ECEN 667

Power System Stability

Lecture 25: Power System Stabilizers

Prof. Tom Overbye

Dept. of Electrical and Computer Engineering
Texas A&M University
[email protected]
 Announcements
• Read Chapter 9
• Homework 6 is due on Tuesday December 3
• Final is at scheduled time here (December 9
from1pm to 3pm)

2
Damping Oscillations: Power System
Stabilizers (PSSs)
• A PSS adds a signal to the excitation system to
improve the generator’s damping
– A common signal is proportional to the generator’s speed;
other inputs, such as like power, voltage or acceleration, can
be used
– The Signal is usually measured locally (e.g. from the shaft)
• Both local modes and inter-area modes can be
damped.
• Regular tuning of PSSs is important

3
 Stabilizer References
• A few references on power system stabilizers
– E. V. Larsen and D. A. Swann, "Applying Power System Stabilizers Part I:
General Concepts," in IEEE Transactions on Power Apparatus and
Systems, vol.100, no. 6, pp. 3017-3024, June 1981.
– E. V. Larsen and D. A. Swann, "Applying Power System Stabilizers Part
II: Performance Objectives and Tuning Concepts," in IEEE Transactions
on Power Apparatus and Systems, vol.100, no. 6, pp. 3025-3033, June
1981.
– E. V. Larsen and D. A. Swann, "Applying Power System Stabilizers Part
III: Practical Considerations," in IEEE Transactions on Power Apparatus
and Systems, vol.100, no. 6, pp. 3034-3046, June 1981.
– Power System Coherency and Model Reduction, Joe Chow Editor,
Springer, 2013

4
 Dynamic Models
in the Physical Structure

Mechanical System Electrical System

Stabilizer Load
Line
Exciter Relay Relay

Supply Pressure Speed Voltage Network Load

control control control Control control control
Fuel Furnace
Turbine Generator Network Loads
Source and Boiler
Machine Load
Char.
Fuel Steam Torque V, I P, Q
Governor

P. Sauer and M. Pai, Power System Dynamics and Stability, Stipes Publishing, 2006.

5
Power System Stabilizer (PSS)
Models

6
Classic Block Diagram of a System
with a PSS

PSS is here

Image Source: Kundur, Power System Stability and Control

7
 PSS Basics
• Stabilizers can be motivated by considering a classical
model supplying an infinite bus
d
    s  
dt
2 H d  E Vs
0
 TM  sin     D
0 dt X d  X ep
• Assume internal voltage has an additional component
E  Eorg
  K 
• This can add additional damping if sin(d) is positive
• In a real system there is delay, which requires
compensation 8
 PSS Focus Here
• Fully considering power system stabilizers can get
quite involved
• Here we’ll just focus on covering the basics, and
doing a simple PSS design. The goal is providing
insight and tools that can help power system
engineers understand the PSS models, determine
whether there is likely bad data, understand the basic
functionality, and do simple planning level design

9
 Example PSS
• An example single input stabilizer is shown below
(IEEEST)
– The input is usually the generator shaft speed deviation,
but it could also be the bus frequency deviation,
generator electric power or voltage magnitude
The model can be
simplified by setting
parameters to zero

VST is an
input into
the exciter

10
 Another Single Input PSS
• The PSS1A is very similar to the IEEEST
Stabilizer and STAB1

IEEE Std 421.5 describes the common stabilizers

11
 Example Dual Input PSS
• Below is an example of a dual input PSS (PSS2A)
– Combining shaft speed deviation with generator electric
power is common
– Both inputs have washout filters to remove low
frequency components of the input signals

12
 Washout Filters and Lead-Lag
Compensators
• Two common attributes of PSSs are washout filters and
lead-lag compensators
Lead-lag compensators

Washout filter

• Since PSSs are associated with damping oscillations

they should be immune to slow changes. These low
frequency changes are “washed out” by the washout
filter; this is a type of high-pass filter.

