Wide-area Measurement Data Analytics Using
FNET/GridEye: A Review
Jidong Chai1, Yong Liu1, Jiahui Guo1, Ling Wu1, Dao
Zhou1, Wenxuan Yao1, Yilu Liu1,2
1
The University of Tennessee, Knoxville
Knoxville, TN USA
Abstract—With the increasing size and complexity of the power
grid, wide-area measurement system (WAMS) has been studied
and implemented more widely. As a pioneering WAMS
deployed at the distribution level, the frequency monitoring
network FNET/GridEye has been providing independent
observation of dynamic performance of U.S. and other
worldwide power grids continuously since 2004. By utilizing
GPS-time-synchronized monitors called frequency disturbance
recorders (FDRs), FNET/GridEye is capable to capture dynamic
grid behaviors (e.g., frequency and voltage phase angle).
Compared to the three-phase phasor measurement units
(PMUs), the FDRs can be conveniently deployed across the grid
at distribution level without complex procedure and heavy
outlay. With years’ continuous development and systematic
management, FNET/GridEye has thrived with multiple
advanced online and offline applications. The system has been
proved to be highly effective in enhancing power grid operators’
situational awareness capabilities. In this paper, the latest
accomplishments of FNET/GridEye will be introduced.
Index Terms-- Data analytics, FNET/GridEye, Frequency
disturbance recorder (FDR), Phasor measurement, Wide-area
measurement system (WAMS).
I.
Thomas Jr. King1, 2, Jose R. Gracia2,
Mahendra Patel3
2
3
Oak Ridge National Laboratory, Oak Ridge, TN USA
Electric Power Research Institute, Palo Alto, CA USA
FNET/GridEye measures from normal single-phase electrical
outlets with a simple procedure at a lower outlay. As a
complete wide-area monitoring system, all the phasor
measurements collected by frequency disturbance recorders
(FDRs) are transmitted to the FNET/GridEye server hosted at
the University of Tennessee, Knoxville (UTK), and Oak Ridge
National Laboratory (ORNL) for cutting-edge research and
development (R&D).
Figure 1. FDR deployment in North America
INTRODUCTION
Due to the increasingly complex behavior exhibited by
large-scale power systems, wide-area measurement system
(WAMS) has been utilized to complement the traditional
supervisory control and data acquisition (SCADA) system to
improve operators’ situational awareness [1]-[3]. By providing
GPS-time-synchronized measurements of grid status at high
time-resolution, it is able to reveal power system dynamics
which cannot be captured before. Especially with more
uncertain renewables introduced to the grid, WAMS has
become an essential tool to deal with current and future power
grid challenges.
As a pioneering WAMS deployed at the distribution level,
the frequency monitoring network FNET/GridEye has been
continuously monitoring the grids for over ten years and
various data visualization and analytics applications have been
developed [4]-[39]. Unlike phasor measurement units (PMU)
which require high manufacturing and installation costs,
This work made use of the Engineering Research Center Shared
Facilities supported by the Engineering Research Center Program of the
National Science Foundation and DOE under NSF Award Number EEC1041877 and the CURENT Industry Partnership Program.
Figure 2. FDR worldwide deployment
After over ten years of development and a number of
improvements, the FNET/GridEye system is widely welcomed
by the academia, industry as well as governments and has
proved to be very stable and reliable. As of 2016, more than
200 FDR units have been deployed across the four North
American Interconnections: Eastern Interconnections (EI),
Western Electricity Coordinating Council system (WECC),
Electric Reliability Council of Texas system (ERCOT), and
Hydro Quebec area. Over 50 FDRs have been deployed
worldwide, e.g., Europe, China, and Egypt. Fig. 1 shows the
current distribution of FDRs across the North American power
grid and Fig. 2 shows the world-wide FDR deployment map.
Both global and local characteristics of frequency and phase
angle variation can be monitored and analyzed based on these
large volumes of data.
