Wide Area Monitoring System (WAMS)
Application in Smart Grids
2021 3rd International Youth Conference on Radio Electronics, Electrical and Power Engineering (REEPE) | 978-1-7281-8398-5/20/$31.00 ©2021 IEEE | DOI: 10.1109/REEPE51337.2021.9388040
D. S. Zolin1, E. N. Ryzhkova2
1, 2 ElectricalEngineering and Electrification
Department
1, 2 National Research University "MPEI"
1, 2 Moscow, Russian Federation
1diffl@mail.ru, 2ryzhkovayn@mpei.ru
Abstract— Digital microprocessor devices are widely used in
modern dispatching. The introduction of wide area
measurement systems (WAMS) is one of the priorities of the
technological development of power systems globally. WAMS
use a well-studied method of measuring phasor (phase vector)
data in various geographically remote sections of power grid
lines. With appropriate software processing, the data coming
from the phasor measurement units (PMUs) allows: providing
dispatch centers with data on the overall system stability;
increasing the reliability of maintaining the power system
operation state; obtaining cost savings by reducing or removing
restrictions on power flow capacities in critical sections in real-
time; maximizing the utilization of most efficient, competitive
power plants, reducing the volume of consumer restrictions.
Keywords— wide area monitoring system (WAMS), wide area
protecting system (WAPS), phasor monitoring unit (PMU), smart
grids, synchronized vector measurements.
I. INTRODUCTION
This paper contains developed methods for accelerated
calculations of electrical modes in relation to the problems of
emergency control within the framework of scientific research
in dispatch control improving and developing field based on
vector control and vector measurements. There are
summarized results of the developed algorithms tests based on
using of vector measurements using PMU - Phasor
Measurement Unit [1, 2] in this paper.
The development and improvement of large power
interconnections management on the basis of new computer
and information technologies is focused on using of electrical
regime parameters vector values synchronized measurements.
Mentioned PMU devices are the basis of EPS WAMS
(Wide Area Measurement System). So far, in domestic
practice, PMUs are connected directly to the measuring
circuits of electromagnetic current and voltage transformers,
as well as to the source of the exact time signal, from which
the PMU measurements are synchronized. The satellite
navigation system GPS or GLONASS is used as a source of
the exact time signal, providing synchronization of
measurements with an accuracy of 1 μs.
Although such measuring system receives instantaneous
current and voltage values, which are programmatically
converted into corresponding complex values vector display,
it does not promote achieving the required accuracy of
measured current and voltage parameters. This is quite
acceptable for WAMS, but does not give the quality effect in
the tasks of electrical system steady-state modes online-
calculations.
From this point of view, it is more promising to use PMUs
based on optical current transformers (OCT) and voltage
(OVT). In this case, the full use of information obtained by
PMU makes possible to solve various problems at a new
978-1-7281-8398-5/21/$31.00 ©2021 IEEE
Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.
qualitative level, including the ability to provide a significant
reduction of modes calculation time, which is especially
important in online operational calculations, in emergency
automation systems operating in the "Before” mode, as well
as in WAMS for monitoring stability at small fluctuations [3].
If we talk about the operational mode calculations based
on telemetry data, then in this case, the task of assessing the
state (SA) based on vector measurements and information
from SCADA inevitably arises [4]. At the first stage of PMU
placement algorithm development in order to ensure steady-
state EPS super-fast calculations [1, 2] based on testing
networks up to 3500 nodes, it was shown that for real high-
dimensional networks the required number of nodes with
PMU turns out to be 20-25% of the total. Vector current
measurements using in the branches incident to the node with
PMU, almost halves the number of PMUs, which is extremely
important from an economic point of view [5]. In this case, an
algorithm is considered when a certain number of PMUs is
already installed in the network in advance according to other
criteria.
Reducing the number of PMUs for electrical mode
accelerated calculation leads to the need of using a small
amount of active and reactive generator and load power
traditional electromagnetic measurements. Since the errors of
these measurements are significantly higher than those of the
PMU, it becomes an important issue to provide an accelerated
electrical network state assessment (SA). An algorithm in
which the SA can also be considered super accelerated is
shown in this paper.
