Jenkins N., Strbac G., Ekanayake J. Distributed Generation

Подождите немного. Документ загружается.

factor, the voltage profile is improved. The voltage profile can be improved further

by operating the distributed generator at 0.95 lagging pf (exporting VArs). The

converse is true when the distributed generator operates at a leading power factor

(importing VArs). Under minimum loading conditions the case without the dis-

tributed generator shows minimum losses and a good voltage profile. The connec-

tion of the distributed generator increases the voltages at the busbars and excess

power is now transmitted back to the grid thus increasing the losses.

2. Effect on load compensation

As shown in Table 3.6, the voltages at remote busbars when the distributed gen-

erator is not in operation are below the limit. It is customary to use line drop

compensation (LDC) on the automatic voltage controller (AVC) of the on-load tap

changing (OLTC) transformer to address this [16]. The LDC controls the voltage at

a remote point downstream of the OLTC transformer. One concern utilities have

over distributed generator connection is possible interaction with LDC when the

distributed generator feeds power back to the grid. This is a reversal of the con-

ventional power flow to which the AVC with LDC is designed to respond.

This case was simulated by designing the AVC with LDC to maintain the

voltage at Bus 6 at 0.95 pu. Figure 3.33 shows the voltages at each busbar for peak

loading condition without the distributed generator and minimum loading condition

with the DG (operating at 0.95 lagging pf). As can be seen from the top diagram of

Figure 3.33, the voltage at Bus 6 has increased from 0.932 to 0.953, thus showing

the expected LDC operation. However, when the DG is connected the voltage at

Without DG and peak loading

condition

Bus 1 Bus 2

Bus 3 Bus 4 Bus 5 Bus 6

Bus 1 Bus 2

Bus 3 Bus 4 Bus 5 Bus 6

For minimum loading condition and

DG operatin

at 0.95 la

1.005 1.024 0.980 0.956 0.954 0.953

1.005

0.993

1.021 1.018 1.018 1.017

P = 6.79

Q = 1.67

P = 7.99

Q = 1.98

P = 0.95

Q = 0.23

Bus 7

Figure 3.33 Voltages with LDC (P in MW and Q in MVAr)

Distributed generators and their connection to the system 87

Bus 6 becomes much higher than the reference setting. This can be explained using

the voltage reference signal to the AVC that is given by Reference 16:

ref

¼ V

þ I

R þ I

X ð3:49Þ

where V

ref

is the reference signal to the AVC, V

is the measured voltage of the

transformer secondary, I

and I

are the active and reactive component of the

current flowing from secondary side to primary side of the generator respectively,

and R and X are the compensation resistance and reactance respectively.

When the DG is connected, both active and reactive currents of line section 2-3

reverse (see the lower diagram of Figure 3.33), whereas the power in other lines

remains the same as in the previous case (but lower values due to minimum loading).

The dominant effect comes from the line section 2-3 where I

, R and X are much

greater than that of other line sections. The reverse power on this line section increases

the reference setting of the AVC.

Thus, for this example, LDC continues to work satisfactorily with a distributed

generator connected to one of the outgoing circuits. This may not always be the

case with multiple outgoing circuits or if the sign of the I

X term is made negative

to give negative reactance compounding [16].

3. Fault studies

A three-phase fault was applied to Bus 5 with and without DG and fault current flows in

MVA are shown in Figure 3.34. It can be seen that with DG the fault current flowing

through switchgear at busbars 3 and 4 has increased from 72 MVA to 91 MVA.

Without DG

Bus 1 Bus 2 Bus 3 Bus 4 Bus 5 Bus 6

DG connected to Bus 3

Bus 1

Bus 2

Bus 3

Bus 4

Bus 5

Bus 6

1.005

63.27 MVA

91.93 MVA

28.72

MVA

72 MVA 72 MVA 72 MVA

Fault

Bus 7

Figure 3.34 Fault current flow to a three-phase fault at Bus 5

88 Distributed generation

3.4.2 Electromagnetic transient studies

The use of mechanically switched capacitors for power factor correction of

induction generators has a number of difficulties. If the degree of compensation

approaches the no-load reactive power requirement of the generator, then there is a

danger of self-excitation as the capacitors can only be switched occasionally and in

discrete steps. The reactive power supplied by a fixed shunt capacitor reduces with

the square of the network voltage so that during network voltage depressions, just

when additional reactive power support is needed, the capability of fixed capacitors

to provide reactive power is reduced. Conventional static var compensators

(SVCs), which use thryristor switched capacitors, have a similar voltage/reactive

power characteristic [17]. These limitations can be overcome if the reactive power

required is supplied by a power electronic compensator rather than by capacitors.

