Wide-Area Synchrophasor based Transient

Instability Prediction and Control

Dinesh Rangana Gurusinghe

Neethu Raju
Athula D. Rajapakse

NASPI 2018 - Albuquerque, NM, USA May 3, 2018

Outline

• Introduction

• Transient Stability Status Prediction Algorithm

• Transient Instability Mitigation Strategy

• Laboratory-Scale Hardware Setup

• Real-Time Simulation Results

• Conclusion

2
Outline
Introduction Outline
Introduction
• Transient instability is a fast phenomenon and a generator
or group of generators potentially lose the synchronism
within a few seconds after a severe disturbance.
• Fast recognition of such instabilities provides opportunity
to initiate appropriate emergency control actions.
• In literature, it is common practice to provide controls
referred as special protection systems (SPSs), which are
rather complicated and expensive.
• Response based wide area protection and control systems
are the best option as they are more simple, effective and
can be implemented using wide area synchrophasor
measurements of a network.

Outline
3 Introduction
Overall Structure of the Proposed Method
Introduction
Objective

Developing a response based novel technique to

predict and control impending transient
instability conditions following a severe
disturbance using wide area synchrophasors.

Outline
4 Introduction
Overall Structure of the Proposed Method
Introduction
Overall Structure of the Proposed Method
PMU

s
PMU
Voltage magnitude

s
Wide area
Severe yes Stability stable
Synchrophasor
disturb.? status?
PMU PMU
Measurements
s

s
no unstable

Frequency
Transient First unstable
stability status generator
PMU prediction (Gen X)
s

algorithm

Identification of coherent groups

of generators
PMU

Generator tripping
s

PMU
(Stage 1) Gen X
s

PMU
yes belongs to
coherent
Under Activate load group ?
yes
s

frequency shedding
? scheme no
PMU
s

Load shedding Over

Generator tripping yes
PMU PMU

(Stage 2) frequency
?
s

Power System Transient instability mitigation scheme

Introduction
5 Overall Structure of the Proposed Method
Transient Stability Status Prediction Algorithm
Overall Structure of the Proposed Method
Transient Stability Status Prediction Algorithm
• This algorithm classifies the transient swings as stable or
unstable, based on rate of change of voltage (ROCOV) vs.
voltage deviation (ΔV) characteristics of the post-
disturbance voltage magnitudes.
• Implementation involves three-steps:
1. Identification of contingencies
that makes generator
A
marginally unstable through

ROCOV (pu/s)
(0,0)
off-line dynamic simulations.
2. Determination of stability B (x1,y1) Stable
Marginally
stable trajectory
boundary for each generator. region
Stability boundary
3. Detection of severe D
Unstable region C (x2,y2)
disturbances and triggering
Voltage deviation, ΔV (pu)
the transient stability status
prediction algorithm. Stability boundary on ROCOV-ΔV plane
Overall Structure of the Proposed Method
6 Transient Stability Status Prediction Algorithm
Transient Instability Mitigation Scheme
Transient Stability Status Prediction Algorithm
Transient Instability Mitigation Scheme
• The mitigation scheme identifies coherent groups of
generators from post-disturbance voltage magnitudes.
• The generators which is becoming unstable first is tripped
together with all other generators in the same coherent
group.
• A priority based load shedding scheme is activated to trip
loads to retain generation-load equilibrium.
• Frequencies at the generator terminals are monitored and
the generators violating under/over frequency limits are
tripped.

Overall Structure of the Proposed Method

7 Transient Instability Mitigation Scheme
Laboratory-Scale Hardware Setup
Transient Instability Mitigation Scheme
Laboratory-Scale Hardware Setup

GPS Real Time Digital Simulator (RTDS)

Antenna PMU

s
(235-0113)
PMU PMU

s
PMU PMU PMU

s
IRIG - B

GPS Clock
GTSYNC
(SEL 2407) Power System Model

GTNET - PMU GTNET - SKT

IEEE
C37.118.2 SocKeT
Protocol

Synchrophasor Communication Network

IEEE
C37.118.2

SocKeT
Protocol

Proposed Transient Instability

Prediction and Control System
Phasor Data Concentrator (PDC)
(OpenPDC V2.0)