In this paper, the latest R&D of FNET/GridEye will be
presented. The rest of the paper is organized as follows.
Section II introduces FNET/GridEye system architecture.
Online applications are described in Section III. Section IV
highlights some of offline data analytics applications and
Section V concludes the paper.
II.
FNET/GRIDEYE SYSTEM ARCHITECTURE
FNET/GridEye system consists of two major parts: sensors
that are deployed across the power grids and data servers
hosted by UTK and ORNL as shown in Fig. 3. The sensor
(FDR) is an embedded microprocessor system with GPS time
synchronization and Ethernet communications capability [4][8]. So far, three generations of FDRs have been developed to
consistently pursue for higher measurement accuracy and
better data quality. The current most-deployed FDR is
Generation-II. Some of the new features of Generation-III
include: 1) added power quality analysis function which can
estimate harmonics composition and detect voltage sag and
swell [9]; 2) higher steady-state phase angle and frequency
measurement accuracy. The error is less than 0.005 ̊ and
0.00006 Hz, respectively, compared with 0.01 ̊ and 0.0005 Hz
for Generation-II; 3) improved dynamic-state measurement
accuracy [10],[11]; 4) use of atomic clock as the GPS timing
backup [12]. FDRs' holdover capability can be up to a day
without losing accuracy. Since FNET/GridEye is installed at
the distribution level, unlike PMU, active power and reactive
power quantities are not included in the system.
Figure 3. FNET/GridEye system architecture
The other part of FNET/GridEye system is the data center,
where the measurements provided by FDRs are systematically
managed, technically processed, and safely archived [13]. The
data center is a multi-layer data management system which is
composed of the data server, application server, web server,
and backup server, etc. Since power system applications
require various time requirements, the applications can be
roughly divided into online applications and offline
applications. In the following two sections, various online
(e.g., data visualization, disturbance and oscillation detection)
and offline (e.g., measurement-driven dynamic modeling and
validation, statistical data analysis) applications run on the
data center will be introduced respectively. Note that the data
are currently transmitted over Internet and thus they cannot be
used for real time control.
III.
ONLINE APPLICATIONS
Real-time applications require response within seconds or
even sub-seconds after receiving the measurement data, while
non-real-time applications have more flexible timing
requirements or are upon request [8]. In this section, some of
the important real-time applications are presented.
Data visualization is important to transform large amounts
of data to enhance grid operators’ situational awareness.
FNET/GridEye correlates streaming wide-area frequency and
voltage angle measurements with corresponding FDR
geographical location information to provide a full coverage
of the current grid status. The real-time frequency and angle
contour maps of the North American power grid can be
accessed online through the FNET/GridEye web services.
Fast detection and location of large disturbances such as
generation trip and transmission line trip are critical for
reliable and safe operation of the power grid. As is well
known, power system frequency is determined by system
generation and load consumption. Through continuously
calculating the rate of average frequency change, disturbances
such as generation trip, load shedding, and line trip can be
detected. Then a geometrical tri-angulation algorithm making
use of the time difference of arrival (TDOA) will be employed
to locate the disturbance [14], [15]. The event information will
be included in a report and automatically sent out to service
subscribers, such as utility operators, in seconds.
Small-signal stability is another key concern for power
system operators. Inter-area oscillations could be monitored
by FNET/GridEye and corresponding modal analysis is
performed using multi-channel matrix pencil algorithm [16].
Oscillation information such as FDRs with largest oscillation
magnitude, oscillation frequency and damping ratio will be
recorded and sent out to service subscribers in real time.
Besides using the ring-down data, FNET/GridEye is able to
utilize ambient data to estimate oscillation mode [17]-[19]. It
has the capability to process hundreds of streaming FDR
measurements at the same time through a multi-channel
parallel processing design.
Islanding detection is also implemented [20], [21].
Islanding is the situation in which a part of the grid becomes
electrically isolated from the remainder of the power system.
With more and more distributed generation in the power grid,
it is becoming imperative to have islanding detection and
protection mechanisms.
