Evaluation of the accuracy of soft computing learning algorithms in performance prediction of tidal turbine

Shahab S. Band, Pezhman Taherei Ghazvinei, Khamaruzaman bin Wan Yusof, Mohammad Hossein Ahmadi, Narjes Nabipour, Kwok Wing Chau

Research output: Journal article publicationJournal articleAcademic researchpeer-review

Abstract

Marine renewable energy has made significant progress in the last few decades. Even after making substantial progress, the cost of electricity produced by tidal turbines is high. Therefore, the current paper concentrated on reducing the cost of transportation and installation of the turbine by performing a model. Extreme Learning Machine and Support Vector Machines as well as Genetic Programming were applied to predict the performance of the turbine model by creating short-term, multistep-ahead prediction models to compute the performance of the H-rotor vertical axis Folding Tidal turbine. The performance of the turbine was verified by a numerical study using the three-dimensional approach for the viscous model with the unsteady flow. Statistical evaluation of the outcomes pointed out that advanced Extreme Learning Machine simulation made the assurance in formulating an innovative forecasting strategy for investigating the performances of the tidal turbine. This study shows that the application of the new procedure resulted in confident generality performance and learns faster than orthodox learning algorithms. In conclusion, the assessment indicated that the advanced Extreme Learning Machine simulation was capable as a promising alternative to existing numerical methods for computing the coefficient of performance for turbines.

Original languageEnglish
JournalEnergy Science and Engineering
DOIs
Publication statusAccepted/In press - 2020

Keywords

  • co-efficient of performance
  • extreme learning machine
  • folding tidal turbine
  • genetic programming
  • support vector machines
  • tidal current turbine

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Energy(all)

Cite this