II. STRUCTURING FUNDAMENTAL WAMS
CONFIGURATION BASED ON REAL POWER GRID
APPLICATIONS
A. WAMS main components
EPS dynamics monitoring can prevent or mitigate the
effects of large-scale disturbances. A WAMS system based on
PMU (Phasor Measurement Unit) measurements provides
dynamic EPS monitoring. Collection of phasor measurements
from EPSs in real to a central node allows establishing the
basic principles of WAMS, WAPS and WACS for real-time
procedures [6].
Conventional supervisory control and data acquisition
(SCADA) systems provide stable, unsynchronized data about
EPSs with a time resolution of 1 to 10 s. Conventional
systems are limited to measurements of steady-state modes
only and cannot be used to analyze system dynamics [7].
WAMS allows to constantly monitor the dynamic mode of
system operation. WAMS provides time-synchronized data
every 20 ms (for a 50 Hz system) with all data timestamped.
Moreover, the data is synchronized with millisecond precision
[8].
A general WAMS configuration is shown in Figure 1.

Fig. 1. WAMS configuration.
The core of WAMS is a complex device consisting of
three main components:
• Phasor data concentrator (PDC) [9] and real-time data
analysis;
• Data archiving;
• Data visualization using Web SCADA.
B. Functions performed by PDC
• Continuously collects measurements from PMU;
• Provides real-time system monitoring and control;
• Stores measurements in a database.
The system receives measurements at a sampling rate that
depends on the utilized PMUs. Server performance determines
the maximum number of PMUs, as well as the algorithms that
can be used.
C. Communications
The system provides a wide range of communication
protocols for collecting and receiving synchronized phasor
data.
For instance, PDC supports conventional IEEE c37.118
[7] and IEEE 1344 communication protocols for PMU
communications. PDC can receive data from at least 100
PMUs with a full resolution of 50(60) samples per second.
As a server, PDC also supports conventional IEEE
c37.118 and IEEE 1344 communication protocols. The server
may connect to other PDCs to exchange the required data in
real-time with other companies, such as transmission network
system operators.
As a server, PDC sends data simultaneously to different
clients at a full resolution of 50(60) samples per second.
Furthermore, the data transfer resolution can be selected by
the client's request.
Additionally, the PDC can communicate with other
systems using standard protocols such as IEC 60870-5-
101/104, DNP3, etc. Such connections can be used to send
data and receive commands from the control center (in
SCADA/EMS systems).
D. Data analysis
This stage involves complex real-time calculations
required for reliable EPS operation:
Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.
• Voltage stability is calculated to monitor the power
transmission over specific transmission lines.
• Frequency stability. Frequency stability calculations
allows detecting frequency changes resulting from
sudden large power shortages or sudden load loss.
• Power fluctuation. The power fluctuation function is
used to identify power fluctuations on lines. This
function allows calculating the dominant fluctuation
frequency, its amplitude and damping factor.
• Phase angle monitoring traces phase angle difference
between two power grid substations.
This difference increases with the system load increase.
Phase angle difference information is also essential for
successful communication between two substations. Thus,
phase angle monitoring includes:
• Measurement of phase angle difference between two
power grid substations;
• Assessment of threshold violation that triggers an
alarm and warns the dispatcher about the hazard.
E. Monitoring function to detect value
increase/decrease
If the value of the analyzed value is greater (or less) than
the threshold value, the monitoring function for this value is
initiated. The principle is similar to the monitoring function
for the phase angle difference.
F. Module difference monitoring function
If the difference between the modules of two values
exceeds the threshold value, the monitoring function is
activated to detect the difference between the modules. The
principle is the same as for the phase angle difference
monitoring function.
G. Custom functions
The system consists of programmable logic tasks and
utilizes a large number of libraries of operations for various
purposes, such as arithmetic, numeric, logical, complex
number arithmetic, comparisons, etc. Thus, custom function
blocks can be created. An example of a custom block is a
troubleshooting diagram.