IV.
OFFLINE DATA ANALYTICS
Power system applications require various time scales.
Besides the online applications presented above, this section
introduces some of offline applications.
A. Worldwide Frequency Distribution Analysis
The power system frequency is an important indicator of a
power grid’s health and stability. The distribution pattern of
frequency measurement data could indicate the overall
performance, especially frequency control, of an
interconnection over time. A frequency distribution
probability calculation method is applied to frequency
measurement data from 2005-2013 collected by the
FNET/GridEye system. The study investigated the distribution
probability of frequency measurements over North American
and worldwide power grids, and compares corresponding
distribution patterns [22].
Figure 4. Frequency distribution probability of different interconnections
(spring 2013)
Fig. 4 shows the comparison of the frequency distribution
probabilities of six interconnections during the spring of 2013.
Since the nominal frequency of North China power grid and
UCTE grid (Continental Europe power grid) is 50Hz, the
deviation from nominal frequency is used for the x axis of the
probability distribution function (PDF). It can be seen that
except ERCOT and North China, the PDF curves of all the
other interconnections are standard normal distributions. For
the spring of 2013, Quebec has the smallest standard
deviation, while EI is the second smallest. The PDF curves of
WECC and EUROPE almost overlap with relatively larger
standard deviation. It is obvious that the PDF curve of
ERCOT presents a roughly bimodal distribution. Unlike other
interconnections, all the frequency data in North China grid
fall in the range between 49.9560Hz and 50.0440Hz.
WECC, and Quebec grid. Very slight distribution differences
over different days of a week can be noticed in ERCOT.
B. Post-event Analysis
One of main synchrophasor applications is that it could be
used for the post-event analysis. With the wide-area
synchronized and high time resolution data, concerned events
such as blackouts, large generation trips and oscillations
could be replayed and analyzed. FNET/GridEye has the event
visualization tool to study the electromechanical wave
propagation throughout the power system [23].
(a)
(b)
(c)
(d)
Figure 6. San Diego blackout replay using FNET/GridEye frequency
measurement
Fig. 6 shows the San Diego blackout happened in
September 2011. It occurred when a 500-kV line connecting
Arizona with San Diego tripped following a capacitor
switchout. Approximately 1.4 million people were affected.
This visualization tool aided post-event analysis is beneficial
for the grid operators to have intuitive understanding of the
cascading response.
C. Measurement-based Dynamic Model Development
Traditionally circuit-based method is the only available
tool for the dynamic study of large-scale power grid.
However, the limitations in model accuracy and calculation
speed can be serious bottlenecks for large interconnected
systems. With the increasing number of real-time wide-area
measurements, measurement-based approaches are becoming
more attractive.
Figure 5. Frequency distribution probability of different days of a week for
different interconnections
Fig. 5 compares the distribution patterns of frequency data
between different days of a week in 2012. Unlike some very
clear changes observed in the seasonal comparisons, the PDF
curves of different days of a week overlap fairly well in EI,
Based on over ten years’ observation of the North
American grids and major grids worldwide and experience of
extensive interconnection-level dynamic simulations, it was
found out that the linearity of large-scale power systems was
underestimated previously. Since only a limited number of
transfer functions of the power system are critical for a certain
dynamic study, a reduced-order equivalent dynamic model
could be derived based on some measured input/output data
with system identification technologies in real time [24]. This
dynamic model could be used to estimate [25]-[28], predict
[29],[30] and control [31] a large-scale interconnected power
grid’s small-signal dynamic behavior.
D. Measurement-aided Model Validation
The accuracy of a power system model is essential to safe
power system operation and economic planning. However,
recent studies revealed that interconnection-wide dynamic
models fail to capture key characteristics of real systems. The
inconsistency between simulation and reality can lead to
incorrect engineering judgement that eventually causes poor
system dynamic performance, over or under utilization of
resources, and even system instability.
