stator’s winding plus the resistance of the stator’s external circuit.
The ratio (1) shows that independent control of the excitation current projections i fx
and ify provides independent control of the active power P (electromagnetic torque) and
reactive power Q (voltage in the stator circuit).
The control signals Ufx and Ufy form the voltage excitation vector Uf in the
synchronous axes “x, y.” For excitation control, these signals are converted into the
coordinate system “d, q” associated with the rotor with the angular position :
U fq  U fy cos  U fx sin 
(2)
U fd  U fy sin   U fx cos
As a result, the AC link is controlled through four control channels (Fig. 3), which
makes it possible to control the following parameters:
 active power flow Pa;
 voltage U1 (reactive power Q1);
 voltage U2 (reactive power Q2);

4
  determined rotor speed equal to half the sum of the frequencies of machine 1
and machine 2.

Fig. 3. AC link control channels.

At present, two options of an AC link are under development: with a rated capacity of
100 MW (ASEMPC 486/200-8) and 200 MW (ASEMPC 486/400-8).
The AC link contains the following basic equipment (Fig. 4):
 step-up package transformer (T1, T2);
 double-feed machine (DF-1, DF-2);
 excitation system (ES1, ES2) with an excitation transformer (TES1, TES2);
 thyristor starter (TS) with a transformer (TSFC).

5
 Fig. 4. AC link block diagram.

The specifications of a 100 MW and 200 MW AC link are provided in Table 1. Fig. 5
shows an outline drawing of AC link with a capacity of 100 MW and 200 MW.
Since the AC link is used in the electricity supply system of a megapolis where the
mains frequency is the same at both ends of the link, it is not necessary to provide a wide
range control of the rotor speed. The allowable slip relative to the synchronous speed is
limited to s=0.005 relative units for dynamic applications and the uniform heating of
excitation windings. The capacity of the excitation system is 2% of the rated capacity of the
unit. The excitation system of each motor generator supplies power to the three-phase rotor
with a current of a variable frequency from 0 to 0.25 Hz. For starting the unit from the idle
position a thyristor starter is used. The time for active power increase from zero to the rated
value is not more than 0.3 second.

Table 1. AC link specifications.

Type ASEMPC 486/200-8 ASEMPC 486/400-8

Rated capacity, MW 100 200

Rated capacity factor 0.85 0.85
Linear stator winding voltage, V 15,750 15,750
Stator current phase, A 4,311 8,625
Excitation current, A* 2,270 2,210

6
Excitation voltage (s=0/s=0.005), V* 130/390 180/730

Rated rotor speed, rpm 750 750

Possible range of motor slip changes ± 0.005 ± 0.005

in each machine, relative units

Vertical Vertical
Version
(horizontal)
Estimated unit weight, tonnes 680 1100
Cooling air air
* - phase amplitude

7
 Fig. 5. 100 MW and 200 MW AC link dimensions.

Fig. 6 shows a diagram of allowable operation modes for a 200 MW AC link. As

demonstrated by the diagram, the link provides reactive power control over a wide range.
With the rated power ±97 MVAr, with P=100 MW from 160 MVAr to -198 MVAr. This
capability is especially useful for megapolises’ electricity supply systems using high-voltage
cable lines with a large charging capacity.

8
 Fig. 6. Diagram of allowable operation modes for a 200 MW AC link.
1 – rated voltage;
2 – voltage decrease -5%;
3 – voltage increase +5%.

Fig. 7 shows the simulation results for a three-phase short circuit of 0.18 seconds. As
shown by the simulation results, the voltage in grid 2 remains stable with a short circuit in
grid 1.
It should be noted that, since the electrical machine is a source of electromotive force,
the AC link creates a fault-current contribution. The greatest value of the fault-current
contribution at the short-circuit point of the 220 kV side is:
5.6 kA in the case of the 200 MW link;
2.4 kA in the case of the 100 MW link.
In spite of its own fault-current contribution, the AC link is an effective means of
limiting short-circuit currents because it prevents a fault-current contribution from the
knowingly larger equivalent electromotive force of the connected power district. For example,
when two disconnected sections of the 220 kV grid in the Moscow CHPP-20 area (points 1
and 2, Fig. 8) are joined, a short-circuit current increases to 77 kA. Connecting these grid
sections through an AC link allows limiting the short-circuit current to acceptable 49 kA (with
fault-current contribution from the AC link).

9
 Fig. 7. Simulation of AC link operation with a short circuit.

Fig. 8. Equivalent design diagram of short-circuit currents with an AC link.

An important characteristic of equipment used in a megapolis is its dimensions. The

cost of land or the absence of available spaces can be the deciding factor when choosing a

10
particular type of equipment. Table 2 contains information on the installation area of 200 MW
AC and DC links. The area occupied by an AC link is far smaller than that of a DC link.
In the emergency mode, an AC link has advantages over a DC link by withstanding a
short-term overload of up to a twofold value.

Table 2. Comparison of the main AC and DC link specifications.

Device Type AC Link DC Link
Rated capacity, MW 200 200
Rated capacity factor 0.85 0.85
Short-term twofold overload yes no
Installation area, m2 736 4,770

The calculations show that the use of an AC link in the Moscow 220 kV power grid
makes it possible to:
 eliminate the partitioning points where the link is installed;
 reduce short-circuit currents in the places adjacent to the 220 kV link
installation site (by 12 kA with one 200 MW link installed, by 30 kA with two
links installed, by 45 kA with four links installed);
 ensure the additional ability to control the voltage in a 220 kV grid (in the
range of –4 kV to +8 kV) and, consequently, reduce the required additional
equipment for reactive power compensation;
 ensure that active power flows be controlled through the link;
 expand the range of allowable modes in a backbone grid, making it possible to
control active power flows through 220 kV shunt links.

CONCLUSION

1. The electromechanical AC link is a voltage inverter–based functional equivalent of

a DC link and has the following main advantages over the latter:
 no higher harmonics;
 allowable short-term current overload of up to a twofold value;
 smaller installation area and lower costs.
2. Design work is currently underway on installing an AC link at a Moscow
substation.

BIBLIOGRAPHY

[1] Y. Shakaryan, P. Sokur. “Experience in the development and operation of

asynchronized turbogenerators and condensers in the Russian Power System” (Archives
of electrical engineering Vol. 64(1), 2015, pages 7-15)

11