A novel Geno-fuzzy based model for hydrodynamic efficiency prediction of a land-fixed oscillating water column for various front wall openings, power take-off dampings and incident wave steepnesses

Altunkaynak A., Celik A.

RENEWABLE ENERGY, vol.196, pp.99-110, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 196
  • Publication Date: 2022
  • Doi Number: 10.1016/j.renene.2022.06.045
  • Journal Name: RENEWABLE ENERGY
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Aerospace Database, Aquatic Science & Fisheries Abstracts (ASFA), CAB Abstracts, Communication Abstracts, Compendex, Environment Index, Geobase, Greenfile, Index Islamicus, INSPEC, Pollution Abstracts, Public Affairs Index, Veterinary Science Database, DIALNET, Civil Engineering Abstracts
  • Page Numbers: pp.99-110
  • Keywords: Fuzzy logic, Hydrodynamic efficiency, Genetic algorithms, Oscillating water column, Wave energy, Physical experimental model, ENERGY CONVERTERS, LOGIC, SYSTEMS, PERFORMANCE, RAINFALL, HEIGHT
  • Istanbul Technical University Affiliated: Yes


Accurate efficiency estimation of a wave energy converter (WEC) is a key concept in the design stage. Oscillating water column (OWC) is a promising type of WEC due to its advantages such as proved concept, operational simplicity, accessibility, reliability etc. In this study, for accurate efficiency estima-tion of an OWC, a novel Geno-fuzzy based model (GENOFIS) was developed, firstly, by improving Adaptive Neuro-Fuzzy inference system (ANFIS) and secondly, incorporating the Genetic algorithms (GAs). Data for training and testing phases of the models were obtained from an extensive wave flume experiments for various OWC underwater opening heights and applied PTO dampings under different incident waves. Both models performed remarkably with a slightly better performance of GENOFIS. The superiority of the GENOFIS model stemmed from that its high performance was attained with sub-stantially low fuzzy rules (only three) where ANFIS required incomparably large fuzzy rules (twenty-seven) and yet achieved a slightly lower performance. Accordingly, very few numbers of fuzzy rules enable the construction of GENOFIS model structure with low complexity, which, in turn, immensely reduce the computational time required. Developed less complicated GENOFIS model is parsimonious, unlikely to suffer from overfitting and has high interpretability and practicality.(c) 2022 Elsevier Ltd. All rights reserved.