Collaborating teams compare wind turbine simulation codes

The wind energy community pairs the actuator line model, one of the most accurate representations of a wind turbine, with large eddy simulation (LES), to study airflow through wind turbines, and to help optimize wind farm operation in the future.

Researchers often use their own LES codes in their simulations. Four independent teams of researchers wanted to find out if their codes agree, or at least conduct basically the same simulation. They ran and compared their simulations of a wind turbine under uniform laminar inflow using their four different codes and published their results in the Journal of Renewable and Sustainable Energy.

The four codes are: LESGO from Johns Hopkins University, SOWFA from National Renewable Energy Laboratory, SP-Wind from KU Leuven, and EllipSys3D from Technical University of Denmark. Each code employs different numerical methods, but the initial parameters for each method were matched to create the most similar simulation possible.

In the simulation where air flow passes through the turbine and continues downstream, all four codes predict that flow would be smooth near the turbine. However, further downstream, where flow becomes more chaotic as it transitions from a nonturbulent state to a turbulent state, the different codes predict different locations for this transition.

These results mean that differences in the numerical methods used by the codes do not create consequential differences when predicting flow near the turbine. Agreement near the turbine is more important than agreement downstream, because it indicates that the actuator line model used to represent the turbine is essentially the same in all codes regardless of numerical methods. Co-author Luis Martínez said this work shows that researchers using different codes are all “on the same page,” a reassurance for the wind energy community.

Next, the researchers will run bigger simulations, modeling hundreds of wind turbines instead of one.

Source: “Comparison of four large-eddy simulation research codes and effects of model coefficient and inflow turbulence in actuator-line-based wind turbine modeling,” by Luis A. Martínez-Tossas, Matthew J. Churchfield, Ali Emre Yilmaz, Hamid Sarlak, Perry L. Johnson, Jens N. Sørensen, Johan Meyers, and Charles Meneveau, Journal of Renewable and Sustainable Energy (2018). The article can be accessed at https://doi.org/10.1063/1.5004710 .