Please use this identifier to cite or link to this item:
Smart Grid Relay Protection and Network Resource Management for Real-Time Communications.
|Title:||Smart Grid Relay Protection and Network Resource Management for Real-Time Communications.|
|Contributors:||Electrical Engineering (department)|
|Date Issued:||Dec 2017|
|Publisher:||University of Hawaiʻi at Mānoa|
|Abstract:||Smart Grid (SG) signi cantly improves the computing and communication capabilities|
of power systems. As a result, smart power generation, distribution, and protection are
realized using more intelligent devices. These improvements provide many opportunities
to better address issues in existing systems, such as cascading failures in traditional grids.
Through the investigation of several large blackouts, malfunctional distance relays have
been recognized as one of the critical causes of cascading failures. To deal with the false
trips of relays and cascading failures, many monitoring and protecting methods have been
proposed on traditional grids. However, many areas are still needed to be further investigated.
In this dissertation, we focus on three issues related to SG protection. We rst
proposed an e ective resource management scheme for agent-based remote relay protection.
Second, we investigated cyber attacks on relay protection systems, analyzed their impacts,
and explored how to mitigate these attacks. Third, we proposed a synchronized routing
framework for real-time delivery of measurement and control data. We brie
these three research areas in the following.
To protect remote relays from hidden failures, several agent-based protection schemes
have been investigated. However, they do not consider the details in network resource management
and di erent failure cases. We propose a set of resource allocation schemes to address
the practical deployment issue of these protection schemes. Our results show that the
basic agent-based protection consumes signi cant amounts of network resources and cannot
e ectively support many new SG applications, and is unable to deal with communication
link failures. We further improve the proposed basic scheme, and develop backup schemes to
handle network failures. Furthermore, when utilizing power system knowledge, we can make
the resource allocation more e cient. Our evaluation shows that our proposed schemes can
use less network resources while achieving the required system reliability. Moreover, we also
investigate Peer-to-Peer (P2P) agent protection schemes. We propose an e ective resource
management scheme and a backup scheme in the P2P mode. The results show that P2P
protection can ful ll the reliability requirement using less resources.
It is well known that the reliability of power systems is a di cult challenge. To address
this problem, basic agent-based protection and agent-reputation-based protection are developed
for future smart grid to enhance the reliability of relay protection. While e ective
in many situations, these schemes may be exploited by malicious attackers to disrupt the
power system, or even trigger cascading failures in the system. Unlike occasional faulty
behaviors, cyber attacks deliberately target various components in power systems, which
cannot be resolved from the perspective of device reliability. Currently there is limited work
on how such attacks are carried out, and to what degree damage can be achieved. The goal
of our research is to better understand the problem, by examining the potential methods
to attack the relay protection system, and evaluating their e ectiveness. We also explore
how to mitigate potential cascading failure damage due to cyber attacks.
As the accurate measurement and sampling of system states are the foundation of SG
monitoring and protection applications, we develop a synchronized framework to achieve
predictable delivery for large-scale real-time applications. With the synchronized time as a
global reference, we design the synchronized routing framework and evaluate its e ectiveness
with di erent packet scheduling schemes. The results show that we can adjust service order
of packets according to their urgency. Compared with existing packet scheduling schemes,
by using both upstream and downstream delay information, the proposed method has a
larger adjustable range, and achieves much better e ciency and predictability.
In summary, we have explored remote relate protection, relay cyber security, and realtime
communication related problems in SG, and proposed several solutions, including
network resource management, analysis and defense of attacks on smart relays, and synchronized
real-time packet delivery. Our future work will focus on more detailed issues
within each problem, such as advanced agent-protection structure, re ned attack and defense
strategies, and improved measurement accuracy of network delays.
|Description:||Ph.D. Thesis. University of Hawaiʻi at Mānoa 2017.|
|Rights:||All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.|
|Appears in Collections:||
Ph.D. - Electrical Engineering|
Please email firstname.lastname@example.org if you need this content in ADA-compliant format.