IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 1, FEBRUARY 2004 139 [615323]

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 1, FEBRUARY 2004 139
Embedding Remote Experimentation in Power
Engineering Education
Mihaela M. Albu , Member, IEEE , Keith E. Holbert , Senior Member, IEEE , Gerald Thomas Heydt , Fellow, IEEE ,
Sorin Dan Grigorescu, and Vasile Trus ¸c˘a
Abstract— Engineering education by its nature is a costly
program in university environments. Perhaps the most costlycomponent is the laboratory facility, usually consisting of spe-cialized equipment. Effective instruction of some topics in powerengineering education requires experience with actual equipment,rather than small-scale replicas or simulation. In this paper, a newlaboratory approach is described, as implemented in a virtual,
Internet-based, experimentation platform. This virtual labora-
tory (VLab) utilizes real equipment distributed among multipleuniversities from which remotely located students can performexperiments. The software solution is a multiuser, client-serverarchitecture developed in the LabVIEW environment. Implemen-tation details including video, chat, archiving, and the hardwareand software platforms are presented in the paper. An examplepresented herein is the study of current and voltage waveformswhile controlling relays and low-voltage contactors. The applica-tions have been tested with student: [anonimizat]— Education, e-learning, relays and breakers clo-
sure, remote experimentation, web-based laboratories.
I. INTRODUCTION
THE higher engineering learning environment has to be at
least contemporary, if not ahead of the leading technolo-
gies of a society. Until now, we can report a significant effort
expended into organizing off-campus delivery of lessons usingmultimedia tools [1]. Laboratories based on simulation tech-niques have also been set up for remote access [2]. The first goal
of implementing such facilities into engineering curriculum is to
attract students to computer-mediated learning [3], [4].
Recently, scheduling flexibility for laboratory classes has
become an important issue, since many students have ex-
tended commitments, especially part-time students enrolled in
master-level programs. One solution addressing these needs isto develop web-based laboratories, on an Internet platform fromwhich real (versus simulation) experiments can be conducted
at anytime, without instructor surveillance or guidance.
Engineering education has also a costly component that is
not directly time-related: sophisticated (and implicitly expen-
Manuscript received April 3, 2003. This work was supported by the National
University Research Council of Romania (CNCSIS) under Project 134/CNCSIS
112/34967/2001.
M. M. Albu, S. D. Grigorescu, and V . Trus ¸c˘a are with the Department of Elec-
trical Engineering, Politehnica University of Bucharest, Bucharest RO-060042,
Romania (e-mail: [anonimizat]; [anonimizat]).
K. E. Holbert and G. T. Heydt are with the Department of Electrical Engi-
neering, Arizona State University, Tempe, AZ 85287-5706 USA (e-mail: hol-
[anonimizat]; [anonimizat]).
Digital Object Identifier 10.1109/TPWRS.2003.821020sive) equipment, whose use might be difficult because of in-
sufficient availability. Issues of equipment availability are often
addressed practically by the time multiplexing of laboratory
schedules and through the use of laboratory teams of studentswho work together rather than individually. But each of these so-lutions has their drawbacks—mainly relating to providing each
student with hands-on experiences in a way that can practi-
cally fit into their education schedule. Existing equipment canbe shared among researchers and students enrolled in differentprograms and with different schedules and knowledge levels.
Offline data processing, often required after completing a lab-
oratory experiment, can be accomplished either in the class-room—which supposes a longer student presence in the labora-tory and an appropriate computer setup––or in other locations,
which means that experimental data have to be somehow trans-
ferred to the offline processing unit, the strongly preferred solu-tion being in this case the use of the Internet layer. When simul-
taneous users are dispersed and remotely located, they are not
sharing the same real location, but a virtual one: the laboratoryplatform.
Institutional collaboration can be achieved through the virtual
exchange of equipment, which can be accomplished in a “web-ring.” In this concept, each institution provides specific experi-
ment(s), for which laboratory equipment is shared together with
experimental data and protocols. The bandwidth and number ofusers from each institution is “paid” in proportion to their con-tribution in terms of equipment and accessibility time. This is an
impetus for overseas cooperation since the time differential fa-
vors extended use during periods when the home institution stu-dents are otherwise asleep. The diversity of power engineering
topics taught and practically performed can be extended in this
way.
Most educators have utilized Java platforms for greater porta-
bility and easier access via web browsers [5]. There are also
projects [6] developed in the LabVIEW environment that can
be executed on a LabVIEW player freely distributed by NationalInstruments.
II. V
IRTUAL LABORATORIES IN POWER ENGINEERING
The dominance of the Internet is acknowledged in the devel-
opment of information and communication technology that hasmade Web-based distributed solutions increasingly attractive.
