Section: Actuator

This section controls the actuator type models. This includes the actuator disk and line models. The prefix is the label set in incflo.physics. For example incflo.physics = FreeStream Actuator Actuator models are meant to simulate aerodynamic objects by using body forces in the momentum equation. There are capabilities to simulate fixed wings as actuator lines and wind turbines as actuator disks and actuator line models.

Actuator.labels

type: String, mandatory

This string is used as an identifier for the current actuator.

Actuator.type

type: String, mandatory

This string identifies the type of actuator to use. The ones currently supported are: UniformCtDisk, JoukowskyDisk, TurbineFastLine, TurbineFastDisk, TurbineKynemaLine, and FixedWingLine.

It is recommended to group common parameters across actuators using the Actuator.[type].[param]. For example:

Actuator.Turb1.type            = UniformCtDisk"
Actuator.Turb1.epsilon         = 5.0 5.0 5.0"
Actuator.Turb2.type            = UniformCtDisk"
Actuator.Turb2.epsilon         = 5.0 5.0 5.0"

becomes:

Actuator.UniformCtDisk.epsilon = 5.0 5.0 5.0"
Actuator.Turb1.type            = UniformCtDisk"
Actuator.Turb2.type            = UniformCtDisk"

FixedWingLine

Example for FixedWingLine:

incflo.physics = FreeStream Actuator
Actuator.labels = F1
Actuator.type = FixedWingLine
Actuator.FixedWingLine.num_points = 21
Actuator.FixedWingLine.epsilon = 3.0 3.0 3.0
Actuator.FixedWingLine.epsilon_chord = 0.25 0.25 0.25
Actuator.FixedWingLine.fllc = 0
Actuator.FixedWingLine.pitch = 4.0
Actuator.FixedWingLine.span_locs = 0.0 1.0
Actuator.FixedWingLine.chord = 2.0 2.0
Actuator.FixedWingLine.airfoil_table = DU21_A17.txt
Actuator.FixedWingLine.airfoil_type = openfast
Actuator.F1.start = 0.0 -4.0 0.0
Actuator.F1.end = 0.0 4.0 0.0
Actuator.F1.output_frequency = 10
ICNS.source_terms = ActuatorForcing

Actuator.FixedWingLine.num_points

type: int, mandatory

This is the number of actuator points along the wing to be used in the simulation.

Actuator.FixedWingLine.epsilon

type: List of 3 real numbers, mandatory

This is the value of epsilon in the chord, thickness and spanwise directions.

Actuator.FixedWingLine.epsilon_chord

type: List of 3 real numbers, optional

This is the value of epsilon/chord. This value will be used to compute epsilon as a function of the chord at every actuator point. A value of epsilon / chord ~ 0.25 is recommended for an optimal representation of the blade aerodynamics. When this variable is specified, the code will choose the maximum value between epsilon_chord * chord and epsilon for every actuator point.

Actuator.FixedWingLine.fllc

type: Bool, optional

This option will activate the filtered lifting line correction (fllc). The correction follows the implementation of Martinez-Tossas and Meneveau (2019) and Blaylock et al (2022). The use of the fllc requires epsilon and an optimal epsilon_chord as an input. The recommended value is 0.25 in all directions for epsilon_chord and a value of epsilon in all directions that would be greater than at least 2.5 times the grid size dx. The default is 0.

Actuator.FixedWingLine.fllc_type

type: String, optional, default = variable_chord

This option tells whether to use the original fllc formulation outlined in Martinez-Tossas and Meneveau (2019), which assumes a constant chord length across blade (specified as constant_chord), or to use a new formulation outlined in Martinez-Tossas et al. (2023), which accounts for chord variations (specified as variable_chord).

Actuator.FixedWingLine.fllc_relaxation_factor

type: Double, optional

The relaxation factor to be applied to the updated velocity see: Martinez-Tossas and Meneveau (2019) The default value is 0.1.

Actuator.FixedWingLine.fllc_start_time

type: Double, optional

The time in the simulation from when to start using the correction. The default value is 0.

