Journal of Energy Engineering · 2020 · 10 citations · 21 references
The aim of this study was to prove that an optimization method combining genetic algorithms with a direct-forcing immersed boundary method as a distinguished numerical method could improve the performance of vertical-axis wind turbine blades. The proposed direct-force immersed boundary (DFIB) flow solver was tested using the benchmark laminar flow problems (lid-driven flow, flow over a stationary cylinder, and vortex-induced vibration of the circular cylinder). Two cases were analyzed, one of a stationary airfoil and another of a rotating airfoil in a vertical-axis wind turbine. The analysis was carried out using two-dimensional flow simulations in a laminar flow regime. A NACA 0012 airfoil was used as the original airfoil cross section of the vertical-axis wind turbine. The proposed method successfully simulated the moving blade in the flow field, and the results revealed that the optimized wind turbine blade designed using the proposed method had an efficiency improvement of 5.61% compared to the original airfoil.
21
Genetic Algorithms and Machine Learning
David E. Goldberg, John H. Holland · Machine Learning · 1988 · 3K citations · Full text