To improve the model quality, FNET/GridEye frequency
measurements have been utilized to validate the dynamic
model of U.S. power grids, such as the EI system. Various
parameters such as governor settings, machine inertias, loads
are tuned to match the simulation results with the
corresponding FNET/GridEye measurements [23]. Fig. 7
shows the improvement of EI frequency response simulation
results by incorporating the previously un-modeled governor
deadbands in the model [32].
load consumption. A large frequency deviation could damage
equipment gradually, degrade load performance, overload
transmission lines and may lead to collapse of the power
system. Thus, how the frequency changes and its statistic
characteristics become an interest, with the large dataset
collected by FNET/GridEye system since 2004, it is possible
to analyze the frequency signature and trend statistically.
Calculation of the yearly average frequency extrema and
corresponding standard deviation may help indicate the
overall performance of an interconnection over time. A
smaller absolute deviation between the extrema and the
nominal frequency value implies better frequency control [34].
Fig. 8 shows that the average minimums in both EI and
WECC drop down to almost the same value in 2010 and stay
nearly identical from 2010 to 2013. Similar trend happens to
the average maximum values: after fluctuating from 2008 to
2010, the values converge from 2010 to 2013. Fig. 9 shows
the standard deviations of the extrema over time. It is obvious
that the standard deviation in EI is smaller than WECC. In
addition, the standard deviations of both interconnections
appear to be tracking each other.
Figure 7. FNET/GridEye-aided model validation
E. System Inertia Status Estimation
The system inertia plays a critical role in determining the
inertial frequency decline following disturbances. Low system
inertia allows fast frequency decline which could lead to
frequency nadir out of protection setting, even further result in
cascading outage. Perception of system inertia condition is
significant to operators to deploy energy, and maintain the
system reliability. However, the true system inertia is hard to
get.
With adequate observation of system behaviors, the
application of system inertia status estimation is developed.
The method utilizes the rate of frequency change during the
inertial response following disturbance to evaluate the system
inertia condition [33]. According to a statistical analysis on the
transients of over 300 events during January 2013 to July 2015
in the EI, the mean rate of frequency change is 0.004Hz/s,
with standard deviation of 0.0016. The statistical analysis also
indicated that the probability of frequency change rate beyond
0.0074Hz/s is as low as 2.5%. Future events with higher
change rate than this threshold should catch more attentions
from system operators since it happens rarely and the system
could face security risk.
F. Frequency Extreme Statistics
The nearly constant frequency could ensure the constancy
of speed of induction and synchronous motors. Meanwhile,
the frequency of power system is a reflection of the
imbalance of the active power between the generation and
(a) Average minimums by year (b) Average maximums by year
Figure 8. Average yearly values of frequency extrema within EI and
WECC, 2008-2013
(a) Minimums of standard deviation by year (b) Maximums of standard
deviation by year
Figure 9. Standard deviations of extrema for EI and WECC, 2008-2013
G. Wide-area Power System Measurements for Digital
Authentication
As an example of innovative interdisciplinary
FNET/GridEye-based applications, the wide-area power
system measurements could be used for authentication of
digital recordings [35]. When a digital recording is made
either by recording devices which are directly connected to the
electric network or by battery-powered devices, the power
grid signals will leave traces in the digital recording [36]. Both
power system frequency and phase angle can be extracted
from the digital recording using digital signal processing
technique and then they are compared with FNET/GridEye
database as shown in Fig. 10. The recording starting time
could be determined by frequency match between extracted
values and frequency database using mean square error or
correlation coefficient method. Tampering detection could be
performed based on frequency and phase angle comparisons
[37].