H. Cable line temperature monitoring
Conductor temperature is an important parameter. It
causes conductor strain and sag. Power lines can be
mechanically damaged if the permissible temperature is
exceeded.
Conductor temperature measurements can be divided into:
• Indirect temperature measurements to determine the
average temperature of the entire line;
• Direct temperature measurements to accurately
measure the line temperature at the data collection
point.
III. INVESTIGATION OF INDIRECT CABLE LINE
TEMPERATURE MEASUREMENT METHODS USING WAMS
A. Indirect cable line temperature
measurement method description
One of the indirect temperature measurement methods is
based on calculating the line impedance [11]. If PMUs are
installed at both ends of the line, the line active impedance can
be determined using the current and voltage phasors. After
that, the average temperature is calculated using the active
impedance.
Using a U-shaped equivalent circuit of the line (Figure 2),
the following equations can be obtained:
I =U Y + U −U Y (1)
I =U Y + U −U Y (2)
Y=
Z= = R + jX
where I1 and I2 – current phasors;
U1 and U2 – voltage phasors at line ends.
Fig. 2. U-shaped equivalent circuit of the power transmission
line.
The relationship between the active impedance and the
line temperature T can be expressed as a ratio:
R=R 1+a T−T
B. Described method disadvantages analisys
The described method has the following disadvantages:
• The calculation gives the average temperature of the
entire line;
• Locations with high temperatures, conductor sagging
and critical spans cannot be detected;
• Two PMUs must be used — one at each end of the
line;
• Instrument errors can significantly impact the results.
C. Described method practice application
The effects of instrument errors are shown in the following
example.
Let us consider a 80 km long 400 kV line. The following
synchronized phasor measurements were obtained for both
ends: U1=396,48кВ, φ1=70,36°, U2=392,94кВ, φ2=66,51°,
I1=600,47кВ, φ1=121,53°, I2=616,98кВ, φ2=52,78°.
Taking into account (4), R=2.345 Ω, while T=21.26°C.
Assuming the voltage accuracy of the voltage phasor U1 is
0.1%, then R=2.656 Ω, while T=54.62°C.
Thus, measurement error significantly affects the
temperature calculations. Calibration allows to overcome this
error.
Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.
Other methods of indirect temperature measurement
include the method based on current measurement [12]:
+ T= I k (6)
!" !"
where k = $%&
is a constant value.
$%&
This equation requires the current temperature and the
ambient temperature to be known. Like the previous one, this
method determines the average temperature of the entire line.
(3) The disadvantage of indirect temperature measurement can
be eliminated by using direct temperature measurement. In
this case, the actual line temperature is measured at the sensor
(4) locations. The resulting solution factor in the effects of
ambient temperature and wind. Temperature sensors are
usually installed in critical sections of the line [13].
A complete EPS monitoring solution is obtained by
enhancing line data in WAMS with accurate line temperature
measurements.
IV. METHODS OF OPERATIONAL DATA STORAGE DURING
WAMS OPERATION
A number of databases with a different set of structural
and functional qualities have been created for WAMS
operation. Their comparative analysis is provided below.
A. Real-time cyclic database
This database allows storing measured and calculated
values of current, voltage, power, various stability indicators,
etc., at the maximum sampling rate of 1 sample per cycle —
50(60) samples per second [14]. The main database stores 14
days of data. If required, data can be stored for longer periods.
B. Event database
Trigger functions can be configured to use measured or
(5) calculated values. When the threshold is exceeded, the trigger
function adds the event to a persistent event database. Events
are used to alert about the danger in real-time. Furthermore,
they can be used for post-accident analysis. Thus, the
dispatcher receives the information required for further
decisions. Events can be used by the following functions:
• For example, to detect increase or decrease of voltage,
• Frequency, any calculated values, etc.
• To detect differences, such as phase angles between
two voltage phasors.
