Agent-based Intelligent Control for Real-time

Operation of a Microgrid
H. N. Aung, Member, IEEE, A. M. Khambadkone, Senior Member, IEEE,
D. Srinivasan, Senior Member, IEEE, and T. Logenthiran, Student Member, IEEE
Abstract- This paper presents an intelligent agent-based
control for real-time operation of a hybrid microgrid in both
islanded and interconnected modes. The proposed control
architecture contains different types of agents such as
Photovoltaic agent, Fuel cell agent, Distributed Generator agent,
Power Converter Building Block agent, Composite Energy
Storage System agent and Load agent, which represent each
major components in the microgrid. In this paper, the control
system implementation will be discussed in details by describing
each agents and the control algorithm associated with ontology
creation. The proposed control system is developed as a Multi
Agent System (MAS) in JADE platform and the microgrid is
implemented in Real Time Digital Simulator (RTDS). A real-time
communication interface between MAS and RTDS is presented
via TCP/IP. To demonstrate the effectiveness of the proposed
control architecture, simulation studies have been performed on
the developed hybrid microgrid. The simulation results indicate
that the proposed agent-based control system effectively
coordinates the distributed energy resources in real-time for
operation of both islanded and grid connected modes.
Index Terms- Microgrid, Multi-Agent System, Real Time
Digital Simulator, Real-Time Operation

I. INTRODUCTION
Over the years, the growing demand of electric power and
the increase in the price of fossil fuels along with the CO2
emission, renewable energy sources have been seen as an
interesting solution. To make use of the renewable energy
sources efficiently, the microgrid concept provides an
effective approach. Microgrid is the interconnection of small,
modular generation sources (PV, fuel cell, wind, microturbines, etc.) to low voltage distribution systems, together
with storage devices (flywheel, super capacitors and batteries)
and controllable loads. Microgrid can be connected to the
main power grid, or islanded, if disconnected from the main
power grid [1] [2].

This work was supported by Science and Engineering Research Council of

the Agency for Science, Technology and Research (A*STAR), Singapore
IEDS programme grant R-263-000-507-305.
Htay Nwe Aung is with the Department of Electrical and Computer
Engineering, National University of Singapore, Singapore (htaynweaung@
nus.edu.sg).
Ashwin M. Khambadkone is with the Department of Electrical and
Computer Engineering, National University of Singapore, Singapore
(ashwinmk@ices.a-star.edu.sg).
Dipti Srinivasan is with the Department of Electrical and Computer
Engineering, National University of Singapore, Singapore (dipti@
nus.edu.sg).
T. Logenthiran is with the Department of Electrical and Computer
Engineering, National University of Singapore, Singapore (logenthiran@
nus.edu.sg).

In [3], the authors proposed multifunctional Power

Converter Building Block (PCBB) to connect hybrid
microgrid (DC+AC bus) to area EPS. Photovoltaic, Fuel Cell
and the energy storage operate with DC voltage, and it is
possible to connect to DC bus. The AC bus is used to connect
all the AC sources and the AC loads inside the microgrid.
Therefore, it could take power from the main grid to feed the
loads inside microgrid directly without extra power conversion
with the combined active power and reactive power control of
PCBB. To apply this concept, a hybrid microgrid (DC+AC
bus) with PCBB is proposed in this paper. The proposed
hybrid microgrid is depicted in Fig.1.
This paper proposes agent-based control system for the
control and management of a hybrid microgrid. The multi
agent system was developed in JADE and the microgrid was
implemented with Real Time Digital Simulator RTDS [4].
The agents in the system communicate and coordinate each
other to achieve the goal of the system such as functionality of
island and reconnection a microgrid, optimum usage of
distributed resources and the management of the power and
the loads. This capability can be recognized as an intelligent
control for a microgrid. The main objective of this research is
to design and implement an intelligent control for microgrid in
real-time.
The remaining paper is organized as follows: In Section II,
agent technology is introduced. Section III illustrates the
architecture of the hybrid microgrid and the control system.
Section IV discusses the implementation of the control system
for microgrid operation. Section V describes the control
algorithm developed for microgrid operation in both islanded
and grid connected modes. Section VI demonstrates the
proposed control system can successfully control a hybrid
microgrid in real-time. Finally, Section VII concludes the
paper.
II. AGENT TECHNOLOGY
Multi-agent system is a system collection of intelligent
agents that work together to achieve a global goal. MAS
allows an intelligent control of a microgrid, where each
necessary element in a microgrid is represented by an
intelligent agent that uses a combination of AI-based and
mathematical models to decide on optimal actions. Each
power source, component and load in a system can therefore
be represented as an intelligent agent. The fundamental
element of a multi-agent system is an intelligent agent which
has the three typical characteristics, namely, reactive,
proactive and social abilities [5] [6]. These characteristics
signify the importance of the agent technology in developing a
complex system. For over a decade, multi-agent system is

