4.10.2. Input Files

The user configures the hydrodynamic model parameters as well as the substructure geometry and properties via a primary HydroDyn input file. When used in standalone mode, an additional driver input file is required. This driver file specifies initialization inputs normally provided to HydroDyn by OpenFAST, as well as the per-time-step inputs to HydroDyn.

No lines should be added or removed from the input files, except in tables where the number of rows is specified.

4.10.2.1. Units

HydroDyn uses the SI system (kg, m, s, N).

4.10.2.2. HydroDyn Driver Input File

The driver input file is only needed for the standalone version of HydroDyn and contains inputs normally generated by OpenFAST, and are necessary to control the hydrodynamic simulation for uncoupled models. A sample HydroDyn driver input file is given in Appendix B.

Set the Echo flag in this file to TRUE if you wish to have HydroDynDriver echo the contents of the driver input file (useful for debugging errors in the driver file). The echo file has the naming convention of OutRootName.dvr.ech. OutRootName is specified in the HYDRODYN section of the driver input file. Set the gravity constant using the Gravity parameter. HydroDyn expects a magnitude, so in SI units this would be set to 9.80665 \(\frac{m}{s^{2}}\). WtrDens specifies the water density and must be a value greater than or equal to zero; a typical value of seawater is around 1025 kg/m3. WtrDpth specifies the water depth (depth of the flat seabed), based on the reference MSL, and must be a value greater than zero. MSL2SWL is the offset between the MSL and SWL, positive upward. This parameter is useful when simulating the effect of tides or storm-surge sea-level variations without having to alter the substructure geometry information. This parameter is unused with WaveMod = 6 and must be set to zero if you are using a potential-flow model (PotMod = 1 or 2). WaveMod and PotMod are specified in the HydroDyn primary input file.

HDInputFile is the filename of the primary HydroDyn input file. This name should be in quotations and can contain an absolute path or a relative path. Note that neither the standalone HydroDyn program nor HydroDyn as part of OpenFAST can run without SeaState. Therefore, in addition to the primary HydroDyn input file, the path to a SeaState input file must also be provided through SeaStateInputFile.

All HydroDyn-generated output files will be prefixed with OutRootName. If this parameter includes a file path, the output will be generated in that folder. Linearize can be set to either TRUE or FALSE. If TRUE, linearized 6-by-6 stiffness, damping, and added-mass matrices will be computed and printed to the calling terminal. NSteps specifies the number of simulation time steps, and TimeInterval specifies the time between steps.

Motion of the structure can be specified in different ways according to PRPInputsMod. Irrespective of the choice of PRPInputsMod (which are explained below), the translational displacement, velocity, and acceleration are always specified in the global inertial-frame coordinate system. With OpenFAST now updated to support potentially large platform rotation, the specification of rotation differs from previous versions. HydroDyn now describes body rotation using Tait-Bryan roll, pitch, and yaw angles with the convention of intrinsic (about body-fixed axis) yaw rotation first, followed by pitch rotation, and roll last. Furthermore, HydroDyn now expects the first and second time derivatives of the Tait-Bryan roll, pitch, and yaw angles in place of angular velocity and acceleration. The standalone HydroDyn driver will convert these inputs to angular velocity and acceleration internally.

Setting PRPInputsMod = 0 forces all platform reference point (PRP) input motions to zero for all time. If you set PRPInputsMod = 1, then you must set the steady-state inputs in the PRP STEADY STATE INPUTS section of the file. Setting PRPInputsMod = 2 requires the time-series input file whose name is specified via the PRPInputsFile parameter. The PRP input file is a text-formatted file. This file has no header lines. Each data row corresponds to a given time step, and the whitespace separated columns of floating point values represent the necessary motion inputs as shown in Table 4.6.