Actuator.FixedWingLine.fllc_nonuniform

type: Bool

The flag to specify if the actuator points used to compute the correction should be non-uniformly distributed. This helps in using less points for the fllc while still maintaining the accuracy of the fllc. The default value is true.

Actuator.FixedWingLine.fllc_epsilon_dr_ratio

type: Double, optional

The ratio of epsilon to actuator point spacing used to create a non-uniform distribution. A value of 1 or greater is recommended. The default value is 1.

Actuator.FixedWingLine.pitch

type: Real number, mandatory

This is the pitch angle of the blade in degrees. All coordinates will be pitched by this angle. In the case of a fixed wing, this would be the angle of attack of the wing with respect to the inflow velocity. This argument is mandatory unless a pitch timetable is specified.

Actuator.FixedWingLine.span_locs

type: List of real numbers, mandatory

These are non-dimensional span locations from 0 to 1. These locations are used to specify the chord values at every span location of the blade.

Actuator.FixedWingLine.chord

type: List of real numbers, mandatory

These are the chord values at every span location. The length of this array needs to be the same length as span_locs.

Actuator.FixedWingLine.airfoil_table

type: String, mandatory

This is the name of the file that contains the lookup table for lift and drag coefficients.

Actuator.FixedWingLine.airfoil_type

type: String, mandatory

This is the type of airfoil table lookup. The currently supported options are openfast and text.

Actuator.F1.start

type: List of 3 real numbers, mandatory

This is the starting point of the wing where the first actuator point will be.

Actuator.F1.end

type: List of 3 real numbers, mandatory

This is the end point of the wing where the last actuator point will be.

Actuator.F1.output_frequency

type: int, optional

This is how often to write actuator output. The default is 10.

Actuator.FixedWingLine.motion_type

type: String, optional

The FixedWingLine actuator allows for motion, though other aspects of the actuator remain fixed (such as the orientation and the dimensions). The currently supported options are none (default), linear, and sine. Linear motion moves the actuator at a constant velocity in a straight line whereas sine motion oscillates the actuator according to a temporal sine signal.

Actuator.FixedWingLine.velocity

type: List of 3 real numbers, mandatory when motion_type = linear

This vector provides the prescribed constant velocity of the actuator motion.

Actuator.FixedWingLine.sine_vector

type: List of 3 real numbers, mandatory when motion_type = sine

This vector provides the actuator displacement from its initial, specified location as it moves according to the oscillatory sine signal. The range of motion of the actuator will be between (initial location + sine vector) and (initial location - sine vector).

Actuator.FixedWingLine.sine_period

type: Real number, mandatory when motion_type = linear

This value specifies the temporal period of the sine signal.

Actuator.FixedWingLine.pitch_timetable

type: String, optional

File name of pitch timetable. This file must specify pitch angles at different times below a one-line header. When this argument is present, the pitch argument is no longer mandatory, and it will not be used.

Actuator.FixedWingLine.disable_spanwise_gaussian

type: Boolean, optional, default = false

When this option is turned on, the actuator Gaussian is disabled in the spanwise Gaussian, making the force distribution uniform in that direction. This option enables quasi-2D simulations with a fixed wing. The code will print warning statements if the detected spanwise direction is not periodic.

Actuator.FixedWingLine.normalize_spanwise

type: Boolean, optional, default = true

When the disable_spanwise_gaussian is true, the default behavior is to normalize the Gaussian and force quantities in the spanwise direction, preventing the number of actuator points or the actuator point spacing from affecting the results. When this option is false, the ordinary treatment of the Gaussian and force quantities in the spanwise direction is used instead. Setting this option to false can be useful for verification studies.

Actuator.FixedWingLine.prescribed_uinf

type: Real, optional, default = -1.0

This input allows the freestream velocity sampled by the actuator routines to be overwritten with a user-prescribed value. This feature becomes active when the prescribed value is non-negative.

Actuator.FixedWingLine.active_force_dirs

type: List of 3 real numbers, optional, default = 1.0 1.0 1.0

By default, the actuator force is computed and applied in every coordinate direction. This input allows actuator force coordinate directions to be deactivated by specifying a 0.0 in for the x, y, or z component of this vector.

