TY - GEN
T1 - Robust Wave Resource Estimation
T2 - The 23rd International Ocean and Polar Engineering Conference
AU - Lavelle, John
AU - Kofoed, Jens Peter
PY - 2013
Y1 - 2013
N2 - An assessment of the wave energy resource at the location of the Danish Wave Energy test Centre (DanWEC) is presented in this paper. The Wave Energy Converter (WEC) test centre is located at Hanstholm in the of North West Denmark. Information about the long term wave statistics of the resource is necessary for WEC developers, both to optimise the WEC for the site, and to estimate its average yearly power production using a power matrix. The wave height and wave period sea states parameters are commonly characterized with a bivariate histogram. This paper presents bivariate histograms and kernel density estimates of the PDF as a function both of Hm0 and Tp, and Hm0 and T0;2, together with the mean wave power per unit crest length, Pw, as a function of Hm0 and T0;2. The wave elevation parameters, from which the wave parameters are calculated, are filtered to correct or remove spurious data. An overview is given of the methods used to do this, and a method for identifying outliers of the wave elevation data, based on the joint distribution of wave elevations and accelerations, is presented. The limitations of using a JONSWAP spectrum to model the measured wave spectra as a function of Hm0 and T0;2 or Hm0 and Tp for the Hanstholm site data are demonstrated. As an alternative, the non-parametric loess method, which does not rely on any assumptions about the shape of the wave elevation spectra, is used to accurately estimate Pw as a function of Hm0 and T0;2.
AB - An assessment of the wave energy resource at the location of the Danish Wave Energy test Centre (DanWEC) is presented in this paper. The Wave Energy Converter (WEC) test centre is located at Hanstholm in the of North West Denmark. Information about the long term wave statistics of the resource is necessary for WEC developers, both to optimise the WEC for the site, and to estimate its average yearly power production using a power matrix. The wave height and wave period sea states parameters are commonly characterized with a bivariate histogram. This paper presents bivariate histograms and kernel density estimates of the PDF as a function both of Hm0 and Tp, and Hm0 and T0;2, together with the mean wave power per unit crest length, Pw, as a function of Hm0 and T0;2. The wave elevation parameters, from which the wave parameters are calculated, are filtered to correct or remove spurious data. An overview is given of the methods used to do this, and a method for identifying outliers of the wave elevation data, based on the joint distribution of wave elevations and accelerations, is presented. The limitations of using a JONSWAP spectrum to model the measured wave spectra as a function of Hm0 and T0;2 or Hm0 and Tp for the Hanstholm site data are demonstrated. As an alternative, the non-parametric loess method, which does not rely on any assumptions about the shape of the wave elevation spectra, is used to accurately estimate Pw as a function of Hm0 and T0;2.
KW - Wave Resource Assessment
KW - Resource Characterisation
KW - Wave Spectra
KW - Wave Parameters
KW - Wave Quality Control
KW - Wave Resource Assessment
KW - Resource Characterisation
KW - Wave Spectra
KW - Wave Parameters
KW - Wave Data Quality Control
M3 - Article in proceeding
VL - 1
T3 - Proceedings of the International Offshore and Polar Engineering Conference
SP - 495
EP - 499
BT - The Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference
PB - International Society of Offshore & Polar Engineers
Y2 - 30 June 2013 through 5 July 2013
ER -