Table 4.6 PRP Inputs Time-Series Data File Contents (PRPInputsMod = 2)

Column Number

Input

Units

1

Time step value
\[s\]

2-4

Translational displacements along X, Y, and Z
\[m\]

5-7

Tait-Bryan roll, pitch, and yaw angles
\[\text{radians}\]

8-10

Translational velocities along X, Y, and Z
\[\frac{m}{s}\]

11-13

First time derivatives of the Tait-Bryan roll, pitch, and yaw angles
\[\frac{\text{radians}}{s}\]

14-16

Translational accelerations along X, Y, and Z
\[\frac{m}{s^{2}}\]

17-19

Second time derivatives of the Tait-Bryan roll, pitch, and yaw angles
\[\frac{\text{radians}}{s^{2}}\]

With PRPInputsMod = 1 or 2, any potential-flow bodies and strip-theory members defined in the primary HydroDyn input file will follow the prescribed motion of the PRP according to rigid-body kinematics. The standalone HydroDyn does not check for physical consistency between motions (displacement, velocity, and acceleration) specified for the PRP in the driver file. The user is responsible for generating consistent kinematics if that is important.

If the HydroDyn model contains one or more potential-flow bodies, PRPInputsMod can be set to a negative number with \(|\ \text{**PRPInputsMod**}\ | = \text{**NBody**}\) in the HydroDyn primary input file. In this case, an alternative form of the PRP input file shown in Table 4.7 can be used to specify different motions for the PRP, which controls the motion of all strip-theory members based on rigid-body kinematics, and for each potential-flow bodies separately. With this option, the user only specifies the translational and rotational displacements. HydroDyn will compute the velocity and acceleration by numerically differentiating the displacement with respect to time.

Table 4.7 PRP Inputs Time-Series Data File Contents (PRPInputsMod < 0)

Column Number

Input

Units

1

Time step value
\[s\]

2-4

Translational displacements of the PRP along X, Y, and Z
\[m\]

5-7

Tait-Bryan roll, pitch, and yaw angles of the PRP
\[\text{radians}\]

8-10

Translational displacements of the 1st potential-flow body along X, Y, and Z
\[m\]

11-13

Tait-Bryan roll, pitch, and yaw angles of the 1st potential-flow body
\[\text{radians}\]

14-16

Translational displacements of the 2nd potential-flow body along X, Y, and Z
\[m\]

17-19

Tait-Bryan roll, pitch, and yaw angles of the 2nd potential-flow body
\[\text{radians}\]

…

…

…

4.10.2.3. HydroDyn Primary Input File

The HydroDyn input file defines the substructure geometry, hydrodynamic coefficients, incident wave kinematics and current, potential-flow solution options, flooding/ballasting and marine growth, and auxiliary parameters. The geometry of strip-theory members is defined by joint coordinates of the undisplaced substructure in the global reference system, with the origin at the intersection of the undeflected tower centerline with MSL. A member connects two joints; multiple members can use a common joint. The hydrodynamic loads are computed at nodes, which are the resultant of member refinement into multiple (MDivSize input) elements (nodes are located at the ends of each element), and they are calculated by the module. Member properties include outer diameter, thickness, and dynamic-pressure, added-mass and viscous-drag coefficients. Member properties are specified at the joints; if properties change from one joint to the other, they will be linearly interpolated for the inner nodes.

The file is organized into several functional sections. Each section corresponds to an aspect of the hydrodynamics model or the submerged substructure. A sample HydroDyn primary input file is given in Appendix A: OC4 Semi-submersible Input File.

If this manual refers to an ID in a table entry, this is an integer identifier for the table entry, and these IDs do not need to be consecutive or increasing, but they must be unique for a given table entry.

The input file begins with two lines of header information for your use, but they are not used by the software. On the next line, set the Echo flag to TRUE if you wish to have HydroDyn echo the contents of the HydroDyn input file (useful for debugging errors in the input file). The echo file has the naming convention of OutRootName.HD.ech. OutRootName is either specified in the HYDRODYN section of the driver input file when running HydroDyn standalone, or by OpenFAST when running a coupled simulation.

4.10.2.3.1. Floating Platform

This and the next few sections of the input file have “Floating Platform” in the title, but the input parameters control the potential-flow model, regardless of whether the substructure is floating or not. The potential-flow solution cannot be used in conjunction with nonzero MSL2SWL or WaveMod = 6 in SeaState.