being used in an increasingly wide variety of applications for

power system engineering applications [7]-[9]. The authors
[10] [11] gave an excellent insight into concepts, approaches,
technical problems and potential values of multi-agent system
application in power systems. The authors discussed the
standards, tools, supporting technologies and design
methodologies that could be incorporated for MAS
implementation in power systems.

operation by negotiating and sharing the responsibilities with

the lower layer agents.

III. ARCHITECTURE OF THE MICROGRID AND THE CONTROL

SYSTEM
This section describes the proposed agent-based control
system architecture for a hybrid microgrid. Fig. 1 shows the
schematic diagram of the architecture. In order to discuss an
agent-based architecture, the structure of the proposed hybrid
microgrid is discussed first.
System under the study consists of three energy sources
namely photovoltaic, fuel cell and distributed generators. To
maximize output power of PV, PV system is connected
through the maximum power point tracking (MPPT) under all
conditions by controlling a DC/DC converter. Because of low
output voltage of FC, DC/DC converter of FC steps up the
voltage to the DC bus voltage of 800 volts. PV, FC and
Composite Energy Storage System (CESS) are connected to
the DC bus of microgrid whose voltage is fixed and controlled
by CESS. CESS consists of batteries and super capacitors that
store energy with high storage efficiency and capable of
deliver high power. The presence of fluctuating due to the
renewable sources and the load, CESS generates or absorbs
power to ensure energy balance at all times. It also maintains
stability and power quality of the microgrid [12].
Distributed generators and all the AC loads inside the
microgrid are connected directly to the AC bus. The microgrid
operation can be classified into islanded mode, gridconnection mode and smooth transition between both
operation modes.
PCBB plays a major role in this area which controls
combined active and reactive power [3]. The major task of
PCBB is to facilitate both the connection of DC and AC bus
within the microgrid and the connection of the microgrid to
main grid simultaneously. Furthermore, PCBB stabilizes the
voltage at point of common coupling (PCC) when the main
grid encounters sag or swell and PCBB smoothly and quickly
shift the microgrid to island [13]. The microgrid connects to
the main grid via PCC and that defines the separation between
the microgrid and the main distribution grid. Loads are
divided into two groups: critical and non-critical controllable
loads.
Each unit in the system is represented by individual agent
and the architecture is composed of several different agents. In
order to achieve the full benefits from the operation of
microgrid, it is important that the integration of the distributed
resources into the LV grids, and their relation with the MV
network upstream [1] [14]. To achieve this goal, hierarchical
control architecture is proposed. MicroGrid Controller (MGC)
is in upper level and responsible for optimizing the microgrid

Fig.1. Proposed Hybrid Microgrid Architecture with PCBB, MC,LC and

MGC

In addition to this, MGC also responsible for local

optimization of the micro-sources power production and loads.
The lower level of control consists of Micro-source controller
(MC), Load controller (LC) and point of common coupling
(PCC). The MC controls PV, FC, CESS and DG, which are
inside the microgrid. The LC controls critical and non-critical
loads. PCC agent is responsible for disconnection and
reconnection of a microgrid by monitoring the grid voltage,
phase-angle and frequency. The lower layer agents have the
autonomy to perform local goal of the individual assigned task
and the interactions with the higher layer agents. The
individual agent coordination and cooperation provide the
distinct benefits to the performance of the control system.
IV. IMPLEMENTATION OF THE CONTROL SYSTEM
A. Agent Platform
JADE (Java Agent DEvelopment framework) was selected
for building the control system. JADE is an agent software
framework fully implemented in Java language. It simplifies
the implementation of the agent systems through a middleware
that comply with the FIPA specifications [15] [16].
JADE facilitates the development of multi-agent peer-topeer applications. It supports an asynchronous agent
programming model, communication between agents either on
the same or different platforms, mobility, security, and other
utilities. JADE provides ontology support and also provides
facility to develop user defined ontology that means it is
possible to define own vocabulary and semantics by user for
the content of the messages exchanged between the agents.
B. Agent Description
In this section, a brief description of the agents that
incorporate in the control system is presented. Based on the
proposed control architecture, it is necessary to define the
functions and roles of each agent accordingly. A function of
an agent is defined by a set of behaviors. Several behaviors