TurbineFastLine

This actuator type requires an AMR-Wind build with OpenFAST coupling enabled. AMR-Wind provides flow quantities at the actuator point locations to OpenFAST and OpenFAST provides the actuator point locations and forces at those points. This tight coupling happens at every time step.

Example for TurbineFastLine:

incflo.physics = FreeStream Actuator
Actuator.labels = WTG01
Actuator.type = TurbineFastLine
Actuator.TurbineFastLine.rotor_diameter = 126.0
Actuator.TurbineFastLine.hub_height = 90.0
Actuator.TurbineFastLine.num_points_blade = 64
Actuator.TurbineFastLine.num_points_tower = 12
Actuator.TurbineFastLine.epsilon = 10.0 10.0 10.0
Actuator.TurbineFastLine.epsilon_chord = 0.25 0.25 0.25
Actuator.TurbineFastLine.fllc = 0
Actuator.TurbineFastLine.epsilon_tower = 5.0 5.0 5.0
Actuator.TurbineFastLine.openfast_start_time = 0.0
Actuator.TurbineFastLine.openfast_stop_time = 1.0
Actuator.TurbineFastLine.nacelle_drag_coeff = 0.0
Actuator.TurbineFastLine.nacelle_area = 0.0
Actuator.TurbineFastLine.output_frequency = 10
Actuator.TurbineFastLine.density = 1.225
Actuator.WTG01.base_position = 5.0191 0. -89.56256
Actuator.WTG01.openfast_input_file = "fast_inp/nrel5mw.fst"
ICNS.source_terms = ActuatorForcing

Actuator.TurbineFastLine.rotor_diameter

type: Real number, required

This is the rotor diameter of the turbine to be simulated.

Actuator.TurbineFastLine.hub_height

type: Real number, required

This is the hub height of the turbine.

Actuator.TurbineFastLine.num_points_blade

type: int, required

This the number of actuator points along the blades.

Actuator.TurbineFastLine.num_points_tower

type: int, required

This is the number of actuator points along the tower.

Actuator.TurbineFastLine.epsilon

Same as Actuator.FixedWingLine.epsilon.

Actuator.TurbineFastLine.epsilon_chord

Same as Actuator.FixedWingLine.epsilon_chord.

Actuator.TurbineFastLine.fllc

Same as Actuator.FixedWingLine.fllc.

Actuator.TurbineFastLine.fllc_relaxation_factor

Same as Actuator.FixedWingLine.fllc_relaxation_factor.

Actuator.TurbineFastLine.fllc_type

Same as Actuator.FixedWingLine.fllc_type.

Actuator.TurbineFastLine.openfast_start_time

type: Real, required

This is the time at which to start the openfast simulation.

Actuator.TurbineFastLine.openfast_stop_time

type: Real, required

This is the time at which to stop the openfast run.

Actuator.TurbineFastLine.nacelle_drag_coeff

type: Real, optional

This is the drag coefficient of the nacelle. If this and the area of the nacelle are specified, a value of epsilon for the nacelle is computed that would provide an optimal momentum thickness of the wake. This is also used to correct the sampled velocity at the location of the nacelle actuator point.

Actuator.TurbineFastLine.nacelle_area

type: Real, optional, default=0

This is the frontal area of the nacelle which is used to compute the force.

Actuator.TurbineFastLine.output_frequency

type: int, optional, default=10

This is how often to write actuator output.

Actuator.TurbineFastLine.density

type: Real, optional

This is the density of the fluid specified in openfast. This is used to non-dimensionalize the forces from openfast.

Actuator.WTG01.openfast_input_file

type: String, required

This is the name of the openfast input file with all the turbine information.

TurbineKynemaLine

This actuator type requires an AMR-Wind build with Kynema coupling enabled. This is a similar coupling to OpenFAST, but Kynema acts as the turbine solver in this instance. Some turbine quantities that the OpenFAST interface needs from the AMR-Wind input file are instead found directly by the code within the Kynema input file, whereas other quantities that OpenFAST has stored within its inputs need to be directly supplied through the AMR-Wind input file for Kynema, especially for initialization.