13
 Washout Filter
• The filter changes both the magnitude and angle of
the signal at low frequencies The breakpoint
frequency is when
the phase shift
is 45 degrees and
the gain is -3 dB
(1/sqrt(2))

A larger T value
shifts the breakpoint
to lower frequencies;
at T=10 the breakpoint
frequency is 0.016 Hz
Image Source: www.electronics-tutorials.ws/filter/filter_3.html 14
 Washout Parameter Variation
• The PSS2A is the most common stabilizer in both the
2015 EI and WECC cases. Plots show the variation in
TW1 for EI (left) and WECC cases (right); for both the
x-axis is the number of PSS2A stabilizers sorted by
TW1, and the y-axis is TW1 in seconds
Tw1
30 Tw1
34
28
32
26 30
24 28

22 26
24
20
22
18
20
16
Tw 1

18
Tw1

14 16
12 14

10 12
10
8
8
6
6
4
4
2 2
0 0
0 200 400 600 800 1,000 0 200 400 600 800
Tw1 Tw1

15
 Lead-Lag Compensators
• For a lead-lag compensator of the below form with
a <= 1 (equivalently a >= 1)
1  sT1 1  sT1 1  asT
 
1  sT2 1  s T1 1  sT
• There is no gain or phase
shift at low frequencies,
a gain at high frequencies but
no phase shift 1  sin  1
 , T1 
• Equations for a design maximum 1  sin  2 f 
phase shift a at a frequency f are 1
sin  
given 1 
16
 Stabilizer Design
• As noted by Larsen, the basic function of stabilizers is
to modulate the generator excitation to damp generator
oscillations in frequency range of about 0.2 to 2.5 Hz
– This requires adding a torque that is in phase with the speed
variation; this requires compensating for the gain and phase
characteristics of the generator, excitation system, and power
system (GEP(s))
– We need to compensate for the
phase lag in the GEP
• The stabilizer input is
often the shaft speed

Image Source: Figure 1 from Larsen, 1981, Part 1 17

 Stabilizer Design

• T6 is used to represent measurement delay; it is usually

zero (ignoring the delay) or a small value (< 0.02 sec)
• The washout filter removes low frequencies; T5 is
usually several seconds (with an average of say 5)
– Some guidelines say less than ten seconds to quickly remove
the low frequency component
– Some stabilizer inputs include two washout filters

Image Source: IEEE Std 421.5-2016

18
 Stabilizer Design Values

• With a washout filter value of T5 = 10 at 0.1 Hz

(s = j0.2p = j0.63) the gain is 0.987; with T5 = 1 at 0.1
Hz the gain is 0.53
• Ignoring the second order block, the values to be
tuned are the gain, Ks, and the time constants on the
two lead-lag blocks to provide phase compensation
– We’ll assume T1=T3 and T2=T4

19
Stabilizer Design Phase Compensation
• Goal is to move the eigenvalues further into the left-
half plane
• Initial direction the eigenvalues move as the stabilizer
gain is increased from zero depends on the phase at
the oscillatory frequency
– If the phase is close to zero, the real component changes
significantly but not the imaginary component
– If the phase is around -45 then both change about equally
– If the phase is close to -90 then there is little change in the
real component but a large change in the imaginary
component

20
 Stabilizer Design Tuning Criteria

• Eigenvalues moves as Ks increases

KOPT is where the damping is

maximized
KINST is the gain at which
sustained oscillations or an
instability occur

• A practical method is to find KINST, then set KOPT as

about 1/3 to ½ of this value
21
 Stabilizer Design Tuning
• Basic approach is to provide enhanced damping at
desired frequencies; the challenge is power systems can
experience many different types of oscillations, ranging
from the high frequency local modes to the slower (<
1.0 Hz usually) inter-area modes
• Usually the PSS should be set to compensate the phase
so there is little phase lag at inter-area frequencies
– This can get modified slightly if there is a need for local
stability enhancement
• An approach is to first set the phase compensation, then
tune the gain; this should be done at full output
22
 PSS2A Example Values
• Based on about 1000 WECC PSS2A models
– T1=T3 about 64% of the time and T2=T4 about 69% of the
time
– The next page has a plot of the T1 and T2 values; the
average T1/T2 ratio is about 6.4