C. Disturbance database
By analogy with event registration, the system allows
registering disturbances. Additional thresholds for
parameterisation of trigger functions for recording
disturbances are the times before and after the event. Time
determines the length of the stored measured and calculated
data record. The duration of disturbance records can reach 30
min of data selected by the user. Disturbance records are
permanently stored on hard drives independently of the
cyclical database and are accessible in real-time [15].
V. PMU DEVICES MEASUREMENT NETWORK
STRUCTURING ALGORITHM
Speaking about these algorithms, the use of not only
vector voltage measurements (module and phase), but also
vector measurements of the incident branch currents (angle '(
between the current vector I)'( and the voltage vector U)' ) is
discussed. Moreover, some of the nodes contain classical supposed to install a PMU) and incident nodes assigned to it
electromagnetic measurements (EMP). as “pseudo-PMU” and denoted as PP, lower the PR and BP in
the rest of the network. Vector measurements of branch
In one of the first works known to us [16], which poses the
currents and voltages of the beginning of power transmission
electrical network full observability ensuring problem with allow to determine with high accuracy the complex value of
minimizing the number of PMUs nodes, an algorithm the voltage of the opposite end. It is further considered as a
forxthese devices arrangement is described and the result for pseudo-dimension of a PP-type node.
the IEEE-14 test network is given, according to which 5 nodes
(numbers: 1 , 4, 6, 10, 14) must be equipped with a PMU. The Sequential elimination of nodes in accordance with the
concept of pseudo-PMU is introduced, hereinafter referred to discussed algorithm leads to a new, transformed matrix of
as PP. In this work, all other nodes turned out to be of the PP conductance Y,∗ , which takes the form shown in Fig. 4 for the
type. IEEE-14 network. An important key point of the algorithm is
In a later work [17], an algorithm with improved local subsystems appearing, in which the voltages of several
computational characteristics was proposed. The results of nodes (most often, one, as can be seen in Fig. 5) are calculated
placing PMUs for the same network are as follows. Received on the basis of information about the voltages in PP-type
4 nodes with PMU: 2, 6, 7, 9; all other nodes are of PP type. nodes and electromagnetic measurements of the power of the
When developing a new algorithm for placing PMUs, for the nodal injections obtained.
sake of generality, we accepted the condition that it is
necessary to take into account the cases when PMU locations
(existing or expected to be installed according to certain
criteria, for example, at large power plants, substations for
WAMS, etc). The task is formulated as a search for placing
additional PMUs to ensure EPS modes super-accelerated
calculations without solving cumbersome systems of
equations for the steady-state EPS mode. The algorithm does
not change its essence if there are no specified PMUs when
setting the problem. In general, we will assume that some
PMUs have a predetermined location or such PMUs already
exist.
At the first stage PMU arrangement algorithm
development in order to ensure EPS modes accelerated
calculations, high-precision vector measurements of the node
voltages - the phase and the modulus of the complex value,
which the PMU can provide - were considered.
First algorithm basis was lower triangular submatrix ***
YΔ of
electrical network conductivity matrix formation, which
provides nodes voltages accelerated calculation without PMU, Fig. 3. Conductivity matrix after rearranging rows and columns according
without solving equations system, due to the sequential to the first algorithm.
determination of voltages from equations with one sought
variable. As seen from Fig. 3, node 3 voltage is calculated
through the solution of one equation and is substituted into all
downstream equations. Now node 2 voltage is also calculated
through the solution of one equation and is also substituted
into all downstream equations, etc. In the considered
algorithm, as in [17], 4 nodes of the PMU type are obtained:
4, 2, 9, 10. However, in this case, the formed conductivity
matrix provides an accelerated mode calculation without
solving the equations system.
Further improvement of the algorithm for arranging vector
measurement devices is based on using high-precision current
measurements in the branches incident to the PMU node. It
leads to a significant decrease in the PMU volume up to 10-
15%. At the same time, the number of traditional
electromagnetic measurements is reducing in the same
proportion.