Example for TurbineKynemaLine:

incflo.physics = FreeStream Actuator
Actuator.labels = WTG01
Actuator.type = TurbineKynemaLine
## Turbine discretization parameters
Actuator.TurbineKynemaLine.num_struct_nodes_blade = 6
Actuator.TurbineKynemaLine.num_points_blade = 64
Actuator.TurbineKynemaLine.num_points_tower = 0 # not enabled yet
## Turbine setup
Actuator.TurbineKynemaLine.rot_speed_rpm    = 12.1
Actuator.TurbineKynemaLine.yaw_deg = 30
Actuator.WTG01.kynema_input_file = NREL-15MW-aero.yaml
Actuator.WTG01.base_position = 5.0191 0. -89.56256
## Turbine - flow coupling parameters
Actuator.TurbineKynemaLine.epsilon = 10.0 10.0 10.0
Actuator.TurbineKynemaLine.epsilon_chord = 0.25 0.25 0.25
Actuator.TurbineKynemaLine.fllc = 0
Actuator.TurbineKynemaLine.nacelle_drag_coeff = 0.0
Actuator.TurbineKynemaLine.nacelle_area = 0.0
Actuator.TurbineKynemaLine.density = 1.225
## Turbine controller parameters and initial state
Actuator.TurbineKynemaLine.controller_shared_library_path = /path/to/libdiscon.so # or libdiscon.dylib
Actuator.TurbineKynemaLine.generator_power_init = 5e6
Actuator.TurbineKynemaLine.hub_wind_vector_init = 9.8726896031426 5.7 0.0
Actuator.TurbineKynemaLine.generator_efficiency = 0.944
## Turbine solver numerical parameters
Actuator.TurbineKynemaLine.dt = 0.01
Kynema.abs_err_tol = 1e-6

Actuator.TurbineKynemaLine.output_frequency = 10

ICNS.source_terms = ActuatorForcing

Actuator.TurbineKynemaLine.num_struct_nodes_blade

type: Int, required

This is the number of structural nodes for Kynema to use when modeling each turbine blade.

Actuator.TurbineKynemaLine.num_points_blade

type: Int, required

This is the number of aerodynamic sections for Kynema to use when modeling each turbine blade. This will correspond to the number of force points and velocity points on each blade in AMR-Wind. This must be the same number as provided in the Kynema input file.

Actuator.TurbineKynemaLine.num_points_tower

type: Int, required

This is the number of aerodynamic sections for Kynema to use when modeling the tower. This feature is still under development, so this argument must be set to 0.

Actuator.TurbineKynemaLine.rot_speed_rpm

type: Real, optional, default = 0

This is the initial rotational speed of the turbine in RPM. This parameter can alternatively be set in radians per second using the input parameter Actuator.TurbineKynemaLine.rot_speed_radps.

Actuator.TurbineKynemaLine.rot_speed_radps

type: Real, optional, default = 0

This is the initial rotational speed of the turbine in radians per second. If this argument is present, Actuator.TurbineKynemaLine.rot_speed_rpm will be ignored.

Actuator.TurbineKynemaLine.yaw_deg

type: Real, optional, default = 0

This is the initial yaw angle of the turbine in degrees, counterclockwise from the -x direction. This parameter can alternatively be set in radians using the input parameter Actuator.TurbineKynemaLine.yaw_rad.

Actuator.TurbineKynemaLine.yaw_rad

type: Real, optional, default = 0

This is the initial yaw angle of the turbine in radians. If this argument is present, Actuator.TurbineKynemaLine.yaw_deg will be ignored.

Actuator.TurbineKynemaLine.kynema_input_file

type: String, required

This is the input file used to initialize the Kynema turbine model. It conforms to the WindIO format. A pre-processing tool is provided in the Kynema repository to change the number of aerodynamic sections per blade, if needed, as well as to address some format edge cases.

Actuator.TurbineKynemaLine.controller_shared_library_path

type: String, optional, default = empty

This is the path to the controller shared library (typically ROSCO). If this parameter is not provided, no controller will be created in the turbine model, and the controller-related input parameters will not be used.

