linearly superimposed to determine the generator blade motion. Figure 7 shows 

 a sample of observed and predicted wave spectra. 



28.00 



23.33 



18,67 



'^ 14.00 - 



9.33 - 



4.67 - 



Predicted Wave Spectrum 



0.25 



Observed Wave Spectrum 



0.50 0.75 1.00 1.25 

 Frequency ( Hz ) 



1.50 1.75 2.00 



Figure 7. Observed and predicted wave spectra (sample of 

 Pierson-Moskowitz spectra) . 



In all tests the water surface time history was recorded at each wave gage 

 for 4,096 data points. A sampling rate of 16 times per second was used for 

 waves with periods less than 3 seconds, and a rate of 8 times per second was 

 used for waves with periods greater than 3 seconds. The data were digitized 

 and recorded on magnetic tape through the use of a Data Acquisition System 

 (DAS) . An analysis of the tapes was performed later on a general-purpose 

 computer. 



Two distinct types of data analysis were performed on the digitized wave 

 records: a fast Fourier transform (FFT) and a zero up-crossing analysis. The 

 FFT was used to determine the spectrum of the wave record, such as shown in 

 Figure 7, which in turn was. used to deterraine the period of peak energy 

 density. The period of peak energy density, Tp, is the reciprocal of the 

 frequency of the midpoint of the 11 consecutive spectral lines with the most 

 wave energy. The zero up-crossing method of analysis is used to determine the 

 wave heights, crest heights, wave periods, and crest durations of individual 

 waves in irregular wave trains or average values for monochromatic wave trains. 



Crest height. 



is measured relative to the mean water level (MWL) and 