23
 Example T1 and T2 Values
Stabilizer_PSS2A Variables

0.48
0.46
0.44
0.42
0.4
0.38
0.36
0.34
0.32
0.3
0.28
Values

0.26
0.24
0.22
0.2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02

100 200 300 400 500 600 700 800 900 1,000
T1 T2

24
 Example
• As an example we’ll use the wscc_9bus_Start case,
and apply the default dynamics contingency of a self-
clearing fault at Bus 8.
• Use Modal Analysis to determine the major modal
frequency Bus 2
Bus 7 Bus 8 Bus 9 Bus 3

and 163 MW
1.016 pu
85 MW

damping
1.025 pu 1.026 pu 1.032 pu 1.025 pu
7 Mvar -11 Mvar

Bus 5 0.996 pu 100 MW Bus 6 1.013 pu

35 Mvar

125 MW
50 Mvar

Bus 4 1.026 pu 90 MW
30 Mvar

Bus1 1.040 pu

slack
72 MW
27 Mvar

25
 Example: Getting Initial Frequency
and Damping
• The new Modal Analysis button provides quick access

Frequency is
Easy access to 1.36 Hz with
plot data 5% damping 26
 PSS Tuning Example:
We’ll Add PSS1As at Gens 2 and 3
• To increase the generator speed damping, we’ll add
PSS1A stabilizers using the local shaft speed as an input
• First step is to determine the phase difference between
the PSS output and the PSS input; this is the value we’ll
need to compensate
• This phase can be determined either
analytically, actually testing the
generator or using simulation results
– We’ll use
simulation
results

27
 PSS Tuning Example:
Using Stabilizer Reference Signals
• PowerWorld now allows reference sinusoidals to be
easily played into the stabilizer input
– This should be done at the desired modal frequency
• Modal analysis can then be used to quickly determine
the phase delay between the input and the signal we
wish to damp
• We’ll use the wscc_9Bus_Stab_Test
– This has SignalStab stabilizers modeled at each generator;
these models can play in a fixed frequency signal

28
 SignalStab Input and Results
• Enable the SignalStab stabilizer at the bus 2 generator
and run the simulation

At time=0 the stabilizer

receives a sinusoidal input
with a magnitude of 0.05
and a frequency of 1.36 Hz

29
 PSS Tuning Example:
Gen2 Reference Signal Results
• Graph shows four signals at bus 2, including the
stabilizer input and the generator’s speed
– The phase relationships are most important
Use modal
analysis to
determine the
exact phase
values for the
1.36 Hz mode;
analyze the
data between 5
and 10 seconds

30
 PSS Tuning Example:
1.36 Hz Modal Values
• The change in the generator’s speed is driven by the
stabilizer input sinusoid, so it will be lagging. The
below values show is lags by
(-161+360) – (-81.0) = 280 degrees
– Because we want to damp the speed not increased it, subtract
off 180 degrees to flip the sign. So we need 100 degrees of
compensation; with two lead-lags it is 50 degrees each

31
 PSS Tuning Example:
1.36 Hz Lead-Lag Values
In designing a lead-lag of the form
1  sT1 1  sT1

1  sT2 1  s T1
to have a specified phase shift of  at a frequency f
the value of a is
1  sin  1
 , T1 
1  sin  2 f 
In our example with  = 50 then
1  sin 
=0.132, T1  0.321, T2   T1  0.042
1  sin 

32
 PSS Tuning Example:
1.36 Hz Lead-Lag Values
• Hence T1=T3=0.321, T2=T4=0.042. We’ll assumed
T6=0, and T5=10, and A1=A2=0

• The last step is to determine Ks. This is done by

finding the value of Ks at just causes instability (i.e.,
KINST), and then setting Ks to about 1/3 of this value
– Instability is easiest to see by plotting the output (VST) value
for the stabilizer
33
 PSS Tuning Example:
Setting the Values for Gen 2
• Instability occurs with KS = 55, hence the optimal value
is about 55/3=18.3
• This increases the damping from 5% to about 16.7%
70

This is saved as case

65
60
55
50
45
wscc_9bus_Stab
40
35
30
25
20
15
10
5
0
-5
-10

0 2 4 6 8 10 12 14 16 18 20

Rotor Angle, Gen Bus 2 #1 Rotor Angle, Gen Bus 3 #1 Rotor Angle, Gen Bus1 #1
gfedcb gfedcb gfedcb

34
 PSS Tuning Example:
Setting the Values for Gen 3
• The procedure can be repeated to set the values for the
bus 3 generator, where we need a total of 68 degrees of
compensation, or 34 per lead-lag

• The values are a = 0.283, T1=0.22, T2=0.062, KS for the

verge of instability is 36, so KS optimal is 12.

