Research
Research Areas:
(1). Computational Intelligence Techniques for Modeling and Control of Smart Grid
Original
research in the field of computational intelligence techniques for modeling and control of distributed energy systems has
been carried out. This research intends to create a framework for the
introduction of distributed power generation in the context of smart grid.
We investigate system security, reliability and planning issues pertaining
to embedded distributed generators, from which the local utilities and
interested parties can benefit.
(2) Dynamic optimization and management
of micro grid
This
project aims to develop intelligent computational tools for efficiency
optimization and management of smart micro grids, and conduct hardware in
the loop testing. In this work, we design, develop, and deploy
innovative technologies for the next generation of smart grids to provide
solutions to the related problems to make the grid more adaptable,
flexible, intelligent and robust.
(3). Intelligent Mobility Modeling and Management
In order to support a wide range of urban transportation options, mobility
management becomes a crucial factor when designing infrastructure for urban
transport networks. Due to the anticipated increase in the number of
vehicles in the future, the networks should be able to support a huge
number of users and their individual requirements while at the same time
optimizing the resources. This research in intelligent mobility management
system aims to develop new decision making technologies in the field of
hybrid intelligent systems which aid the vehicle scheduling, intelligent congestion-aware
routing and energy management. A new architecture that applies hybrid
computational intelligence concepts to implement a cooperative multi-agent
system for real-time traffic signal control of a complex traffic network
has also been developed. The large-scale traffic signal control problem is
divided into various sub-problems and modularized, hierarchical agent
architecture have been used. Coordinated control
is achieved by intelligent agents, which adopt the cooperative distributed
problem solving approach.
(4). Evolutionary Computation for Optimization and
Scheduling
Evolutionary
algorithms, hybridized with heuristic approaches are being developed for
solving real world problems with large number of objectives in constrained
environments. Various types of operators and control parameter settings
that are likely to produce the best results are also investigated.
(5). Multi-agent Systems
This
research involves design of multi-agent systems that deliver high levels of
cooperation and reliability in dynamic and rapidly evolving environments.
These systems are being developed and tested on two large test beds: the
first for traffic network systems and the
other for distributed energy systems.
|