Download files
Abstract
The vision of a smart grid is to provide a modern, resilient and secure electric power
grid in a highly reliable, stable and efficient environment through effective use of its
information and communication technology (ICT). However, a number of technical
and economical challenges have to be addressed for successfully integrating ICT with
the current grids. The deployment of an intelligent decision support application, that
is, a distributed multi-agent system (MAS) can be an effective solution to mitigate
the constraint related to stability and security of future smart grids.
The first contribution of this thesis is the development of a distributed agentbased
framework for detecting and isolating faults in power systems, since they are
susceptible to large disturbances. To act effectively against faults, the agents cooperate
and communicate with each other, exchange information and make autonomous
decisions regarding their control of circuit breakers (CBs) in order to efficiently
maintain the reliability of power systems.
Based on the idea of fault detection and isolation, a unique contribution to
transient stability enhancement using multi-agent approach is achieved by proper
relay coordination to operate with the critical clearing time (CCT) information.
A hybrid approach, combination of direct and time-domain simulation methods,
is used for CCT calculation. The agents dynamically adapt online measurements
and use the CCT information for relay coordination to improve the online transient
stability. The performances of the proposed approach fulfil the criteria for robust
stabilisation and produce adequate stability margins. This effectiveness is validated
by detailed simulations which demonstrate that transient stability can be better
improved by using agents rather than conventional approaches.
As the relationship between the amount of integrated renewable energy sources
(RESs) and volatge variations is a prominent and crucial issue, another important
contribution of this thesis is to develop a new agent-based reactive power management
scheme in order to enhance the dynamic voltage stability. In this scheme,
agents properly estimate the desired amount of reactive power from distribution
networks and take into account the changing nature and dynamic behaviour of wind
generators. Also, a distributed agent-based robust control methodology for distribution
static synchronous compensator (DSTATCOM) is designed which offers several
benefits compared to conventional approach under various operating conditions.
Since smart grids are vulnerable to various cyber attacks and noises, the final
contribution of this research is to analyse the potential impacts of these events
on different domains of smart grids. Cyber attacks in the protection system may
cause false tripping of CBs at an undesirable time, even if there is no fault and also
affect the phasor measurement unit (PMU) by tempering with its information. The
distributed MAS scheme can enhance the security of smart grids against possible
cyber attacks. The simulations undertaken demonstrating that the intelligent agents
have the potential to work together to counteract the undesirable events and return
the system to reliable and stable operation by taking appropriate remedial actions.