Apart from providing other services, the World Wide Web is
being looked upon as an environment for hosting modeling andsimulation applications [2], [7]. One of the major advantagessuch models provide is their ability to help students develop
0885-8950/04$20.00 © 2004 IEEE

140 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 1, FEBRUARY 2004
technical skills. The users manipulate virtual hardware (a sim-
ulator) to develop proficiency for operating the correspondingreal world system. To preserve the advantage of the Internet,
web-based laboratories require a tradeoff between simulation
and actual operation of the laboratory equipment. A laboratoryis commonly defined [8] as: 1) a place equipped for experi-mental study in a science or for testing and analysis; a place pro-
viding opportunity for experimentation, observation, or practice
in a field of study, or 2) an academic period set aside for labo-ratory work [9]. A virtual laboratory is defined [9] as an inter-active environment for creating and conducting simulated ex-
periments: a playground for experimentation. It consists of do-
main-dependent simulation programs, experimental units calledobjects that encompass data files, tools that operate on these ob-jects, and a reference book.
Consider the scientific meaning of simulation as “goal-di-
rected experimentation with dynamic models ”[10]. Also, de-
note a virtual laboratory as being based on the first definition of
laboratory presented above, that is, a place providing opportu-nity for experimentation. In the work described below, a virtuallaboratory becomes a virtual location inside the web, having
a distributed nature and a dynamic configuration. Laboratory
equipment available in the web-ring can be shared together withexperimental data and protocols. In the field of remote experi-ments used for distance learning purposes, one identifies two
major solutions. The first approach uses one or more data ac-
quisition cards as a versatile interface between the physical phe-nomenon and the digital realm. The experiments are accessedeither synchronously or asynchronously by at least one user via
different software.
A second pedagogical method uses standard digital instru-
ments that can perform either standalone or connected to an ex-ternal processing unit, usually a local personal computer (PC).
The communication layer allows either serial or parallel data
streaming and drastically determines the overall performanceof the experiment. The first approach is more suitable for com-plex experimental setups and imposes special requirements in
terms of security access [11]. The second method is mostly ori-
ented toward realizing a more realistic software replica of theequipment itself. In this latter technique, a special concern is theequipment safety. Topics from distributed measurement systems
field of study are very suitable for this second type of remote ex-
periment. Fig. 1 shows the front panel of a client application asused in a Internet-based calibration of a digital voltmeter [e.g.,the device under test (DUT), physically connected —at the re-
mote location —to a calibrator unit (CLD)] and a hardware plat-
form connected to the Internet [12].
III. R
EMOTE EXPERIMENTS OVER THE INTERNET
In this section, an experiment designed for students enrolled
in electrical engineering is presented. The particular hardwareemployed here is organized around a data-acquisition (DAQ)
board. Ultimately, this online laboratory seeks to present
fundamental topics in commutation such as synchronouscontact closure in three-phase systems, closing time dispersionof a switches lot, relay command, and electromagnetic noise
Fig. 1. Example of the front panel (client application) of the remote calibration
process, using a step-by-step calibration procedure. The equipment is located in
a measurement laboratory at Politecnico di Milano, Italy. The actual calibration
process is fully controlled from a web location (this example is hosted on a
computing platform at the Politehnica University of Bucharest, Romania).
measurements. The application is also intended to serve as
online course support for the in-class lectures on measurementssystems and other power engineering topics.
This virtual laboratory (VLab) utilizes equipment and
software distributed between two universities from which re-
motely located students can perform experiments. The softwaresolution is a multiuser, client-server architecture developed inthe LabVIEW environment. The server application is unique
to the communication interface and protocol (e.g., parallel,
Centronics-based) between the computer and the specificlaboratory hardware. Fig. 2 shows the particular architecture ofVLab. One of its nodes, VLab PUB, hosts a total of eight PC
stations, and a network server, interconnected via a fiber-optic
link with the main node of the Romanian Academic RoEduNetwide-area network. Each remotely accessed device is connectedto a PC running a specific server application. Then, the user
is operating the equipment via a client application that can
be formed by multiple client modules. All servers, includingthe one responsible for user authentication are running onPCs located in VLab PUB. It is not necessary for all server
applications to be housed on the same machine; in fact, the
bandwidth availability can be improved if the video server, forexample, is running on a separate computer.