35
 PSS Tuning Example:
Final Solution
70

65

60 With stabilizers
55

50
at buses 2 and 3
45

40
the damping has
35 been increased to
25.7%
30

25

20

15

10

-5

-10

0 2 4 6 8 10 12 14 16 18 20

Rotor Angle, Gen Bus 2 #1 Rotor Angle, Gen Bus 3 #1 Rotor Angle, Gen Bus1 #1
gfedcb gfedcb gfedcb

36
 Example 2: Adding a PSS to a 42
Bus System
• Goal is to try to improve damping by adding one
PSS1A at a large generator at Lion345 (bus 42)
– Example event is a three-phase fault is applied to the middle
of the 345 kV transmission line between Prairie (bus 22) and
Hawk (bus 3) with both ends opened at 0.05 seconds
42 Bus Case
26 7 MW
0 M var 2 67 MW 26 7 M W 2 67 MW

Dolphin
-5 0 M var -5 0 Mvar -50 Mv ar
2 67 MW 2 67 MW
0 M var 0 Mvar

Viking
Metric: Unserved MWh: 0.00
64%
50% 5 05 M W 5 05 MW

1 50 M W 27%
Badger
178 MW 16 2 MW 17 7 MW 30 Mvar
78%

Unserved Load: 0.00 MW 56 M var 18 5 M W 2 10 MW

33 Mvar 4 5 M var 58%
3 00 MW

The starting case

1 16 Mvar 60 Mvar
Apple
122 Mvar Steel138
1 20 Mvar 25 0 M W 1 93 M W
50 Mvar Ash138
Peach138 3 0 M var 89%
56%
46% 85%
95 M W 94 M W
7 5 MW 23 M var Rose138 19 8 MW
15 Mvar
75%
35 Mvar 28% 35 Mvar

Oak
22%

30% 31% 11 4 Mvar 11 2 M W 53%

4 0 M var 55%

name is
25%
70%
Parkway138
20 0 M W 40 0 M W 7 7 M W
Spruce138
38%
28%
Bear
53% 2 40 MW
48% 1 10 Mvar
20 0 M W 2 00 M W
45 Mvar 20 0 MW
60%
Palm 20 7 M W
58% 45 Mvar 4 5 M var 26 8 M W
12 8 M v ar
54 Mv ar 2 68 MW
20 2 MW 1 90 MW 2 00 MW 1 58 MW 1 28 Mvar

Bus42_PSS
3 2 Mvar 42 Mvar 4 0 M var
1 90 MW 20 0 M W
6 3 M var 82 Mv ar
4 3 Mvar 2 05 MW
6 5 M var
42%
25%
Sidney 86%

Valley
1 60 M W 94 MW
2 7 M var 23%
3 0 M var
Prairie 51% 25%
2 01 MW 2 01 MW
5 00 MW 5 2 M var 6 2 M var

15 8 M W 75%
65%
Hickory138 4 3 M var 99 Mvar
13 2 M W 68%
1 60 M W
63% 69% Tiger 72% 1 5 Mvar
36%
Hawk Grafton
3 2 M var Owl138 22 65 MW
57%
61% slack

Walnut138
7 2 M var 13 0 M W 57%
2 12 M W 4 5 Mvar 45%
37% 17 5 M W
3 0 M var 15 6 M W
1 40 M W 21 2 M W 32 Mvar 50%
3 3 M var 30 Mvar 2 3 M var
52% Willow138 1 76 MW 2 34 MW 23 4 M W
79%
53%
53% 12 8 MW 1 5 M var 29% 5 5 Mvar 45 M var
1 35 MW
Lion
47%
28 M var 16 50 MW

Eagle 20 Mv ar

85% Homer138 65%

Elm138 Lark138 Savoy138

16 1 M W 87%
14 0 M W
1 86 M W 21 Mv ar 40%
Lake
20 Mvar
58%
5 6 M var
8 8 MW 2 40 M W 2 40 MW
47%
-49 Mvar 0 M var 0 Mv ar
1 10 0 MW

