Bass, J. H. and Rebbeck, M. and Landberg, L. and Cabre, M. and Hunter, Andrew (2000) An improved measure-correlate-predict algorithm for the prediction of the long term wind climate in regions of complex environment. Project Report. European Commission. (Unpublished)

	PDF An_improved_measure-correlate-predict_algorithm_for_the_prediction_of_the_long_term_wind_climate_in_regions_of_complex_environment-Part_1.pdf - Whole Document Restricted to Repository staff only Download (3787Kb) | Request a copy
	PDF An_improved_measure-correlate-predict_algorithm_for_the_prediction_of_the_long_term_wind_climate_in_regions_of_complex_environment-Part_2.pdf - Whole Document Restricted to Repository staff only Download (3812Kb) | Request a copy
	PDF An_improved_measure-correlate-predict_algorithm_for_the_prediction_of_the_long_term_wind_climate_in_regions_of_complex_environment-Part_3.pdf - Whole Document Restricted to Repository staff only Download (4Mb) | Request a copy

Abstract

This final report contains a complete description of the work undertaken in fulfilment of contract number JOR3-CT98-0295. The Project comparises three partners (Renewable Energy Systems Limited, Riso National Laboratory and Ecotecnia) and one sub-contractor (University of Sunderland). Renewable Energy Systems Ltd is the Project Co-ordinator. Teh aim of the project is to investigate the application of neural network techniques to the assessment of wind climate at potential wind farm sites. The current state-of-the-art for such assessment involves 'Measure-Correlate-Predict' (MCP) methods, which are statistical in nature. One of the limiations of MCP methods is that in regions of complex terrain or complex climatology, large prediction errors can result. It is anticipated that he use of neural networks, which are very good at identifying patterns in noisy data, should significantly improve predictions of the wind climatology of a site, so reducing uncertainty in the available energy yield and, in turn, the risk of financial investment in a wind farm.

Item Type:	Monograph (Project Report)
Additional Information:	This final report contains a complete description of the work undertaken in fulfilment of contract number JOR3-CT98-0295. The Project comparises three partners (Renewable Energy Systems Limited, Riso National Laboratory and Ecotecnia) and one sub-contractor (University of Sunderland). Renewable Energy Systems Ltd is the Project Co-ordinator. Teh aim of the project is to investigate the application of neural network techniques to the assessment of wind climate at potential wind farm sites. The current state-of-the-art for such assessment involves 'Measure-Correlate-Predict' (MCP) methods, which are statistical in nature. One of the limiations of MCP methods is that in regions of complex terrain or complex climatology, large prediction errors can result. It is anticipated that he use of neural networks, which are very good at identifying patterns in noisy data, should significantly improve predictions of the wind climatology of a site, so reducing uncertainty in the available energy yield and, in turn, the risk of financial investment in a wind farm.
Keywords:	Neural networks, Meausre correlate predict, long term wind climate, complex environment
Subjects:	G Mathematical and Computer Sciences > G400 Computer Science
Divisions:	College of Sciences > Faculty of Science > Lincoln School of Computer Science
Depositing User:	Tammie Farley
Date Deposited:	26 Sep 2010 16:03
Last Modified:	13 Mar 2013 08:47
URI:	http://eprints.lincoln.ac.uk/id/eprint/3389

Actions (login required)

View Item