Fig. 3 shows the schematic of the experimental setup in the
case of three-phase contactors. The three voltages and three cur-rents are acquired on six differential inputs of the data acqui-
sition card (PCI-AT-E4 from National Instruments), while its
maximum sampling rate (200 kHz) assures a maximum delaybetween the input channels of 30
. This value is acceptable
when considering the closing time of typical contactors. Stu-
dents are required to save data and then process it in a special-
ized environment (e.g., Matlab) to find the dependence of con-tacts nonsynchronicity from the load type. The control signal,sent from a DAQ digital output, is the time reference.

ALBU et al. : EMBEDDING REMOTE EXPERIMENTATION IN POWER ENGINEERING EDUCATION 141
Fig. 2. Virtual laboratory (Vlab). a) General diagram of the VLab architecture;
b) an experimental setup.
Fig. 3. Simple schematic for studying the three-phase contactor for motor-type
load using a DAQ card.
Figs. 4 and 5 show typical waveforms as seen on the front
panel of the client application. In this example, there are two on-line users, one of them becoming the master user after selectingthe control switch. All of the actual selections (i.e., number of
Fig. 4. V oltage and current waveforms at low-voltage contactor terminals
(located in Europe) as visualized at the remote-user location (U.S.).
Fig. 5. V oltage and current waveforms at low-voltage contactor terminals(located in Europe) as visualized at the remote-user location (U.S.).
acquired points, acquisition rate, etc.) made by the master user
are visible on each user ’s front panel.
Fig. 6 presents the schematic used for a synchronous con-
trolled eight-relay bank using a parallel port with the corre-
sponding voltages given in Fig. 7. The application is realized in
the LabVIEW environment. A client of the LabVIEW applica-tion is embedded into another application, dedicated to the timedispersion study of the relay switches.
IV. C
ONCLUDING REMARKS
Designing a multiple-access experimentation platform is
especially effective for power systems educational purposes.Students can practice utilizing the power devices in a tuto-rial mode, before they actually perform experiments. The
equipment —usually sensitive to voltages and currents outside
the rated range —is better protected from accidental misuse
by implementing various levels of (inexpensive) softwareprotection. Such an implementation allows real-time operation
of distributed research teams. Similarly, the specific master
(and client) software can be installed on the computing stationof an industrial testing unit, which could then be exploited forresearch purposes. Overall, the educational exposure versus

142 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 1, FEBRUARY 2004
Fig. 6. Simple schematic for controlling a bank of eight relays from the
computer parallel port.
Fig. 7. V oltage across contacts of eight synchronously controlled relays.
cost favors the use of remote electronic laboratory experiments
in many circumstances.
Some educators feel that virtual laboratories may not be as
effective as manipulating real equipment. For example, it can
be argued that students do not attain respect for power mea-
surements, and actual manipulation of test equipment affordedby hands-on experience; however, remotely located hardwaremay be superior to simulation. Further, students appear to be
attracted to computer use in this education application, thus
taking advantage of the students ’expertise in computer software
applications. Often overlooked is the fact that psychologicallyspeaking, the remote student is shielded against the adverse con-
sequences of the misconnection of equipment. Software protec-
tion interlocks prohibiting the student from making such mis-takes increases the longevity of the equipment. For this reason,some institutions might consider initially utilizing the remote
method to “train”the student before his/her use of the equipment
for in-laboratory experiments. Another drawback to simulationis the fact that the student does not enter the actual laboratory,so there is less hardware troubleshooting (e.g., loose wiring orconnections) whereas the (nonsimulation) approach taken here
retains these possibilities but in a restricted form.
One of the main advantages offered by VLab is that students
from all over the world can use the equipment located in a partic-
ular laboratory. Each university that is part of the ring of virtuallaboratory users can provide a different subset of experiments.The diversity of power engineering topics taught and practically
performed can be extended in this way.
A
CKNOWLEDGMENT
The authors gratefully acknowledge the contribution of F.
Mihai, member of the VLab team at Politehnica University ofBucharest.
R
EFERENCES
[1] L. Harasim, “A framework for online learning: the virtual-U, ”Com-
puter , vol. 32, no. 9, pp. 44 –49, Sept. 1999.
[2] A. Guru, P. Savory, and R. Williams, “A web-based interface for
storing and executing simulation models, ”inProc. Winter Simulation
Conf. , J. A. Joines, R. R. Barton, K. Kang, and P. A. Fishwick,
Eds. Lincoln, NE: Dept. Ind. Manage. Syst. Eng., Univ. Nebraska,
2000, pp. 1810 –1814.
[3] G. Karady, G. Heydt, M. Michel, P. Crossley, H. Rudnick, and S.
Iwamoto, “Review of power engineering education worldwide, ”in
Proc. IEEE Summer Power Meeting , vol. 2, Edmonton, AB, Canada,
July 1999, pp. 906 –915.
