Wind energy has been emerged as one of important alternative energy from nature sources. Especially in part of Europe, for example, Denmark and Germany, wind has become a definite choice of the national power generation policy to cope with energy problem in terms of environment and energy shortage.
In modern wind power system of large capacity above 1MW, horizontal axis wind turbine(HAWT) is a common type. And, the optimum design of wind turbine to guarantee excellent power performance and its reliability in structure and longevity is a key technology in wind industry.
In this study, mathematical expressions based upon the conventional blade element momentum theory(BEMT) applying to turbine blade design was first analyzed systematically to secure accurate power prediction from the basic aerodynamic parameters such as lift and drag coefficients, Prandtl's tip loss coefficient, tangential and axial flow induction factors and twist angle. X-FOIL open software was used to acquire lift and drag coefficients of the 2-D airfoils used in power prediction procedure. Three kinds of turbine rotor blade(named as FIL series, FIL-1000:1MW, FIL-100:100KW and FIL-20:20KW) were selected as examples of the optimum blade design and power prediction. Especially, in FIL-1000 case, three kinds of airfoil(FFA, DU and NACA series) were combined to produce maximum aerodynamic performance around blade tip and strong structure around blade root.
Furthermore, optimum blade design and its power prediction software, named as "POSEIDON", was also developed from the BEMT discussed in the present study. Several input data are necessary and power curve is acquired easily to show the overall characteristics of the suggested wind turbine performance. Three kinds of blades(FIL-1000, FIL-100 and FIL-20) were again selected to apply the suggested software. In case of FIL-1000, power coefficient was 0.47 at TSR=7. At FIL-100, power coeffi- cients were compared between the experimental aerodynamic data and X-FOIL prediction data, resulting in fairly good agreement. In case of FIL-20, single blade shape(NREL S809) was adopted and comparison between experimental aerodynamic parameters and X-FOIL data was made with good agreement in power prediction.
Finally, systematic computational fluid dynamics(CFD) analysis by commercial code CFX ver.5.7.1 was performed to study the detailed flow characteristics upon the blade surface and wake region. Turbulence model, k-ω SST was selected to guarantee stall phenomena, one of 3-D separation flow occurring in wind turbine. ICEM-CFD, reliable grid generation commercial software was also adapted to secure good quality of grid generation necessary for the reliable CFD simulation.
Three kinds of turbine were again selected to represent the complex 3-D stall flows appearing in the blade surface at various TSR condition. Centrifugal acceleration and pressure difference between hub and tip played a major role to govern the stall phenomena.
Power prediction from the CFD data was also made and compared with BEMT prediction in case of FIL-20, with satisfactory agreement.
In the future, aero-elastic analysis of the blade will be performed by the two-way FSI method software such as CFX-ANSYS and the more reliable blade design procedure and its performance prediction will be established.