Loading...

Multi-objective Optimization of a Megawatt Wind Turbine Blade Geometry Using an Evolutionary Algorithm

Mahboubi Fouladi, Hossein | 2014

767 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46529 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Darbandi, Masoud
  7. Abstract:
  8. Without optimization, it is impossible to reduce the cost of power generation from an efficient megawatt wind turbine. In this work, we present a multi-objective algorithm to optimize the megawatt blade geometry. In this algorithm, the mass of blade and the wind turbine annual energy production (AEP) are considered as the objective functions. The design variables are blade chord, blade twist, airfoil thickness, spar geometry and blade curvature distribution.The constraints are maximum allowable strain for the chosen materials, maximum tip deflection and maximum tip speed. For the internal geometry, we consider two spars and four panels. The blade element momentum method (BEM) is used for aerodynamic calculations and the Euler-Bernoulli beam theory for the structural calculations. The design point is chosen based on minimum cost of energy (COE) among Pareto front solutions, which are obtained within a two-objective optimization procedure. The current presented algorithm benefits from two main innovations, i.e., using the blade curvature distribution as a design variable and using a novel approach to satisfy the maximum blade tip deflection constraint. The current developed algorithm is validated against the results of reliable softwares in the field. To analyze the performance of presented algorithm, the optimization procedure is used to optimize the geometry at one baseline blade, i.e., the NREL 5MW reference wind turbine blade. The results of this optimization shows that the optimum blade is 23% lighter than the base blade and that it provides 0.2% more AEP than baseline blade. In the second stage, we use the current developed algorithm and optimize the rotor radius for a 2MW turbine blade in a site specific design process. In this design, hub height is assumed to vary linearly with rotor radius. The optimization is performed for different rotor radii and we optain the COE curve as a function of rotor radius. We calculate the optimum rotor radius for three different sites. Then, these three blades are compared with a blade designed to operate at these three sites with predicted capacities. The latter design is called a robust design (RD). It is shown that the RD turbine rotor produces about 0.2 to 3% lower total COE than its cumulative value for three site specific turbines. Eventually, the blade geometry is suitably optimized at the optimal rotor radius
  9. Keywords:
  10. Multiobjective Genetic Algorithm (MOGA) ; Multidisciplinary Optimization ; Turbine Blades ; Megawatt Wind Turbine ; Blade Geometry Optimization ; Site Specific Design

 Digital Object List

 Bookmark

No TOC