can be executed in parallel or concurrently by an agent. Each

agent has the autonomy to perform its goal and the behavior of
each agent is constructed based on their goals.
Micro-source Agent: It is responsible for monitoring and
controlling power levels and the status of connect/ disconnect.
Micro-sources in this study are PV, FC and DG. All of them
are controlled by corresponding agents that model and
represent them.
Load Agent: It is capable of monitoring and controlling power
level of loads and its connect/ disconnect status. Our
microgrid has separate load agents for critical and non-critical
loads.
CESS Agent: It monitors SOC level. CESS requests power to
MGC when the SOC level is low. CESS also can accept or
reject a MGC power offer depending on its SOC level.
PCC Agent: It is responsible for monitoring the grid voltage,
phase-angle and frequency, and it performs actions to isolate
or reconnect the microgrid by operating the main switch. It
also keeps informing the status changes to the MGC.
PCBB Agent: It monitors voltage, current and power level of
the PCBB. MGC will provide current/voltage reference set
point to PCBB agent, whether it needs to draw/supply power
from/to the main grid.
CB Agent: It is capable of working as a switch for energy
sources and loads; on/off control status is based on the
command sent by individual MC and LC agent. Each CB
agent interacts with the corresponding breaker in the
microgrid.
Bus Agent: It monitors voltage magnitude and phase angle and
maintains the voltage not to exceed the safety limits. Our
proposed architecture has the separate DC and AC bus agents.
RTDS Agent: It is capable of sending and receiving commands
to and from RTDS through TCP/IP.

MGC Agent: It is responsible for monitoring, scheduling and

management of the micro-sources and loads; providing power
to CESS when power supply is more than demand or CESS
SOC is low; providing power set-point to PCBB to supply or
draw power from the main grid.
C. Ontology Creation
Ontology provides a way to share common understanding
of information among agents [17]. Agents communicate by
exchanging messages and the ontology is used in the structure
of the messages. The content of the message complies with
Content Language (e.g., SL) and ontology. In this project,
ontology has been created to share common understanding of
the information between agents for real-time microgrid
operation. The structure of this ontology, called the
MicroGridOntology is shown in Fig.2. In fact, in the JADE
ontology can be composed of three types of elements:
predicates, concepts and agent actions.
Concepts are the expressions that indicate entities with a
complex structure that can be defined in terms of role or
property. Concepts represent the agents that relevant to the
microgrid operation and forming the microgrid such as PV,
FC, DG, CESS, PCBB, Load, PCC and Switches. As depicted
in Fig.2, each concept can consist of one or more subconcepts. For example, three sub-concepts of Load are
Loadnumber, Voltagelevel and Powerlevel.
Predicates are expressions that say something about the status
of the concepts and can be true or false. Our Micro Grid
Ontology consists of six predicates; Islanded, GridConnected,
RequestPower, InformPower, LoadShed and LoadConnect.
The predicates, LoadShed and LoadConnect, are used by the
MGCAgent to inform the non-critical load agent to connect or
disconnect the load, while the predicate RequestPower
informs the micro-source agents and load agent to update the
current available power.
Agent actions indicate actions that can be performed by some
agents. Here, it consists of four agent actions; SwitchIn,
SwithOut, ProvidePower and Charging. The agent actions,
SwitchIn and SwitchOut, are used by PCC, MC and LC agents
to request a corresponding switch agent to connect or
disconnect a switch.
V. MGC ALGORITHM FOR MICROGRID OPERATION

Fig.2 Ontology for Microgrid Operation

In this section, the MGC algorithms implemented in

microgrid which can operate in grid-connected and islanded
are described. The objective of this application is to maximize
the use of renewable energy sources and minimize the use of
the diesel generator. It is also assumed that there is no limit to
the power that can be drawn from the grid. In both operation
modes, the power balance between the generation and the
consumption is controlled by MGC. To produce the optimum
power assignments, MGC gathers information of the energy
produced from the micro-sources and the load periodically.
Islanded Mode: PCC agent measures grid voltage, phase-angle
and frequency. Base on the measured information, PCC
disconnects from the grid if the frequency and voltage at the
PCC exceed the limits. The standard thresholds, according to