Time-aligned Measurements

Transient Instability Mitigation Scheme

8 Laboratory-Scale Hardware Setup
Real Time Simulation Results
Laboratory-Scale Hardware Setup
Simulation Results : IEEE 39-Bus Test System

Laboratory-Scale Hardware Setup

9 Real Time Simulation Results
Conclusion
Real Time Simulation Results
Real Time Simulation Results
240
200 Gen.-30 Gen.-31 Gen.-32 Gen.-33
Rotor Angle (deg)

160 Gen.-34 Gen.-35 Gen.-36 Gen.-37

120 Gen.-38 Gen.-39
80
40
0
-40
-80
1.2
Voltage magnitude (pu)

1.0
0.8
0.6
0.4
0.2
0.0

UTC Time
Variations of rotor angle and voltage magnitude
Fault on line 16-17 (95% of the length) cleared by removing the line after 6 cycles
Laboratory-Scale Hardware Setup
10 Real Time Simulation Results
Conclusion
Real Time Simulation Results
Real Time Simulation Results
Gen.-30 Gen.-31 Gen.-32
2 2 2
ROCOV (pu/s)

ROCOV (pu/s)

ROCOV (pu/s)
1 1 1

0 0 0

-1 -1 -1
-0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5
∆ V (pu) ∆ V (pu) ∆ V (pu)

Gen.-33 Gen.-34 Gen.-35

2 2 2
ROCOV (pu/s)

ROCOV (pu/s)

ROCOV (pu/s)
1 1 1
0 0 0
-1 -1 -1
-0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5
∆ V (pu) ∆ V (pu) ∆ V (pu)

Gen.-36 Gen.-37 Gen.-38

2 2 2
ROCOV (pu/s)

ROCOV (pu/s)

ROCOV (pu/s)
1 1 1
0 0 0

-1 -1 -1
-0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5 -0.5 -0.25 0 0.25 0.5
∆ V (pu) ∆ V (pu) ∆ V (pu)

Variations of ROCOV vs. voltage deviation

Fault on line 16-17 (95% of the length) cleared by removing the line after 6 cycles
Laboratory-Scale Hardware Setup
11 Real Time Simulation Results
Conclusion
Real Time Simulation Results
 Real Time Simulation Results
240
200 Gen.-30 Gen.-31 Gen.-32 Gen.-33 Gen.-34
Rotor angle (deg)

160 Gen.-35 Gen.-36 Gen.-37 Gen.-38 Gen.-39

120
80
40
0
-40
-80
1.4
1.2
Voltage magnitude

1.0
0.8
(pu)

0.6
0.4
0.2
0.0
Generator trip signals
(at LDC)

UTC
Laboratory-Scale Hardware Setup
12 Real Time Simulation Results
Conclusion
Real Time Simulation Results
 Real Time Simulation Results
62.0
61.6 Gen.-30 Gen.-31 Gen.-32
W/o load shedding

61.2
Gen.-33 Gen.-34 Gen.-35
Frequency (Hz)

60.8
60.4 Gen.-36 Gen.-37 Gen.-38
60.0
59.6 Gen.-39
59.2
58.8
58.4
58.0
62.0
61.6
With load shedding

61.2
Frequency (Hz)

60.8
60.4
60.0
59.6
59.2
58.8
58.4
58.0
Load trip signals
(at LDC)

UTC
Laboratory-Scale Hardware Setup
13 Real Time Simulation Results
Conclusion
Real Time Simulation Results
Conclusion
• Based on the results of numerous experiments, it can be
concluded that using the post-disturbance trajectories of
generator terminal buses on ROCOV-∆V plane, stability
status of the system can be determined.
• The proposed prediction algorithm pinpoints the unstable
generator, which is very important in determining
emergency control actions.
• A generator tripping and load shedding scheme
implemented in the RTDS and a laboratory scale
communication network demonstrated the effectiveness
of the proposed approach.

14
Real Time Simulation Results
Conclusion Conclusion
Thank you