Computational Flow Analysis of Wind Farms Using a Simplified Rotor Disc Model With Radially Varying Thrust Coefficient
Large scale horizontal axis wind turbines are one of the most promising renewable energy technologies. When they are installed in wind farms (onshore and offshore) they can exploit the most of the available wind energy of a site. The accurate calculation of their aerodynamic interaction due to the wake development is crucial for the design of the layout and the operation of a wind farm. Simulating a wind farm with more than one fully detailed wind turbines and possibly complex terrain geometry requires significant computational power and time. Therefore, the turbine rotors are approximated as discs which behave as momentum sinks. This approach has been adopted in the present study which focuses on the development of a simplified rotor disc model. However, in the present contribution, in order to approximate the axial thrust across the disc in a more accurate manner, a novel model that involves a radially varying thrust coefficient is utilized which is extracted from the CFD full rotor transient analysis results. The analysis is carried out with the use of two commercial CFD codes, ANSYS CFX and ANSYS Fluent for the full rotor and the simplified model, respectively. For the development of the radial distribution of thrust along the blades, both steady and transient computations are carried out and the results are compared against available experimental data. For the full rotor simulation the time-averaged transient results are compared against the steady ones and with the results of the actuator disc approach. Two different turbulence models, k-epsilon and Shear Stress Transport, were used along with the three dimensional RANS equations. The detailed assessment of the differences in the flow field as it is obtained from the steady and transient analysis of both full rotor and actuator disc approximations indicated that a good agreement exists between the two distributions and the existing differences are identified and quantified.