Neural Network Modeling for Predicting Electric Power Generation From Sea Waves , , ,

Abstract

Thispaper presents a development of a prototype 1 kW electricpower generation system from sea waves in the gulf of Thailand. The system consists of electrical generator and buoy, and installed on a steel frame. This system is far from shore about 200 meters and sea level 3 meters deep. The purpose of this study is to investigate the relationship between electric power and wave height including the other parameters in the system. The neural networks are used to train the wave height and electric power and consequently the electric power from sea wave can be predicted. The predicted values from neural network models and polynomial regressionwill be compared. The predicted values from neural network are more accuracy in linking the inputs to the outputs in a non-linear manner. The prototype power units have been performed at the Sirindhorn international environment park to verify the ability of neural network models.