Show simple item record  

dc.contributor.authorWilson, Bretten_NZ
dc.contributor.authorWakes, Sarahen_NZ
dc.contributor.authorMayo, Michaelen_NZ
dc.coverage.spatialHonolulu, Hawaiien_NZ
dc.date.accessioned2018-04-17T23:34:45Z
dc.date.available2017en_NZ
dc.date.available2018-04-17T23:34:45Z
dc.date.issued2017en_NZ
dc.identifier.citationWilson, B., Wakes, S., & Mayo, M. (2017). Surrogate modeling a computational fluid dynamics-based wind turbine wake simulation using machine learning. In Proceedings of 2017 IEEE Symposium Series on Computational Intelligence, 27 November -1 December 2017, Honolulu, HI, USA (pp. 1–8). Washington, DC, USA: IEEE. https://doi.org/10.1109/SSCI.2017.8280844en
dc.identifier.isbn9781538627259en_NZ
dc.identifier.urihttps://hdl.handle.net/10289/11785
dc.description.abstractThe Wind Farm Layout Optimization problem involves finding the optimal positions for wind turbines on a wind farm site. Current Metahueristic based methods make use of a combination of turbine specifications and parameters, mathematical models and empirically produced power production equations to estimate the energy output of a real wind farm [15]. The overarching variable in any optimisation function is wind speed - this is what used to determine the power generated. Therefore, accurate predictions of wind speeds at specific points across the volume of the site are needed. In this paper, Computational Fluid Dynamics (CFD) was used to simulate a full scale rotating wind turbine blade with fluid (air) at various wind speeds flowing past the turbine. The wake effect can be observed and leads to decrease in wind speeds, as expected. Wind speed at specific x,y and z (3D) coordinates were sampled and used as input to common Machine Learning regression algorithms to create different surrogate models. This was needed as each individual CFD experiment takes approximately 8 hours to complete, so it is not feasible to continuously repeat these simulations inside a metaheuristic optimiser.
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherIEEEen_NZ
dc.rightsThis is an author’s accepted version of an article published in the proceedings of 2017 IEEE Symposium Series on Computational Intelligence. © 2017 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
dc.sourceSSCIen_NZ
dc.subjectcomputer scienceen_NZ
dc.subjectsurrogate modelen_NZ
dc.subjectwind turbineen_NZ
dc.subjectwind farm layout optimisation problemen_NZ
dc.subjectMachine learningen_NZ
dc.subjectcomputational fluid dynamicsen_NZ
dc.titleSurrogate modeling a computational fluid dynamics-based wind turbine wake simulation using machine learningen_NZ
dc.typeConference Contribution
dc.identifier.doi10.1109/SSCI.2017.8280844en_NZ
dc.relation.isPartOfProceedings of 2017 IEEE Symposium Series on Computational Intelligenceen_NZ
pubs.begin-page1
pubs.elements-id221279
pubs.end-page8
pubs.finish-date2017-12-01en_NZ
pubs.place-of-publicationWashington, DC, USA
pubs.start-date2017-11-27en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record