1000
Extracted phase
FDR measurement
[1]
[2]
[3]
Phase(degree)
0
[4]
-1000
-2000
[5]
-3000
0
500
1000
Time(s)
1500
2000
Figure 10. Comparison of frequency and phase angle from digital recording
and FNET/GridEye measurement
H. Other Applications
Some other applications are highlighted here. Based on
FNET/GridEye measurement data, the impact of social events
like the FIFA World Cup or the NFL SuperBowl on the
power grids can be analyzed [38]. Data mining and statistical
techniques are used to perform the analysis of FNET/GridEye
event database. Power grid disturbance analysis studies the
relationship between frequency change and the power loss of
disturbances [39]. Statistical oscillation analysis is exploring
the relationship among oscillation magnitude, event type,
amount, and location.
V.
CONCLUSIONS
As a pilot WAMS, FNET/GridEye was developed at the
distribution level and specifically applied to power system
dynamics monitoring. FNET/GridEye is a GPS-synchronized,
highly accurate, high time resolution, low-cost and widelydeployed system. After over ten years’ development, it proves
to be very stable and reliable. With its easy installation and
wide deployment, it has been being a situational awareness
enhancement tool for electric utilities, independent system
operators, and regulatory agencies. Due to the installation
location and data communication limitations, the system does
not include active and reactive power information and cannot
be used for real time control. In this paper, the latest
developments of FNET/GridEye are presented from a data
analytics viewpoint. A variety of online and offline
applications are described.
FNET/GridEye will continue to develop and explore its
applications in power system dynamic monitoring, modelling
and other areas. With more sensors deployed, besides
applications targeting for the wide-area transmission level,
studies in the distribution level will be conducted.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
ACKNOWLEDGMENT
The authors would like to express gratitude to all the past
FNET/GridEye group members for their pioneering work that
contribute
to
the
successful
implementation
of
FNET/GridEye. Also, the authors would like to thank all the
past and current FNET/GridEye sponsors and hosts.
REFERENCES
[19]
[20]
A. G. Phadke, J. S. Thorp, and M. G. Adamiak, "A new measurement
technique for tracking voltage phasors, local system frequency, and rate
of change of frequency," IEEE Trans. Power App. Syst., vol. PAS-102,
no. 5, pp. 1025-1038, May 1983.
A. G. Phadke and R. M. de Moraes, "The wide world of wide-area
measurement," IEEE Power Energy Mag., vol. 6, no. 5, pp. 52-65,
2008.
D. Karlsson, M. Hemmingsson, and S. Lindahl, "Wide area system
monitoring and control-Terminology, phenomena, and solution
implementation strategies," IEEE Power Energy Mag., vol. 2, no. 5, pp.
68-76, 2004.
Z. Zhong, C. Xu, B. J. Billian, L. Zhang, S.-J.S. Tsai, R. W. Conners,
V. A. Centeno, A. G. Phadke, and Y. Liu, "Power system frequency
monitoring network (FNET) implementation," IEEE Trans. Power
Syst., vol. 20, no. 4, pp. 1914–1921, Nov. 2005.
Y. Zhang, P. N. Markham, T. Xia, L. Chen, Y. Ye, Z. Wu, Z. Yuan, L.
Wang, J. Bank, J. Burgett, R.W. Conners, and Y. Liu, "Wide-area
frequency monitoring network (FNET) architecture and applications,"
IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 159-167, Sept. 2010.
Z. Lin, Y. Liu, Y. Jia, Y. Zhang, T. Xia, Y. Liu, and F. Wen, "Dynamic
performance test of single-phase phasor measurement units," in Proc.
2011 IEEE Power Eng. Soc. Gen. Meet., pp. 1-7.
Y. Zhang, Y. Liu, L. Chen, J. Guo, and Y. Liu, "Visualization of
distribution level voltage magnitude pattern trend in EI system using
FNET data," in Proc. 2014 IEEE T&D Conference and Exposition, pp.
1-5.
Y. Liu, et al. "Wide-area measurement system development at the
distribution level: an FNET/GridEye example," IEEE Trans. Power
Del., vol. PP, no. 99, pp.1-1, 2015.
L. Zhan, J. Zhao, S. Gao, J. Cullis, Y. Liu, and Y. Liu, "Universal grid
analyzer design and development," in Proc. 2015 IEEE Power Eng.
