Application of Soft Computing Tools for Wave Prediction at Specific Locations in the Arabian Sea Using Moored Buoy Observations

Abstract

The knowledge of design and operational values of significant wave heights is perhaps the single most important input needed in ocean engineering studies. Conventionally such information is obtained using classical statistical analysis and stochastic methods. As the causative variables are innumerable and underlying physics is too complicated, the results obtained from the numerical models may not always be very satisfactory. Soft computing tools like Artificial Neural Network (ANN) and Adaptive Network based Fuzzy Inference System (ANFIS) may therefore be useful to predict significant wave heights in some situations. The study is aimed at forecasting of significant wave height values in real time over a period of 24hrs at certain locations in Indian seas using the models of ANN and ANFIS. The data for the work were collected by National Institute of Ocean Technology, Chennai. It was found that the predictions of wave heights can be done by both methods with equal efficiency and satisfaction.