RECINTRO Info about RECDEMO: A statistical procedure for reconstruction of 1D signals. The files in this directory can reproduce the figures in the article of Brodtkorb, P.A., Myrhaug, D. and Rue, H (1999) 'Joint distribution of wave height and wave crest velocity from reconstructed data.' in Proceedings of 9th ISOPE Conference, Vol III, pp 66-73 The paper shows a method for reconstruction of missing and or spurious data points of a recorded time series by using a transformed Gaussian process. The application of the result is illustrated by fitting a joint distribution to the maximum of the crest front velocity and crest back velocity (V = max(Vcf,Vcb)) and the corresponding wave height (H = Hd or Hu) obtained from the reconstructed data set. The joint distribution seem to be physically sound and much better described using the reconstructed than the original time series. (Note that the extracted data are normalized with the rms values, i.e., v=V/Vrms and h=H/Hrms) The joint probability distribution given as the product of the Rayleigh distribution of wave heights and the Weibull distribution for the conditional crest velocity (v) given wave height (h) seem to be a reasonable model. DATA: GFAKS89.DAT ~~~~~~~~~~~~~~~~~~~ The wave data considered was measured 24th December 1989 at the Gullfaks C platform in the North Sea from 17.00 to 21.20. The period from 20.00 to 20.20 is missing and contains NaNs. The water depth of 218 m is regarded as deep water for the most important wave components. There are two EMI laser sensors named 219 and 220. This data set is obtained from sensor 219, which is located in the Northwest corner approximately two platform leg diameters away from the closest leg. Thus the wave elevation is not expected to be significantly affected by diffraction effects for incoming waves in the western sector. The wind direction for this period is from the south. Some difficulties in calibration of the instruments have been reported resulting in several consecutive measured values being equal or nearly equal in the observed data set. Hm0 = 6.8m, Tm02 = 8s, Tp = 10.5 Tp2=19 Wind data from Statfjord A, 1989 are also given to justify the assumption that the data may be regarded as stationary.

- recintro
- help

Info about RECDEMO: A statistical procedure for reconstruction of 1D signals. | |

Display help text in Command Window. |

- recdemo
- recintro

Show Rec figures with point and click interface | |

Info about RECDEMO: A statistical procedure for reconstruction of 1D signals. |

001 function recintrorecintro 002 % RECINTRO Info about RECDEMO: A statistical procedure for reconstruction of 1D signals. 003 % 004 % The files in this directory can reproduce the figures in the 005 % article of 006 % 007 % Brodtkorb, P.A., Myrhaug, D. and Rue, H (1999) 008 % 'Joint distribution of wave height and wave crest velocity 009 % from reconstructed data.' 010 % in Proceedings of 9th ISOPE Conference, Vol III, pp 66-73 011 % 012 % 013 % The paper shows a method for reconstruction of missing and or spurious 014 % data points of a recorded time series by using a transformed Gaussian 015 % process. The application of the result is illustrated by fitting a 016 % joint distribution to the maximum of the crest front velocity and crest back velocity 017 % (V = max(Vcf,Vcb)) and the corresponding wave height (H = Hd or Hu) 018 % obtained from the reconstructed data set. The joint distribution seem to be physically 019 % sound and much better described using the reconstructed than the 020 % original time series. (Note that the extracted data are normalized with 021 % the rms values, i.e., v=V/Vrms and h=H/Hrms) 022 % The joint probability distribution given as the product of the 023 % Rayleigh distribution of wave heights and the Weibull distribution 024 % for the conditional crest velocity (v) given wave height (h) seem to be 025 % a reasonable model. 026 % 027 % DATA: GFAKS89.DAT 028 % ~~~~~~~~~~~~~~~~~~~ 029 % The wave data considered was measured 24th December 1989 at the Gullfaks C platform 030 % in the North Sea from 17.00 to 21.20. The period from 20.00 to 20.20 031 % is missing and contains NaNs. The water depth of 218 m is 032 % regarded as deep water for the most important wave components. 033 % There are two EMI laser sensors named 219 and 220. This data set is 034 % obtained from sensor 219, which is located in the Northwest 035 % corner approximately two platform leg diameters away from 036 % the closest leg. 037 % Thus the wave elevation is not expected to be significantly 038 % affected by diffraction effects for incoming waves in the western sector. 039 % The wind direction for this period is from the south. 040 % Some difficulties in calibration of the instruments have been reported 041 % resulting in several consecutive measured values being equal or nearly equal 042 % in the observed data set. 043 % 044 % Hm0 = 6.8m, Tm02 = 8s, Tp = 10.5 Tp2=19 045 % 046 % Wind data from Statfjord A, 1989 are also given to justify the 047 % assumption that the data may be regarded as stationary. 048 049 050 051 052 % By pab 28.01.2000 053 054 %more on 055 help recintro 056 %more off 057

Mathematical Statistics

Centre for Mathematical Sciences

Lund University with Lund Institute of Technology

Comments or corrections to the WAFO group

Generated on Thu 06-Oct-2005 02:21:16 for