Let us introduce terminology of primary and secondary
Fig. 4. Conductivity matrix Y,∗ after permutation of rows and columns
ranks of a node. The primary rank (PR) of a node is the according to the second algorithm.
number of lines (NL branches) incident to a given node. For
example, node 3 has PR3 = 2. By the secondary rank of a node
we mean the nodes primary ranks sum incident to this node.
Considering that PR2 = 4 and PR4 = 5, we get secondary rank
of node 3 - NL3 = 9.
The second algorithm is based on ranking of all nodes by
BPi in decreasing order, starting from the maximum.
Elimination of the node with the maximum BP (this node is

Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.

Fig. 5. IEEE – 14 network. There are 2 nodes with PMUs: 6 and 4; PP type
nodes: 2, 3, 5, 7, 9, 11, 12, 13; EMI type nodes: 1, 8, 10, 14. Shown 4
local networks for an autonomous OS.
VI. RESULTS
It is noted in [4, 18] that the maximum PMU using effect
in EPS state assessing can be achieved with SCADA system
PMU and traditional TI data joint use. If reliable
measurements are available, sufficient to perform a flux
distribution calculation, state estimation procedure can be
replaced by a steady state calculation. Moreover, the PMU
measurements are considered accurate and can participate in
EPS state assessing as constant values. These values remain
unchanged, but they affect the feedback process themselves
(affect the estimates of the phase and voltage modulus in
nodes that do not have PMUs). Therefore, further all points of
the OS algorithm are performed.
Since appearing “island type” local subsystems in matrix
formation algorithm, as shown in Fig. 5, in which some EMP
type nodes set is surrounded by PP type nodes, then traditional
OS problem breaks down into a set of local OS. The dotted
lines in Fig. 5 shows local networks of small dimension. The
local network state assessment can be carried out 2 magnitude
orders faster than the original network state assessment. This
allows us to speak about super-accelerated EPS state
estimation.
The most important result of algorithm development for
steady-state modes super-accelerated calculations based on
vector measurements was not only minimizing the number of
nodes with PMUs, but also providing the possibility of an
accelerated state assessment due to the fact that it is necessary
to use a part of information from electromagnetic measuring
devices with an error of magnitude order higher than PMU.
VII. CONCLUSION
Reducing the number of nodes with PMUs for electrical
network mode accelerated calculation leads to using some of
the traditional active and reactive power electromagnetic
measurements (EMP), and its errors are significantly higher
than that provided by the PMU. Evaluation becomes an
important issue. Interest in sharing information from PMU
and SCADA appeared immediately with vector measurement
systems WAMS appearance.
The developed system is designed to control EPSs in real–
time and to warn system operators if the measured or
calculated values exceed the thresholds. Furthermore, the
system allows to exchange measurements with other PDC and
SCADA systems. WAMS potential comes from the operation
Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.
of WAPS and WAMS in real-time, though widespread use of
such scheme will require a more reliable communication
infrastructure compared to the currently available.
Future development shall be mainly focused on the
following:
• EPS WAPS, preventive and repair procedures,
including control actions for tap changers and reactive
shunts, division into subsystems, etc.;
• Development of an expert system to prevent system
failures;
• System relay protection for conventional protection
schemes;
• Complex state assessment functions.
REFERENCES
[1] P.I.Bartolomey, S.I.Semenenko, "Vector measurements number
minimization for EPS modes accelerated calculations", Electric power
engineering through the eyes of youth: scientific works of the 6th
International scientific and technical Conf .: in 2 volumes - Ivanovo, T.
1, p.207–212, 2015.
[2] P.I.Bartolomey, S.I.Semenenko, "Arrangement of PMUs for EPS
modes calculation accelerating in emergency control problems in
WAMS environment", Modern trends in power systems relay
protection and automation systems development: collection of articles.
Sochi, p.5–9, 2015.
[3] B.I.Ayuev, A.V.Zhukov, "New approaches to electric power systems
stability margin monitoring", Power system: management; competition,
education: collection. Yekaterinburg, USTU-UPI, T.1, p. 9–15, 2008.