the IEEE Std. 929-2000, are 88% and 110% for voltage and
49.3 Hz and 50.5 Hz for frequency. PCC agent informs the
status changes to the MGC, MC and LC. MGC sends
RequestPower to MC and LC to provide the current
available power. MGC uses the PV power as much as possible
because of the non-controllable nature and clean energy. MGC
performs power calculation and will request to turn on DG
when it is necessary. MGC informs the LC to shed noncritical
loads if power consumption is higher than power production.
Grid Connected Mode: Reconnection is acceptable if voltage
error is below 3%, frequency error is below 0.1Hz and phaseangle error is below 10 degree. PCC agent informs the status
changes to the MGC, MC and LC. MGC provides current
reference set point to PCBB to draw power from the main grid
and power balance between the generation and the
consumption is guaranteed by the main grid via PCBB. Next
step will be reconnection of the non-critical load to the
microgrid if the load shedding has occurred during islanded
mode.
Fig.3 illustrates the MGC algorithm for microgrid
operation in both islanded and interconnected modes. CESS is
responsible for the power balance between generation and
loads during the transition by injecting the power balance.
MGC also needs to respond for providing power to CESS
when CESS SOC is low.

Fig.4. Simulation setup for RTDS, and the control system

In case of any energy excesses, the MGC sends a message to

inform (a) CESS needs power for charging and/or (b) reduce
the output power by controlling the micro-sources in the
micro-grid. The CESS agent is capable to decide if there is a
need for charging according to their SOC.
The main objective of MGC is to obtain the desired
performances and functionality in both operating modes. In
order to achieve this, all the agents cooperate and coordinate
with the MGC. Using the defined MicroGridOntology and the
selected content language, all the agents can relatively easily
deal for all the interactions and accomplished through the
exchange of messages.
VI. SIMULATION AND RESULTS
Agent-based control system was developed and
implemented in JADE platform and analyzed it for a hybrid
microgrid system which was set up in a Real Time Digital
Simulator (RTDS). The RTDS is a special simulation tool for
real-time simulation. RTDS software includes accurate power
system component models required to represent many of the
complex elements which make up physical power systems.
RTDS also includes a graphical user interface, RSCAD
software [4].
The microgrid model including PV, FC, CESS, PCBB and
controllable loads are implemented in RSCAD software and
run in real-time. The control system in JADE and the
microgrid model in RTDS are two different pieces of software
existing in two separate environments.

Fig.3. MGC algorithm for grid-connected and islanded operation

Fig.5. Hybrid Microgrid simulation in RSCAD/RTDS

Fig.6. Real-time simulation result in grid-connected mode

Fig.7. Real-time simulation result in Islanded mode

In order to exchange messages among them, the two

platforms have been connected together via TCP/IP. For the
microgrid
model,
the
ListenOnPort
and
ListenOnPortHandshake commands are used to instruct
RTDS - RUNTIME to listen the script commands generated
from an external program. The ListenOnPort script command
provides a way for an external program to control RSCAD by
sending script commands over a TCP/IP connection. When
ListenOnPort command is executed, a server socket is
opened on assigned port and the script waits for an external
process to connect to it. However, there is no way to send
multiple commands to the RTDS at one time. To overcome
this, we implemented RTDS agent as a Client Socket and a
buffer inside our control system. All the agents inside the
control system will send requests/commands to RTDS agent
and RTDS agent consistently providing first-in-first-out
operations. Fig.4 shows the simulation setup of real-time
simulation.
To demonstrate the effectiveness of the proposed agentbased control, simulation studies have been performed in a
real-time simulation environment. The main objective of this