[4] J. Kyyra, “Challenges in the teaching of electrical power engi-
neering —motivation of students and change of paradigma, ”inProc.
Int. Conf. Eng. Educ. , Oslo, Norway, Aug. 2001, pp. 8D6-26 –8D6-30.
[5] M. Bertocco, F. Ferraris, C. Offelli, and M. Parvis, “A client-server ar-
chitecture for distributed measurement systems, ”IEEE Trans. Instrum.
Meas. , vol. 47, pp. 1143 –1148, Oct. 1998.
[6] K. Holbert and M. Albu, “A signal processing laboratory employing
on-line teaming for remote experiments, ”in Proc. Amer. Soc. Eng. Educ.
Annu. Conf. Expo., Nashville, TN, June 2003.
[7] M. Joler and C. G. Cristodoulou, “Virtual laboratory instruments and
simulations remotely controlled via the Internet, ”inProc. IEEE An-
tennas Propagat. Soc. Int. Symp. , vol. 1, Boston, MA, July 2001, pp.
388–391.
[8] Webster ’s Ninth New Collegiate Dictionary, 1983.
[9] The Virtual Laboratory Environment, L. Mercer, P. Prusinkiewicz,
and J. Hanan. (1990 –91). http://www.cpsc.ucalgary.ca/Re-
search/bmv/vlab.orig/vlab.html. [Online]
[10] T. Oeren, “Impact of data on simulation: from early practices to feder-
ated and agent-directed simulation, ”inProc. Europe. Simulation Soc.
Conf. , Shaping Future with Simulation, T. L.Th. L. van Stijn, I. Smit,
H. de Swaan Arons, L. Dekker, and A. W. Heemink, Eds., Delft, The
Netherlands, June 2001.
[11] X. Vilalta, D. Gillet, and C. Salzmann, “Contribution to the definition
of best practices for the implementation of remote experimentation so-
lutions, ”inProc. IFAC Workshop Internet-Based Contr. Educ. , Madrid,
Spain, Dec. 2001, pp. 37 –42.
[12] M. Albu, A. Ferrero, F. Mihai, and S. Salicone, “Remote calibration
using mobile, multi-agent technology, ”inProc. IEEE Instrum. Measure.
Technol. Conf. , Vail, CO, May 2003, pp. 21 –23.
Mihaela M. Albu (M’96) is from Craiova, Romania.
She received the Ph.D. degree from the Universitatea
Politehnica Bucuresti, Bucharest, Romania.
Currently, she is a Professor of Electrical Engi-
neering at the Universitatea Politehnica Bucuresti.
Her research interests include power quality,
instrumentation, and remote experimentation. She is
spending a leave at Arizona State University, Tempe,
as a Fulbright Fellow during 2002 –03.

ALBU et al. : EMBEDDING REMOTE EXPERIMENTATION IN POWER ENGINEERING EDUCATION 143
Keith E. Holbert (M’88–SM’96) received the Ph.D.
degree in nuclear engineering from the University of
Tennessee, Knoxville, in 1989.
Currently, he is an Associate Professor and Asso-
ciate Chair for Undergraduate Studies in the Elec-
trical Engineering Department at Arizona State Uni-versity, Tempe. His research interests focus on power
plant instrumentation, signal processing, and fuzzy
logic applications for power plants. He has been pub-
lished in many journal and conference papers.
Gerald Thomas Heydt (S’62–M’64–SM’80–F’91)
received the Ph.D. degree in electrical engineering
from Purdue University, West Lafayette, IN.
Currently, he is the Director of a Power Engi-
neering Center program at Arizona State University,
Tempe, where he is also a Regents ’Professor. His
industrial experience is with the Commonwealth
Edison Company, Chicago, IL, and E. G. & G.,
Mercury, NV .
Dr. Heydt is a member of the National Academy
of Engineering.
Sorin Dan Grigorescu is from Bucharest, Romania.
He received the Ph.D. degree from the Universitatea
Politehnica Bucuresti, Bucharest, Romania.
Currently, he is a Professor of Electrical Engi-
neering at the Universitatea Politehnica Bucuresti.
His research interests include electronic instrumen-tation, power transformers monitoring, and signal
converters.
Vasile Trus ¸c˘ais from Ramnicu, Valcea, Romania. He received the Ph.D. degree
from the Universitatea Politehnica Bucuresti, Bucharest, Romania.
Currently, he is a Professor of Electrical Engineering at the Universitatea
Politehnica Bucuresti. His research interests include power apparatus, electric
drives, and high-voltage measurement techniques.