The STATCOM is a voltage source converter based reactive power compensator,

which may be used to generate (or absorb) controlled reactive power [17]. It con-

sists of a voltage source converter and a coupling reactor. The STATCOM operates

by creating a voltage of controlled magnitude and phase at the converter terminals,

and so exchanges real and reactive power with the network across the coupling

reactor. Just as with a synchronous generator, the real power exchanged is con-

trolled by the angle between the converter output voltage and the network voltage,

while the reactive power exchanged is controlled by the relative magnitudes of the

two voltages.

The VSC of the STATCOM is switched at a high frequency and the resulting

output voltage is non-sinusoidal and contains harmonics. For accurate modelling

of such system the load flow and fault calculation packages that solve a set of

algebraic equations are not adequate. Usually a numerical integration of a set of

differential equations is used to simulate this type of circuit. The analysis of this

nature is called electromagnetic transient simulation and they normally demand a

smaller time step for simulations.

Figure 3.35 shows the power circuit of the STATCOM. The voltage source

inverter is a six pulse IGBT-based converter, which operates with space vector

pulse width modulated (PWM) modulation. The STATCOM was simulated using

the electromagnetic transient simulation programme EMTDC/PSCAD. Figure 3.36

shows the STATCOM terminal voltage (V

), injected current (I

) and current

through a switch (I switch) when it is generating only reactive power.

Volta

e source inverter

Grid supply

STATCOM current (I

)

Figure 3.35 STATCOM used for simulations

Distributed generators and their connection to the system 89

A3.1 Appendix: Unbalanced faults

Line-to-line fault

For a line-to-line fault as shown in Figure 3.37, I

¼ 0, I

¼I

¼ I

and

 V

¼ I

STATCOM-ES

Terminal voltage

+1.25

+0.75

+0.25

−0.25

−0.75

−1.25

0.759

0.767

0.771

0.775

0.7790.763

Injected current and switch current

I_switch

0.759 0.763 0.767 0.771 0.775 0.779

Time

(

)

−0.04

+0.04

Figure 3.36 STATCOM voltage and current

Figure 3.37 Line-to-line fault

90 Distributed generation

From symmetrical components:

11 1

1 ll

1 l

I

¼0 and I

¼I

ð3:50Þ

Substituting the symmetrical components of V

and V

into V

 V

¼ I

the following equation was obtained:

ðl

 lÞþV

ðl  l

Þ¼I

ð3:51Þ

Since I

¼ I

, from symmetrical components:

¼ I

þ lI

þ l

ð3:52Þ

Substituting for sequence cur rent from (3.50) into (3.52):

¼ðl  l

ÞI

ð3:53Þ

Substituting for I

from (3.53) into (3.51) and divided by ðl  l

Þ, the fol-

lowing equation was obtained:

 V

¼ I

ð3:54Þ

From (3.50) and (3.54), the connection of sequence networks for a line-to-line

fault was obtained (Figure 3.38).

′

Figure 3.38 Connection of sequence networks under a line-to-line fault

Line-to-line-to-earth fault

For a line-to-line-to-earth fault shown in Figure 3.39, V

¼ V

¼ 0.

Distributed generators and their connection to the system 91

From symmetrical components:

11 1

1 ll

1 l

¼ V

ð3:55Þ

Equation (3.55) suggests that three sequence networks should be connected in

parallel as shown in Figure 3.40.

References

1. Chapmen S.J. Electrical Machinery Fund amentals . McGraw-Hill; 2005.

2. Hindmarsh J. Electrical M achines and their Applications. Pergamon Press;

1970.

3. McPherson G. An Introduction to Electrical Machines and Transformers.

John Wiley and Sons; 1981.

′

Figure 3.40 Connection of sequence networks under a line-to-line-earth fault

Figure 3.39 Line-to-line-earth fault

92 Distributed generation

4. Grainger J.J., Stevenson W.D. Elements of Power Systems Analysis.

McGraw-Hill; 1994.

5. Weedy B., Cory B.J. Electric Power Systems. John Wiley and Sons; 2004.

6. Hurley J.D., Bize L.N., Mummert C.R. ‘The adverse effects of excitation

system Var and Power Factor controllers’. Paper No. PE-387-EC-1-12-1997.

Presented at the IEEE Winter Power Meeting; Florida, 1997.

7. Heier S. Grid Integration of Wind Energy Conversion Systems. John Wiley

and Sons; 1998.