[4] A.M.Glazunova, I.N.Kolosok, E.S.Korkina, "PMU data application in
EPS state assessing by control equations method", Operational
management in the electric power industry. Staff training and
qualifications maintenance. No. 1, p.44–49, 2009.
[5] A.V.Danalin, "Solutions based on electrical modes parameters
synchronized vector measurements technologies in the world",
[Electronic resource]. URL:
http://www.alteropower.ru/upload/medialibrary/f4e/f4efb302290ad637
daf4bb5c2eb9904c.pdf (accessed 05.08.20).
[6] P.K.Mahesh, M.Gavande; M.F.A.R Satarkar, "Phasor Measurement
unit technology and its applications – a review", 2015 International
Conference on Energy Systems and Applications, 30 Oct.–1 Nov.
2015.
[7] R.Luís, T.Cruz; P.Simões, E.Monteiro; L.Lev, "Attacking SCADA
systems: A practical perspective", 2017 IFIP/IEEE Symposium on
Integrated Network and Service Management (IM), 8-12 May 2017.
[8] A.Popelka; D.Jurik; P.Marvan; V.Povolny, "Advanced applications of
WAMS", 22nd International Conference and Exhibition on Electricity
Distribution (CIRED 2013), 10-13 June 2013.
[9] J.S.Bhonsle, A.S.Junghare, "An optimal PMU-PDC placement
technique in wide area measurement system", 2015 International
Conference on Smart Technologies and Management for Computing,
Communication, Controls, Energy and Materials (ICSTM), 6-8 May
2015.
[10] I.V.Gusev, "Overview of the IEEE C37.118 standard and its
requirements for synchronized vector measurement systems", Digital
energy. [Electronic resource]. URL –
http://digitalsubstation.com/blog/2014/03/05/obzor-standarta-ieee-c37-
118-i-ego-trebovanijj-k-sistemam-sinkhronizirovannykh-vektornykh-
izmerenijj/ (accessed: 10.11.2020).
[11] M.Zima, M.Larsson, P.Korba, C.Rehtanz and G.Andersson, "Design
Aspects for Wide-Area Monitoring and Control Systems", in
Proceedings of the IEEE, vol. 93, no. 5, pp.980-996, May 2005, doi:
10.1109/JPROC.2005.846336.
[12] IEC 255-8 Standard, Electrical relays. Thermal electrical relays,
Second edition, 1990-09.
[13] V.Lovrenиiи, M.Polak, R.Tomaћiи, S.Peuliж, "Diagnostics of
overhead lines provides benefits to a transmission line operator", 9
Konf. Slovenskih elektroenergetikov (CIGRE B2-12), Kranska Gora
(Slovenia), 2009.
[14] R,Barumerli; M.Geronazzo; F.Avanzini, "Round Robin Comparison of
Inter-Laboratory HRTF Measurements – Assessment with an auditory
model for elevation", 2018 IEEE 4th VR Workshop on Sonic
Interactions for Virtual Environments (SIVE). 18-18 March 2018.
[15] S.Sejal, M.Agarwal, "Real time database management system", 2018
2nd International Conference on Inventive Systems and Control
(ICISC), 19-20 Jan. 2018.
[16] T.L.Baldwin, L.Mili, M.B.Boisen, R.Adapa, "Power system
observability with minimal phasor measurement placement", IEEE
Transactions on Power Systems, Vol. 8(2), p. 707–715, 1993.
[17] N.M.Manousakis, G.N. Korres, "A weighted least squares algorithm
for optimal PMU placement", IEEE Transactions on Power Systems",
Vol. 28(3), p. 3499–3500, 2013.
[18] A.Z.Gamm, A.M.Glazunova, Yu.A.Grishin, I.N.Kolosok, E.S.Korkina,
"Algorithms development for electric power system state assessing
based on SCADA and PMU data integration", Electricity, No.6, p.2–9,
2009.

Authorized licensed use limited to: University of Prince Edward Island. Downloaded on July 03,2021 at 16:53:12 UTC from IEEE Xplore. Restrictions apply.