Transcript Voltage Stability with Respect to Distributed Generation

1
Distribution Capacitor Placement With
Distributed Generation Concerning
Voltage Drop Reduction
Dr. Adly A. Girgis
Thomas M. Haire
Clemson University
Clemson, SC, USA
March 13, 2002
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
2
Topics
•
•
•
•
•
•
Background Information
Procedure and System
PQ Solution
PV Solution
Solution Comparison
Conclusion
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
3
Background
• Sizing and placement standard “two-thirds
rule.”
– A capacitor may be placed two-thirds the length
of the line and may be two-thirds the size of the
reactive load.
• Does not hold for economic consideration.
– This paper desires to make the voltage profile
as flat as possible.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
4
“Two-Thirds Rule”
2 n  1  i 
xi 
2n  1
I Ci
2 I S

2n  1
Where,
xi distance from substation to ith capacitor
n number of capacitors
Ici capacitor load size (Amps or VARs)
 1, to maximize peak power loss reduction
Is reactive load (Amps or VARs)
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
Procedure
• Analyze using Newton-Raphson LoadFlow
• PQ Solution
– Specify generator real and reactive power.
– Allow generator voltage to float.
– Design capacitors for constant generator power
factors of 1, 0.9, 0.8.
– Capacitors placed and sized according to “TwoThirds Rule.”
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
5
6
Procedure (cont.)
• PV Solution
– Specify generator real power and voltage.
– Allow generator reactive power to float
between 0.8 and 1.
– Design capacitors for estimated generator
power factors of 1, 0.9, 0.8.
– Capacitors placed and sized according to
“Two-Thirds Rule.”
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
7
System Studied
• 300 A circuit
SS
• (From observed conditions)
– Zone 1, 6 mi, 4.374 MW, evenly
distributed
– Zone 2, 0.75 mi, 1.458 MW, evenly
distributed
– Power factor is 0.9 lagging.
0-6.0 mi.
4.374 MW
• Wire
– 477 ACSR, 18/1 str.
• DG
– 2 Natural gas engines
– 1.062 MW each
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
6.0-6.75 mi.
1.458 MW
DG DG
7.5 mi.
8
PQ Solution
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
9
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
10
Voltage Conclusions
• The more reactive power produced by the DG,
the less voltage drop for any given number of
capacitors.
• Excluding unity power factor at all loads.
• Voltage Support from both real and reactive
power flowing from both directions.
• In practice, design generator settings and
capacitor placement for DG producing
maximum reactive power.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
11
Power Reduction Conclusions
• Real and reactive power consumed
by the wires is the least when all
loads have capacitors and DG is
operated at unity power factor.
• This formation will produce the least
current flowing in the wires.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
12
PV Solution
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
Voltage Drop Vs. Number
of Capacitors
pf=1
pf=0.9
pf=0.8
120
100
80
Voltage
60 Drop (V)
40
0
20
3
Number of
Capacitors
6
0
pf=1
pf=0.9
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
pf=0.8
13
14
Voltage Conclusions
• Best profile is a result of designing the
capacitors as if no DG was present.
• This would be a design for the DG power
factor to be 1; however, the DG will not
operate at unity power factor.
• In practice, use the “Two-Thirds Rule” as
normal and let the DG chase the power
factor of the system.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
15
Power Reduction Conclusions
• The real and reactive power loss decreases
as the number of capacitors increases.
• Most graphs show no true trend to the
change in power as a result of design
changes related to different DG power
factors.
• In designs other than unity power factor,
the load flow had difficulties finding a
solution without lowering the DG voltage.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
16
Solution Comparison
• The PV solution provided the better solution
for voltage reduction.
• This results from not forcing any source in
the system to supply a specific power.
• In these tests, the DG operated near the low
power factor setting as in the PQ solution.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
17
Conclusions
• The more reactive power produced by the
generator, the flatter the voltage profile.
• If the real and reactive power from the
generator are kept constant, design the
capacitors for max. reactive power from the
generator.
• If the voltage and real power are to be kept
constant, design the capacitors as if the DG
does not exist. Then allow DG to “chase”
the system reactive power.
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION
18
Questions?
Thank You!
Thomas M. Haire
[email protected]
(864) 656-7219
Power System 2002 Conference: Impact of Distributed Generation
CLEMSON UNIVERSITY ELECTRIC POWER RESEARCH ASSOCIATION