Wavelet Transform as a Filtering Technique to Ship Motion Time Series

Abstract

Noise is a problem inherent to any measurement and may jeopardize the achievement of test objectives. In the design of a ship hull, noise in the processed data can lead to under or over estimation of certain parameters, such as acceleration. This can create economic loss and risk to human lives. For vertical acceleration measurements, the application of low-pass filters is not recommended due to the high frequency nature of relevant events. The present work proposes the application of wavelet transforms as a filtering technique for vertical acceleration time series. An influence investigation of wavelet parameters is performed along with a parameter determination test using measured data from high-speed hull towing tank tests conducted at Davidson Laboratory. Commercially available software is used for the data processing and determination of design parameters, such as third-highest and tenth-highest acceleration.