wayve#

WAYVE is an open-source code Python framework for the atmospheric perturbation model (APM).

The APM is an atmospheric perturbation model developed to simulate the interaction between wind farms and the atmosphere. It focuses on the generation and feedback effects of gravity waves. This code implements the model described in the following papers:

  • Allaerts D., Meyers J. Sensitivity and feedback of wind-farm-induced gravity waves. Journal of Fluid Mechanics, 2019. DOI

  • Devesse K., Lanzilao L., & Meyers J. A meso-micro atmospheric perturbation model for wind farm blockage. Preprint, 2023. DOI

  • Devesse K., Stipa S., Brinkerhoff J., Allaerts D., & Meyers J. Comparing methods for coupling wake models to an atmospheric perturbation model in WAYVE. Journal of Physics: Conference Series, 2024. DOI

Configuration Options#

WAYVE-specific options are configured in the analysis.yaml file:

APM Grid Settings:

apm_grid:
  Lx: 1.e6              # Grid size in x-direction [m]
  Ly: 1.e6              # Grid size in y-direction [m]
  dx: 500               # Grid spacing [m]
  L_filter: 1.e3        # Spatial filter length [m]

Layer Description:

layers_description:
  farm_layer_height: 238.       # Lower layer height [m]
  number_of_fa_layers: 1        # Number of free atmosphere layers (1 or more)

Wake Model Coupling:

wm_coupling:
  method: "PB"          # Options: PB (Pressure-Based), VM (Velocity Matching), US (Upstream)
  settings:             # Method-specific settings
    alpha: 1.0          # VM method parameter
    distance: 500       # US method parameter [m]
  subgrid:
    include_subgrid: false
    D_to_dx: 0.5

Additional APM Terms:

APM_additional_terms:
  momentum_entrainment:
    mfp_type: "constant_flux"   # Options: None, constant_flux
    apm_mfp_settings:
      a_mfp: 0.120              # Momentum flux coefficient
      d_mfp: 27.80              # Distance parameter
  apm_disp_stresses:
    ds_type: subgrid            # Dispersive stresses type

Wake Model Settings:

wake_tool: wayve        # Options: wayve, foxes
wind_deficit_model:
  name: Bastankhah2014
  wake_expansion_coefficient:
    k_a: 0.0
    k_b: 0.04
  ceps: 0.2

Input Requirements#

WAYVE requires atmospheric profile data for proper ABL initialization:

Hub-height time series (minimal):

  • Wind speed and direction at hub height

  • Optional: LMO (Monin-Obukhov length), friction velocity, roughness height

Vertical profiles (recommended for stratified flows):

  • Wind speed and direction profiles

  • Potential temperature profile

  • Turbulence intensity profile

  • Optional: TKE and dissipation profiles

Capping inversion data (for gravity waves):

  • ABL height

  • Inversion strength (dtheta)

  • Inversion depth (dH)

  • Free atmosphere lapse rate

API Reference#

Usage#

Command Line:

wifa_wayve path/to/system.yaml

Python:

from wifa.wayve_api import run_wayve

# Run WAYVE simulation
run_wayve("path/to/system.yaml", output_dir="results")
wifa.wayve_api.run_wayve(yamlFile, output_dir='output', debug_mode=False)#

Run a WAYVE (APM) simulation from WindIO input.

Parameters:

  • yamlFile (str or dict) – Path to WindIO YAML file or pre-loaded dictionary

  • output_dir (str) – Output directory for results

  • debug_mode (bool) – Enable debug mode (wake model only, no APM solve)

Example:

from wifa.wayve_api import run_wayve

run_wayve("system.yaml", output_dir="wayve_results")

Outputs#

  • turbine_data.nc: Power and rotor effective velocity per turbine

  • flow_field.nc: Wind speed and direction on specified grid (if requested)