D5.1. Transient or Steady

Wind is inherently a time-varying phenomenon, and its simulation can capture various outcomes, such as temporal averages of velocity and turbulence, or dynamic flow effects, depending on the chosen method (see Table 6).

Table 6: Flow Quantities Calculable with RANS, URANS, and LES

According to Shimada and Ishihara, an instantaneous flow quantity Φ can be decomposed into a temporal average Φ ‾, a periodic fluctuation Φ ˜, and a stochastic-turbulent fluctuation Φ^'. In the simplest form, the RANS simulation (Reynolds-Averaged Navier-Stokes) only calculates the temporal averages Φ ‾ of the flow quantities, leaving out dynamic effects. To capture unsteady flow effects, a transient simulation is required. URANS (Unsteady RANS) enables the calculation of temporal averages Φ ‾ plus periodic fluctuations Φ ˜, though it tends to underestimate the overall fluctuation. In URANS, besides the appropriate choice of grid and time step, the turbulence model and discretization scheme play a critical role. Overly dissipative turbulence models can dampen unsteady effects, so non-dissipative discretization schemes are recommended for unsteady simulations.

The most comprehensive approach is LES (Large Eddy Simulation), which can represent temporal averages Φ ‾, periodic Φ ˜, and stochastic-turbulent fluctuations Φ^'. This decomposition illustrates the capabilities and limitations of different simulation methods for capturing unsteady flow phenomena. The choice of the appropriate method ultimately depends on the specific requirements of each study and must be carefully considered to adequately capture the relevant aspects of wind flow.