simulation is to verify the feasibility of the agent-based control

system for real-time microgrid and the time delay between
communications is not presented here. Fig.5 shows the
RTDS/RSCAD schematics view of 100kW hybrid microgrid
under study.
The 50-kW PV, 30-kW FC, 65-kW CESS has been
connected to DC bus. The DC bus is regulated at high voltage
800V for connecting PCBB interfacing a 415 V 3 phase
system. 100kW resistive loads (including 75% critical loads
and 25% non-critical loads) are connected to AC bus. The
power converters were represented by switching average
models and described in [3] [12] [13].
In Fig.6, the simulation is in grid-connected mode. When
PV is available, the total internal supply is 80 kW (50kW PV
+ 30kW FC) and MGC provides current reference signal to
PCBB to take 20% of power from the main grid via PCBB. To
demonstrate the capability of the control algorithm, solar
insolation value was set to zero with full load 100kW. CESS is
responding to the power balance during this period, and MGC
takes action for power generation and consumption. As can be
seen from Fig.6, PCBB and grid current changed due to MGC
request PCBB to get additional 50% of power from the main
grid by proving new current reference setting. Fig.7 presents
the result when an islanded condition is applied. The scenario
of no PV is simulated in an islanded mode. There is no change
in load and FC is the only available source inside the
microgrid hence power demand is more than power supply. It
can be seen from Fig.7, MGC requests LC to disconnect noncritical load to fulfill the power balance.
Simulation studies demonstrate that the control system can
manage the power of each energy source and load properly,
the negotiation process between agents has worked
accordingly and the microgrid works reliably. Simulation
results also show that the MGC control algorithm is valid for
any operating conditions. Simulation results prove that the
proposed control system is capable of fulfilling real-time
microgrid management and control.
VII. CONCLUSION
This paper discusses and presents an agent-based intelligent
control for real-time microgrid. The control system includes
an algorithm for the management of the microgrid operation in
both grid connected and islanded modes, functionality of
power management, load sharing, isolating microgrid and
securing critical loads during the power outage. Details
architecture and development of the control system using
JADE were described. The implemented control system was
tested in real-time microgrid with the Real Time Digital
Simulator (RTDS) in order to have a real-time control and
management as close as possible to the real physical
implementation. This can be extended with actual hardware of
CESS, PCBB and the power converters as hardware in loop
(HIL) simulation. The communication middleware was
implemented to exchange data between JADE and RTDS via
TCP/IP sockets. Simulation studies demonstrate that that the
proposed agent-based control system is effective for managing
the real-time microgrid operation. The proposed control

system has provided a flexible and scalable architecture that

allows the agents constituting a microgrid control in a
coordinated way.
VIII. REFERENCES
[1] R.Lasseter, A.Akhil, C.Marnay, J.Stephens, J.Dagle, R. Guttromson,
A.Meliopoulos, R.yinger and J.Eto, White paper on Integration of
consortium Energy Resources. The CERTS MicroGrid Concept.
CERTS, CA, Rep.LBNL-50829, Apr.2002.
[2] A.Dimeas and N.D. Hatziargyriou Operation of a Multiagent System
for Microgrid Control Power Systems, IEEE Transactions on Volume
20, Issue 3, Aug. 2005 Page(s):1447 1455 Digital Object Identifier
10.1109/TPWRS.2005.852060
[3] X. Yu and A. Khambadkone,Multi-functional Power Converter
Building Block to Facilitate the Connection of Micro-grid Control and
Modeling for Power Electronics, 11th Workshop, Aug. 2008.
[4] RTDS Technologies Inc., http://www.rtds.com/
[5] M. Wooldridge, N. R. Jennings, Intelligent agents: Theory and
Practice, The Knowledge Engineering Review, Vol. 10, Issue 2, pp
115 152, 1995
[6] Wooldridge, M., Weiss, G., Ed.: Intelligent Agents, in Multi-agent
Systems. Cambridge, MA: MIT Press, Apr. 1999, pp. 351
[7] A. Dimeas and N.D Hatziargyriou, A multi-agent system for
microgrids, In Proc. 2004 IEEE Power Engineering Society General
Meeting, Vol. 1, pp. 55-58.
[8] L. Phillips, M. Link, R. Smith and L. Weiland, Agent-based control of
distributed infrastructure resources, Sandia National Laboratories,
SAND2005-7937,Jan2006[Online].Available:
www.sandia.gov/scada/documents/sand_2005_7937.pdf.
[9] Logenthiran. T, Srinivasan. D, David Wong, "Multi-Agent Coordination
for DER in MicroGrid ", ICSET 2008, 2008.
[10] McArthur, S.D.J., Davidson, E.M., Catterson, V.M., Dimeas, A.L.,
Hatziargyriou, N.D.,Ponci, F. and Funabashi, T., Multi-Agent Systems
for Power Engineering Applications Part I: Concepts, Approaches,
and Technical Challenges, IEEE Transaction on Power Systems, Vol.
22, Issue 4, pp. 1743 1752, Nov. 2007
[11] McArthur, S.D.J., Davidson, E.M., Catterson, V.M., Dimeas, A.L.,
Hatziargyriou, N.D., Ponci, F. and Funabashi, T., Multi-Agent Systems
for Power Engineering Applications Part II: Technologies, Standards,
and Tools for Building Multi-agent Systems, IEEE Transaction on
Power Systems, Vol. 22, Issue 4, pp. 1753 - 1759, Nov. 2007
[12] Haihua Zhou, Tanmoy Bhattacharya and Ashwin M Khambadkone,
Composite Energy Storage System Using Dynamic Energy
Management in Microgrid Applications, International Power
Electronics Conference (ECCE Asia), 2010.
[13] Xiaoxiao Yu and Ashwin M Khambadkone, Control of Paralleled
Power Converter Modules to Facilitate the Efficient Operation of
Microgrid, International Power Electronics Conference, Sapporo,
Japan, June 2010.
[14] MicroGrids A Possible Future Energy Configuration, N.
Hatziargyriou, G. Strbac, IEA Workshop Distributed Generation: Key
issues, Challenges, Roles, Paris, 1 March 2004.
[15] FIPA: www.fipa.org
[16] JADE (Java Agent Development Framework) at http://jade.tilab.com/
[17] G. Caire. JADE Tutorial. Application-Defined Content Languages and
Ontologies, 2002.

