Optimal Feature Identification for Machine Prediction of Wind-Wave Parameters at Wave Energy Converter Site

• Main Authors: Umair, M., Hashmani, M.A., Keiichi, H.
• Format: Conference or Workshop Item
• Published: Institute of Electrical and Electronics Engineers Inc. 2020
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85097560138&doi=10.1109%2fICCI51257.2020.9247677&partnerID=40&md5=95f609b6ce635ad28b1de7e44d093fe2; http://eprints.utp.edu.my/29884/

The hike in fossil-fuel prices and increasing environmental damage due to the subsequent Carbon Monoxide (CO) emission from burning fossil-fuel is becoming a major concern for every nation. The possibility of generating power from natural sources such as solar, wind, and sea waves is thus considered as an alternative. In the case of the sea waves, the kinetic energy of surface waves can be converted into single direction motion which runs a turbine to generate electricity. A Wave Energy Converter (WEC) is such an installation that converts the wave energy into electrical energy. In this study, we have conducted a literature investigation to identify the significant meteorological and wind-wave data parameters which determine wave-energy potential at a wave energy converter site and then identified optimal feature sets from buoy data for machine prediction of those identified parameters. The authors hope that by suggesting optimal feature sets, the outcomes of this study will help in improving the computational efficiency of machine learning models specially designed for wave parameter prediction at WEC sites. © 2020 IEEE.