Actuator.TurbineKynemaLine.generator_power_init

type: Real, optional, default = 0

Power of the generator at the start of the simulation.

Actuator.TurbineKynemaLine.hub_wind_vector_init

type: Vector<Real>, optional, default = 0 0 0

This is the initial wind vector that the turbine hub is exposed to. It does not have to be the actual wind there at initialization; this number is converted to a wind speed that is used as the controller’s initial guess.

Actuator.TurbineKynemaLine.generator_efficiency

type: Real, optional, default = 1

This is the efficiency of the generator. If not populated, the efficiency is assumed to be 1, i.e., 100%.

Actuator.TurbineKynemaLine.dt

type: Real, optional, default = same as AMR-Wind dt

This is the time step size chosen for the Kynema turbine model. It must be a factor of the AMR-Wind time step so that Kynema can take an integer number of sub-steps for each AMR-Wind time step. If not populated, the Kynema time step size will be the same as the flow solver time step, and, due to the robustness of Kynema, this is typically fine.

Actuator.TurbineKynemaLine.output_frequency

type: Int, optional, default = 10

This is how often, in time steps, to output actuator data from AMR-Wind. Note, this does not govern how often Kynema outputs turbine data. Kynema automatically outputs data every AMR-Wind time step.

Kynema.abs_err_tol

type: Real, optional, default = 1e-5

This turbine solver parameter is not turbine-specific; rather, it informs the solution parameters of Kynema overall. This, in particular, sets the absolute tolerance of the Kynema solver.

Kynema.rel_err_tol

type: Real, optional, default = 1e-4

This parameter sets the relative tolerance of the Kynema solver.

Kynema.max_nonlinear_iterations

type: Int, optional, default = 12

This parameter sets the maximum number of nonlinear iterations for the Kynema solver.

Kynema.damping_factor

type: Real, optional, default = 0

This parameter sets the numerical damping (time-based) of the Kynema solver. Counterintuitively, full damping corresponds to 0 and no damping corresponds to 1.

Active Wake Control with Joukowsky Disk

There is preliminary support for exploring Active Wake Control (AWC) strategies with the Joukowsky disk model. The current implementation follows Cheung et. al (2024). The following input options allow for enabling AWC:

Actuator.WTG01.awc_angular_frequency

type: Real, optional, default=0

Sets the temporal angular frequency for AWC (in radians)

Actuator.WTG01.awc_amplitude

type: Real, optional, default=0

Sets the amplitude of the forcing term in AWC relative to the axial force

Actuator.WTG01.awc_azimuthal_mode

type: Int, optional, default=0

Sets the azimuthal mode for the AWC (e.g. 0 denotes a pulse mode, 1 denotes a helical mode)

Actuator.WTG01.awc_clocking_angle

type: Real, optional, default=0

Sets the clocking angle to adjust the orientation of the modes in the azimuthal direction (in radians)

ActuatorSourceTagging

It is possible to seed a passive scalar in the flow field at locations where the actuator source term value is above a certain threshold. This is useful for wake visualization and for dynamic adaptation of the mesh to the wake location. This is activated by adding ActuatorSourceTagging to incflo.physics. It has the following input options:

ActuatorSourceTagging.actuator_source_threshold

type: Real, optional, default=0.1

Threshold value for the actuator source term above which the passive scalar will be set to 1.0.

Additional input parameters are transport.passive_scalar_laminar_schmidt and transport.passive_scalar_turbulent_schmidt to set the diffusion of the passive scalar. This can be combined with the FieldRefinement criteria for mesh adaptation:

tagging.labels = tracer
tagging.tracer.type = FieldRefinement
tagging.tracer.field_name = passive_scalar
tagging.tracer.field_error = 0.3 0.3 0.3 0.3

where the field_error is the value above which the cells should be tagged for refinement. Here is an example using the uniform_ct_disk_dynamic_adaptation regression test:

Warning

This is an experimental feature and there is no guidance yet on the values that should be used for the passive scalar and tagging criteria.