IX. BIOGRAPHIES
Htay Nwe Aung obtained her BEng degree in
Electrical Engineering from Yangon Institute of
Technology, Myanmar, and M.Sc. degree in
Electrical Engineering from National University of
Singapore (NUS). She was with Seagate Technology
International, Singapore from 2001 to 2009. Since
2009, she has been with the National University of
Singapore as a Research Engineer. Her research
interests include power system operation and control,
distributed
generation,
artificial
intelligence
application in power system.

Ashwin M. Khambadkone received the Dr.Ing.

degree from the University of Wuppertal,
Wuppertal, Germany, in 1995. He joined the
National University of Singapore, Singapore, in
1998, where he is currently an Associate Professor
in the Department of Electrical and Computer
Engineering. Since April 2010 he holds the
Programme Director position at Experimental Power
Grid Center (EPGC), A*STAR concurrently with
his position at NUS. His current areas of research
are distributed-energy-resource networks, renewable
energy sources, control of power-electronics-based energy systems, and
digital control of power factor correction and multilevel inverters. From 1987
to 2001, his research was in the areas of pulsewidth modulation methods,
field-oriented control, parameter identification, and sensorless vector control.
He was with the University of Wuppertal; the University of Queensland,
Brisbane, Australia; and the Indian Institute of Science, Bangalore, India. Dr.
Khambadkone was the recipient of the Outstanding Paper Award in 1991 and
the Best Paper Award in 2002 from the IEEE TRANSACTIONS ON
INDUSTRIAL ELECTRONICS, the Prize Paper Award from the IEEE
Industry Applications Society Industrial Power Converter Committee in 2005,
and the Outstanding Educator Award in 2008 from the National University of
Singapore.
Dipti Srinivasan obtained her MEng and PhD
degrees in Electrical Engineering from the National
University of Singapore (NUS) in 1991 and 1994,
respectively. She worked at the University of
California at Berkeleys Computer Science Division
as a post-doctoral researcher from 1994 to 1995. In
June 1995, she joined the faculty of the Electrical
and Computer Engineering department at the
National University of Singapore, where she is an
associate professor. From 1998 to 1999, she was a
visiting faculty in the Department of Electrical and Computer Engineering at
the Indian Institute of Science, Bangalore, India. Her main areas of interest are
neural networks, evolutionary computation, intelligent multi-agent systems
and application of computational intelligence techniques to engineering
optimization, planning and control problems in intelligent transportation
systems and power systems. Dipti Srinivasan is a senior member of IEEE and
a member of IES, Singapore. She has published over 180 technical papers in
international refereed journals and conferences. She currently serves as an
associate editor of IEEE Transactions on Neural Networks, a social editor of
IEEE Transactions Intelligent Transportation Systems, area editor of
International Journal of Uncertainty, Fuzziness and Knowledge-based
Systems, and as a managing guest editor of Neurocomputing.
T. Logenthiran obtained his BSc degree in
Electrical and Electronic Engineering at University
of Peradeniya, Sri Lanka. He is currently pursuing
PhD degree in the department of Electrical and
Computer Engineering, National University of
Singapore (NUS), Singapore. His main areas of
interest are distributed power system and,
application of intelligent multi-agent systems and
computational intelligence techniques to power
engineering problems.