Soc. Gen. Meet, pp. 1-5.
L. Zhan, Y. Liu, J. Culliss, J. Zhao, and Y. Liu, "Dynamic single-phase
synchronized phase and frequency estimation at the distribution level,"
IEEE Trans. Smart Grid, vol. 6, no. 4, pp. 2013-2022, July 2015.
L. Zhan, Y. Liu, Y. Liu, "A Clarke Transformation-based DFT phasor
and frequency algorithm for wide frequency range," IEEE Trans. Smart
Grid, in press.
L. Zhan, Y. Liu, W. Yao, J. Zhao, and Y. Liu, “Utilization of chip scale
atomic clock for synchrophasor measurements,” IEEE Trans. Power
Del. vol. PP, no. 99, 2016.
D. Zhou, J. Guo, Y. Zhang, J. Chai, H. Liu, Y. Liu, C. Huang, X. Gui,
and Y. Liu, “Distributed data analytics platform for wide-area
synchrophasor measurement systems,” IEEE Trans. Smart Grid, in
press.
R. M. Gardner, J. K. Wang, and Y. Liu, "Power system event location
analysis using wide-area measurements," in Proc. 2006 Power Eng.
Soc. Gen. Meet., pp. 1-7.
R. M. Gardner, J. N. Bank, J. K. Wang, A. J. Arana, and Y. Liu, "Nonparametric power system event location using wide-area
measurements," in Proc. 2006 IEEE Power Syst. Conf. Expo., pp.16681675.
Z. Yuan, T. Xia, Y. Zhang, L. Chen, P. N. Markham, R. M. Gardner,
and Y. Liu, "Inter-area oscillation analysis using wide area voltage
angle measurements from FNET," in Proc. 2008 IEEE Power Eng. Soc.
Gen. Meet., pp.1-7.
J. Guo, H. Liu, D. Zhou, J. Chai, Y. Zhang, and Y. Liu, "Real-time
power system electromechanical mode estimation implementation and
visualization utilizing synchrophasor data", in Proc. 2016 IEEE T&D
Conference and Exposition, In Press.
S. You, J. Guo, G. Kou, Y. Liu, and Y. Liu, “Oscillation mode
identification based on wide-area ambient measurements using
multivariate empirical mode decomposition”, Electric Power Systems
Research, vol. 134, pp. 158-166, 2016.
S. You, J. Guo, W. Yao, S. Wang, Y. Liu, and Y. Liu, “Ring-down
oscillation mode identification using multivariate empirical mode
decomposition”, in Proc. 2016 IEEE Power and Energy Society
General Meeting, pp. 1-5.
Z. Lin, T. Xia, Y. Ye, Y. Zhang, L. Chen, Y. Liu, K. Tomsovic, T.
Bilke, and F. Wen, "Application of wide area measurement systems to
islanding detection of bulk power systems," IEEE Trans. Power Syst.,
vol. 28, no. 2, pp. 2006-2015, May 2013.
[21] J. Guo, Y. Zhang, M. A. Young, M. J. Till, A. Dimitrovski, Y. Liu, P.
Williging, and Y. Liu, "Design and implementation of a real-time offgrid operation detection tool from a wide-area measurements
perspective," IEEE Trans. Smart Grid, vol. 6, no. 4, pp. 2080-2087,
July 2015.
[22] Y. Zhang, "Frequency Monitoring Network (FNET) Data Center
Development and Data Analysis," Ph.D. dissertation, Dept. of EECS,
Univ. of Tenn., Knoxville, TN, 2014.
[23] L. Chen, "Wide-area measurement application and power system
dynamics," Ph.D. dissertation, Dept. of EECS, Univ. of Tenn.,
Knoxville, TN, 2011.
[24] Y. Liu, "Wide-area measurement and applications in power system
dynamic modeling and control." Ph.D. dissertation, Dept. of EECS,
Univ. of Tenn., Knoxville, TN, 2013.
