The University of Southampton
University of Southampton Institutional Repository

Decentralised coordination of smart distribution networks using message passing

Decentralised coordination of smart distribution networks using message passing
Decentralised coordination of smart distribution networks using message passing
Over the coming years, distribution network operators (DNOs) face the challenge of incorporating an increased number of electrical distributed generators (DGs) into their already capacity-constrained distribution networks. To overcome this challenge will require the DNOs to use active network management techniques, which are already prevalent in the transmission network, in order to constantly monitor and coordinate these generators, whilst ensuring that the bidirectional flows they engender on the network are safe. Therefore, this thesis presents novel decentralised message passing algorithms that coordinate generators in acyclic electricity distribution networks, such that the costs (in terms of carbon dioxide (CO2) emissions) of the entire network are minimised; a technique commonly referred to as optimal dispatch. In more detail, we cast the optimal dispatch problem as a decentralised agent-based coordination problem and formalise it as a distributed constraint optimisation problem (DCOP). We show how this DCOP can be decomposed as a factor graph and solved in a decentralised manner using algorithms based on the generalised distributive law; in particular the max-sum algorithm. We go on to show that max-sum applied naively in this setting performs a large number of redundant computations. To address this issue, we present both a discrete and a continuous novel decentralised message passing algorithm that outperforms max-sum by pruning much of the search space. Our discrete version is applicable to network settings that are entirely composed of discrete generators (such as wind turbines or solar panels), and when the constraints of the electricity network have been discretised. Our continuous version can be applied to a wider range of network settings containing multiple types of generators, without the need to discretise the electricity distribution network constraints. We empirically evaluate our algorithms, using two large real electricity distribution network topologies, and show that they outperform max-sum (in terms of computational time and total size of messages sent)
Miller, Sam
d1210662-75ea-4009-9500-7dd8173f6aee
Miller, Sam
d1210662-75ea-4009-9500-7dd8173f6aee
Ramchurn, Sarvapali
1d62ae2a-a498-444e-912d-a6082d3aaea3
Rogers, Alex
f9130bc6-da32-474e-9fab-6c6cb8077fdc

Miller, Sam (2014) Decentralised coordination of smart distribution networks using message passing. University of Southampton, Physical Sciences and Engineering, Doctoral Thesis, 124pp.

Record type: Thesis (Doctoral)

Abstract

Over the coming years, distribution network operators (DNOs) face the challenge of incorporating an increased number of electrical distributed generators (DGs) into their already capacity-constrained distribution networks. To overcome this challenge will require the DNOs to use active network management techniques, which are already prevalent in the transmission network, in order to constantly monitor and coordinate these generators, whilst ensuring that the bidirectional flows they engender on the network are safe. Therefore, this thesis presents novel decentralised message passing algorithms that coordinate generators in acyclic electricity distribution networks, such that the costs (in terms of carbon dioxide (CO2) emissions) of the entire network are minimised; a technique commonly referred to as optimal dispatch. In more detail, we cast the optimal dispatch problem as a decentralised agent-based coordination problem and formalise it as a distributed constraint optimisation problem (DCOP). We show how this DCOP can be decomposed as a factor graph and solved in a decentralised manner using algorithms based on the generalised distributive law; in particular the max-sum algorithm. We go on to show that max-sum applied naively in this setting performs a large number of redundant computations. To address this issue, we present both a discrete and a continuous novel decentralised message passing algorithm that outperforms max-sum by pruning much of the search space. Our discrete version is applicable to network settings that are entirely composed of discrete generators (such as wind turbines or solar panels), and when the constraints of the electricity network have been discretised. Our continuous version can be applied to a wider range of network settings containing multiple types of generators, without the need to discretise the electricity distribution network constraints. We empirically evaluate our algorithms, using two large real electricity distribution network topologies, and show that they outperform max-sum (in terms of computational time and total size of messages sent)

Text
thesisFINALHardBound.pdf - Other
Download (1MB)

More information

Published date: February 2014
Organisations: University of Southampton, Agents, Interactions & Complexity

Identifiers

Local EPrints ID: 362616
URI: http://eprints.soton.ac.uk/id/eprint/362616
PURE UUID: e9da0d12-cab8-47d8-8bbd-210c769d3f07
ORCID for Sarvapali Ramchurn: ORCID iD orcid.org/0000-0001-9686-4302

Catalogue record

Date deposited: 28 Feb 2014 16:35
Last modified: 15 Mar 2024 03:22

Export record

Contributors

Author: Sam Miller
Thesis advisor: Sarvapali Ramchurn ORCID iD
Thesis advisor: Alex Rogers

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×