Spectral analysis of sea state wave and influence of autocovariance and smoothing parameter

This paper presents the results of the sensitivity test performed on the spectral plot of wave data on the Malaysian water related to the oil and gas industries. The presence of white noise in spectral analysis produces unwanted spectral peak which may lead to incorrect interpretation of the result. For oil and gas development, the impact is immense as this may lead to improper design criteria and may impede offshore facilities activities. This jeopardizes the huge capital investment. The objectives of the study are to determine the appropriate number of autocorrelation lag as well as the number of the smoothing function´s span for the spectral analysis and to plot the wave spectrum corresponding to the number of autocorrelation lag and the number of the smoothing function´s span. As a first step of the study, the sensitivity test was carried out to determine the appropriate number of autocorrelation. Subsequently, the smoothing test was performed to obtain smoothing function´s span number. From the study, it was found that the suitable number of autocorrelation lag and the number of smoothing function´s span to be used in the spectral analysis are 400 and 37 respectively. The plotted spectrums at these values appear to be appropriate and representative in shape.