[25] Y. Liu, K. Sun, and Y. Liu, "Measurement-based power system
dynamic model for response estimation," in Proc. 2012 IEEE Power
and Energy Society General Meeting, pp.1-6.
[26] F. Bai, Y. Liu, Y. Liu, K. Sun, X. Wang, N. Bhatt, A. Del Rosso, and
E. Farantatos, "Methods to establish input-output relationship for
system identification-based models," in Proc. 2014 IEEE Power and
Energy Society General Meeting, pp.1-5.
[27] Y. Liu, K. Sun, and Y. Liu, "A Measurement-based power system
model for dynamic response estimation and instability warning,"
Electric Power Systems Research, vol. 124, pp. 1-9, July 2015.
[28] F. Bai, Y. Liu, Y. Liu, K. Sun, N. Bhatt, A. D. Rosso, E. Farantatos,
and X. Wang, "Measurement-based correlation approach for power
system dynamic response estimation," IET Generation, Transmission &
Distribution, vol. 9, no.12, pp.1474-1484, Aug. 2015.
[29] C. Li, J. Chai, Y. Liu, N. Bhatt, A. D. Rosso, and E. Farantatos, "Power
system dynamics prediction with measurement-based autoregressive
model," in Proc. 2014 CIGRE US Conference, pp. 1-5.
[30] C. Li, Y. Liu, K. Sun, Y. Liu, N. Bhatt, "Measurement based power
system dynamics prediction with multivariate AutoRegressive Model,"
in Proc. 2014 IEEE Power and Energy Society T&D Conference and
Exposition, pp.1-5.
[31] H. Liu, L. Zhu, Z. Pan, F. Bai, Y. Liu, Y. Liu, M. Patel, E. Farantatos,
and N. Bhatt, "ARMAX-Based Transfer Function Model Identification
Using Wide-Area Measurement for Adaptive and Coordinated
Damping Control," IEEE Trans. Smart Grid, in press.
[32] G. Kou, P. N. Markham, S. Hadley, T. J. King, and Y. Liu, "Impact of
Governor Deadband on Frequency Response of the U.S. Eastern
Interconnection," IEEE Trans. on Smart Grid, In press.
[33] L. Wu, Y. Liu, D. Zhou, J. Guo, and Y. Liu, “Observation of Inertial
Frequency Response of Main Power Grids Worldwide Using
FNET/GridEye”, in Proc. 2016 IEEE Power and Energy Society
General Meeting, pp. 1-5.
[34] P. N. Markham, Y. Liu, T. Bilke and D. Bertagnolli, "Analysis of
frequency extrema in the eastern and western interconnections," in
Proc. 2011 IEEE Power and Energy Society General Meeting, pp. 1-8,
July 2011.
[35] Y. Liu, Z. Yuan, P. N. Markham, R. W. Conners, and Y. Liu,
"Application of power system frequency for digital audio
authentication," IEEE Trans. Power Del., vol. 27, no. 4, pp. 1820-1828,
Oct. 2012.
[36] J. Chai, F. Liu, Z. Yuan, R. W. Conners and Y. Liu, "Source of ENF in
battery-powered digital recordings," in Proc. The Audio Engineering
Society 135th International Convention, pp.1-7.
[37] J. Chai, Y. Liu, Z. Yuan, R. W. Conners and Y. Liu, "Tampering
detection of digital recordings using electric network frequency and
phase angle," in Proc. The Audio Engineering Society 135th
International Convention, pp. 1-7.
[38] Y. Lei, Y. Zhang, J. Guo, D. Zhou, and Y. Liu, "The impact of
synchronized human activities on power system frequency", in Proc.
2014 IEEE Power Eng. Soc. Gen. Meet., pp.1-5.
[39] L. Liu, J. Chai, H. Qi and Y. Liu, "Power grid disturbance analysis
using frequency information at the distribution level," in Proc. 2014
IEEE International Conference on Smart Grid Communications, pp.
523-528.