University of Khartoum

A Selection Hyper-heuristic for Wind Farm Layout Optimisation

A Selection Hyper-heuristic for Wind Farm Layout Optimisation

Show full item record

Title: A Selection Hyper-heuristic for Wind Farm Layout Optimisation
Author: Daffalla, Alaa Ahmed Yousif
Abstract: Wind energy is a promising source of renewable energy. The past decade has witnessed a tremendous increase in wind farm installations globally and it's still envisioned to increase in the coming years. Hence, e cient wind farm layout - that which is subject to various nancial and engineering objectives and constraints - is needed to ensure the sustainability and continuity of wind energy production. The wind farm layout optimisation problem is considered a highly complex NP-hard problem, for which exact methods are unsuitable. There is a wide and varied literature on the use of evolutionary algorithms for the optimisation of wind farm layouts. These algorithms have proven to be very e ective at nding near-optimal solutions to a large number of problems in the energy industry. However, a new genre of optimisation algorithms known as hyper-heuristics is gaining the interest of researchers and is being applied to such problems. Hyperheuristics are automated search methodologies or learning mechanisms for selecting or generating heuristics to solve multiple computationally di cult optimisation problems. They combine simple heuristics to create bespoke algorithms for speci c problem domains, and have proven successful on other optimisation problems. This work examines the use of a selection hyper-heuristic when applied to the wind farm layout optimisation problem. A selection hyper-heuristic mixes and controls a prede ned set of low level heuristics with the goal of improving an initially generated solution by choosing and applying an appropriate heuristic to a solution in hand and deciding whether to accept or reject the new solution at each step under an iterative framework. Current preliminary results clearly indicate that selection hyper-heuristics could possibly outperform conventional evolutionary approaches such as GA in terms of solution quality, and run-time. Further comparisons with previous work, show that there are many more opportunities in applying varied combinations of selection hyper-heristics to WFLOP.
URI: http://khartoumspace.uofk.edu/123456789/25818
Date: 2017-10


Files in this item

Files Size Format View

This item appears in the following Collection(s)

Show full item record

Share

Search DSpace


Browse

My Account