Smart grid evolutionary planning and modelling future power networks
| dc.contributor.advisor | Savill, Mark A. | |
| dc.contributor.advisor | Kipouros, Timoleon | |
| dc.contributor.author | Nieto Martin, Jesus | |
| dc.date.accessioned | 2020-03-31T09:03:24Z | |
| dc.date.available | 2020-03-31T09:03:24Z | |
| dc.date.issued | 2017-02 | |
| dc.description.abstract | The volatility and associated uncertainties of a rapid evolving power sector, along with the digitalisation of the sector, have triggered the necessity of answering questions faster while dealing with more granularity in data. The primary hypothesis underlying this research is that evolutionary meta-heuristics methods could be used to provide planners exploration capabilities of trade off in system when conflicting objectives appear. The aim of this research is to apply a set of novel evolutionary techniques to make better informed decisions that are capable of a) develop detailed quantitative representation of real-world power systems suitable for being optimised, b) fit existing meta heuristics evolutionary techniques to real-world size problems, c) evaluating non-traditional system flexibility services, d) validate, visualise, and evaluate performance metrics for power systems optimisation. Dynamic optimisation encompasses the important challenge in real-world applications of capturing evolving behaviours of complex systems. The literature review identifies key problems in the sector for evolving pathways to a low-carbon 2050. Issues on power networks relate to the reactive nature of in tervention planning, which leads to horizoning and locally optimal solutions. In that context, and as interventions are triggered by network failures, locational case studies are presented in this research. Applying a bespoke Graph search algorithm (A*) and Multi-Objective Evolutionary Algorithms (MOEAs) can be say that where the first evaluates just one solution at a time, MOEAs are a better approach for global optimisation due to its capability of developing multiple alternative solutions to a problem simultaneously. Historically, electricity distribution networks have been designed to provide reliable connections to the customers by virtue of asset ratings sufficient to cope with peak demand. With the proliferation of low carbon technologies such as electric vehicles, heat pumps and distributed generation, the network is starting to experience congestion both, load and generation driven. The congestion restricts further deployment of distributed energy generation, making it more difficult to meet the emission reduction targets. This motivated three case studies contained in this thesis: a large power system case study modelling the Independent System Operator of New England in the US with high wind penetration and storage; A 11kV distribution network for investment planning in the UK evaluating smart grid interventions, and finally, a non-traditional flexibility service propositions evaluation using Real Options for Multi-Utility dynamic investments. | en_UK |
| dc.identifier.uri | http://dspace.lib.cranfield.ac.uk/handle/1826/15344 | |
| dc.language.iso | en | en_UK |
| dc.rights | © Cranfield University, 2015. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder. | |
| dc.subject | Power distribution | en_UK |
| dc.subject | Evolutionary optimisation | en_UK |
| dc.subject | Strategic planning | en_UK |
| dc.subject | Meta-heuristics | en_UK |
| dc.title | Smart grid evolutionary planning and modelling future power networks | en_UK |
| dc.type | Thesis | en_UK |