77% 73% 1 520 MW 1 52 0 M W

58%
Monarch138
83%
Ram

37
42%
63%
 Example 2: Decide Generators to
Tune and Frequency
• Generator speeds and rotor angles are observed to
have a poorly damped oscillation around 0.6 Hz.
1.0008 60
1.0006 55
50
1.0004
45
1.0002
40
1.0000 35
0.9998 30
25
0.9996
20
0.9994 15
0.9992 10
5
0.9990
0
0.9988 0 2 4 6 8 10 12 14 16 18 20
0 2 4 6 8 10 12 14 16 18 20
Rotor Angle, Gen Dolphin345 (23) #1 Rotor Angle, Gen Dolphin345 (23) #2
Speed, Gen Dolphin345 (23) #1 Speed, Gen Dolphin345 (23) #2 g
b
c
d
e
f Rotor Angle, Gen Grafton345 (1) #1 fedcbg Rotor Angle, Gen Lake345 (2) #1
fedcbg Speed, Gen Grafton345 (1) #1 fedcbg Speed, Gen Lake345 (2) #1 gfedcb gfedcb
Rotor Angle, Gen Lake345 (2) #2 Rotor Angle, Gen Lion345 (42) #1
gfedcb Speed, Gen Lake345 (2) #2 gfedcb Speed, Gen Lion345 (42) #1 gfedcb Rotor Angle, Gen Oak345 (18) #1 gfedcb Rotor Angle, Gen Oak345 (18) #2
gfedcb Speed, Gen Oak345 (18) #1 gfedcb Speed, Gen Oak345 (18) #2
gfedcb gfedcb gfedcb Rotor Angle, Gen Oak345 (18) #3 gfedcb Rotor Angle, Gen Prairie345 (22) #1
Speed, Gen Oak345 (18) #3 Speed, Gen Prairie345 (22) #1
gfedcb Speed, Gen Ram345 (35) #1 gfedcb Speed, Gen Viking345 (19) #1
gfedcb Rotor Angle, Gen Ram345 (35) #1 gfedcb Rotor Angle, Gen Viking345 (19) #1
gfedcb Speed, Gen Viking345 (19) #2 gfedcb Speed, Gen Viking345 (19) #3 gfedcb Rotor Angle, Gen Viking345 (19) #2 gfedcb Rotor Angle, Gen Viking345 (19) #3
gfedcb gfedcb gfedcb gfedcb

38
Aside: Visualizing the Disturbance in
PowerWorld Dynamics Studio (DS)

39
Example 2: Response Quantified
Using Modal Analysis

For 0.6 Hz mode the damping

is 2.89%
40
 Example 2: Determine Phase
Compensation
• Using a SignalStabStabilizer at bus 42 (Lion345), the
phase lag of the generator’s speed, relative to the
stabilizer input is 199 degrees; flipping the sign requires
phase compensation of 19 degrees or 9.5 per lead-lag
• Values are a = 0.72; for 0.6 Hz, T1= 0.313, T2=0.225;
set T3 and T4 to match; gain at instability is about 450, so
the gain is set to 150.

The case with the test signal is Bus42_PSS_Test

Adding this single stabilizer increases the damping to 4.24%
41
 Example 2: Determine Phase
Compensation For Several Generators
• Adding and tuning three more stabilizers (at
Grafton345 and the two units at Lake345) increases
the damping to 8.16%
60
55
However, these
50
45
changes are
40
35
impacting
30
25
modes in other
20
15
areas of the system
10
5
0
0 2 4 6 8 10 12 14 16 18 20

Rotor Angle_Gen Dolphin345 (23) #1 Rotor Angle_Gen Dolphin345 (23) #2

g
b
c
d
e
f Rotor Angle_Gen Grafton345 (1) #1 fedcbg Rotor Angle_Gen Lake345 (2) #1
gfedcb Rotor Angle_Gen Lake345 (2) #2 gfedcb Rotor Angle_Gen Lion345 (42) #1
gfedcb Rotor Angle_Gen Oak345 (18) #1 gfedcb Rotor Angle_Gen Oak345 (18) #2
gfedcb Rotor Angle_Gen Oak345 (18) #3 gfedcb Rotor Angle_Gen Prairie345 (22) #1
gfedcb Rotor Angle_Gen Ram345 (35) #1 gfedcb Rotor Angle_Gen Viking345 (19) #1
gfedcb Rotor Angle_Gen Viking345 (19) #2 gfedcb Rotor Angle_Gen Viking345 (19) #3
gfedcb gfedcb
42