8. Allan C.L.C. ‘Water-turbine driven induction generators’. Proc. IEE, Paper

No. 3140S, December 1959.

9. Olimpo Anaya-lara O., Jenkins N., Ekanayake J., Cartwright P., Hughes M.

Wind Energy Generation Modelling and Control. John Wiley Press; 2009.

10. Burton T., Sharpe D., Jenkins N., Bossanyi E. Wind Energy Handbook. John

Wiley and Sons; 2001.

11. Kundur P. Power System Stability and Control. McGraw-Hill; 1994.

12. Krause P.C., Wasynczuk O., Shudhoff S.D. Analysis of Electric Machinery

and Drive System. John Wiley Press; 2002.

13. Ekanayake J.B., Holdsworth L., Jenkins N. ‘Control of DFIG wind turbines’.

Power Engineer. 2003;17(1):28–32.

14. National Grid Company plc. The Grid Code. Issue 3, Revision 25, 1 February

2008.

15. Anderson P.M. Analysis of Faulted Power Systems. IEEE Press; 1995.

16. Hingorani N.G., Gyugyi L. Understanding FACTS: Concepts and Technology

of Flexible AC Transmission Systems. Wiley-IEEE Press; 1999.

17. Thomson M. ‘Automatic voltage-control relays and embedded generation. Part

II’. Power Engineering Journal [see also Power Engineer]. 2000;14(3):93–99.

Distributed generators and their connection to the system 93

Chapter 4

Fault currents and electrical protection

Black Hill Wind Farm, Duns, Scotland (28.6 MW) [RES]

4.1 Introduction

Electrical faults, caused by the breakdown of insulation, are inevitable in any

electrical power system. Faults may be caused by mechanical damage to the

equipment or created by the degradation of insulation over time. Electrical pro-

tection is then used to operate circuit breakers that isolate the faulty equipment

rapidly. Distributed generators must be protected against internal electrical faults,

with fault current flowing from the network, and conversely the distribution net-

work must be protected against fault current from the distributed generators.

Islanded operation of smaller distributed generators is not generally permitted and

so this condition is detected by protection systems designed to detect islanding or

loss of mains and the generator then disconnected. Finally, the addition of dis-

tributed generation to a distribution network may alter the flows of network fault

current in subtle ways and so lead to maloperation of conventionally designed

distribution network protection systems.

When the insulation fails in a power system and a short-circuit fault occurs,

excessive currents, perhaps as high as 20 times the load current, will flow.

These

high currents may further damage the plant in which the insulation failure has

occurred or damage the other items of equipment through which the fault current

flows. These large fault currents can lead to fires or create hazardous voltages and

are not allowed to persist for more than a second or two. Electrical protection

schemes are used to isolate rapidly the faulty piece of plant while maintaining the

sound pieces of equipment in service in order to ensure minimum disruption of

supplies to customers. The large short-circuit currents disturb the operation of the

power system, particularly by depressing the voltage, and so faults are isolated

rapidly to maintain both voltage quality and stability of the system.

Different protection schemes have been developed to protect items of plant or

sections of the distribution network. It is conventional to divide the distribution net-

work into a number of zones to ensure discrimination is obtained and only the smallest

possible section of network is isolated for any fault. Most distribution network pro-

tection is designed to respond to fault current that is supplied by the large central

generators. These are electrically remote from faults on the distribution system and so

the magnitude of fault currents is determined mainly by the impedance of the trans-

mission and distribution networks. Thus, in traditional distribution network protection

design, the fault currents are well defined and quite easy to calculate. The addition of

generation located within the distribution network leads to more complex flows of

fault current that no longer comes only from the transmission network. In addition

many of the new types of generation use either induction generators or are connected to

the network through power electronic converters, whose ability to provide fault current

is determined both by the capability of the inverter and also by their control systems.

4.2 Fault current from distributed generators

All directly connected, spinning electrical machines (motors as well as generators)

supply fault current into a short circuit in the network. Too much short-circuit

current is dangerous, as it will over-s tress circuit breakers that try to break the

current and distort cables and other plant through which it passes. Too little short-

circuit current from generators is also of concern, as most protection systems on

distribution networks work by detecting the over-currents caused by faults, and so

will not work correctly with too little fault current.

If a distribution system is designed to provide good power quality (i.e. limited

variation in voltage with changes in load), it will have a high short-circuit level

The terms ‘short circuit’ and ‘fault’ are used interchangeably in this chapter.

The short circuit or fault level is the short-circuit or fault current that would flow in the event of a short

circuit or fault multiplied by the nominal prefault voltage at that point of the network.

96 Distributed generation