If the load contributions from potential-flow theory are to be included, set PotMod to 1 to use frequency-to-time domain transforms based on WAMIT output or 2 to use FIT (FIT is not yet documented in this manual). The remaining parameters in this section are only used when PotMod = 1.

ExctnMod can be set to 0 for no wave excitation, 1 for frequency-to-time domain wave excitation using discrete Fourier transform, or 2 for the state-space wave-excitation model. Depending on the choice of ExctnMod, suitable hydrodynamic input files must be provided through the PotFile input. More information below.

ExctnDisp specifies if and how structure displacement in the horizontal plane should be considered when evaluating the potential-flow wave excitation. Setting ExctnDisp = 0 ignores structure displacement, and wave excitation will be computed using the undisplaced structure position as in previous versions of OpenFAST. If ExctnDisp = 1, HydroDyn will compute the potential-flow wave excitation using the unfiltered instantaneous PRP position in the horizontal plane. If ExctnDisp = 2, HydroDyn will instead compute the wave excitation based on the low-pass filtered PRP position in the horizontal plane. The cutoff frequency is specified through ExctnCutOff in Hz. This option is useful when second-order potential-flow wave excitation is enabled. The cutoff frequency should be set to filter out as much of the wave-frequency PRP motion as possible while retaining the low-frequency drift motion to prevent double counting the contributions from first-order structural motion already included in the second-order potential-flow wave excitation.

HydroDyn now supports large but slow (well below wave frequencies) transient platform yaw motion with both strip-theory only and hybrid potential-flow models. To enable this capability, the inputs PtfmYMod, PtfmRefY, PtfmYCutoff, and NExctnHdg must be set appropriately. Note that HydroDyn still requires the platform roll and pitch angles to be small, i.e., within +/-15 deg.

To conform with the first- and second-order potential-flow theory, which limits the structure to small displacement about a reference mean position, a constant or slowly varying reference platform yaw orientation must be established.

Setting PtfmYMod = 0 lets HydroDyn use a constant reference yaw angle given by PtfmRefY in degrees. In this case, the platform yaw rotation during the simulation, as given by the PRPYaw output channel, must stay within +/-15 deg of PtfmRefY specified by the user. A severe warning will be displayed if this requirement is not met at any point during the simulation, while still allowing the simulation to continue if possible. With a hybrid potential-flow model, the potential-flow wave excitation input file needs to cover a suitable range of wave headings relative to the platform after a yaw offset of PtfmRefY is applied.

Alternatively, PtfmYMod = 1 lets HydroDyn update the reference yaw position PtfmRefY dynamically based on the low-pass-filtered platform yaw rotation, analogous to the modeling of slow-drift motion with ExctnDisp = 2 above. In this case, the PtfmRefY input allows the user to specify the initial reference yaw position at t = 0. The cutoff frequency of the first-order low-pass filter for platform yaw rotation can be set with PtfmYCutoff in Hz. Ideally, PtfmYCutoff should be placed between the wave frequency region and the characteristic frequency of any slow but large change in platform heading to filter out as much wave-frequency platform motion as possible while minimizing the phase shift in the low-frequency heading change. Throughout the simulation, the instantaneous platform yaw rotation should stay within +/-15 deg of the now time-dependent PtfmRefY. A severe warning will be displayed if this requirement is not met at any point during the simulation, while still allowing the simulation to continue if possible.

With PtfmYMod = 1, HydroDyn requires the first- and second-order (mean- or slow-drift loads from Newman’s approximation only) potential-flow wave excitation input file(s) to cover the full range of possible wave headings with the first (smallest) wave heading being exactly -180 deg and the last (largest) wave heading being exactly +180 deg (the duplicated wave headings of +/-180 deg are intentional). HydroDyn will error out if this requirement is not met by the input files. HydroDyn uses this information to precompute the wave excitation on the platform for NExctnHdg evenly distributed platform yaw/heading angles over the range of [-180,+180) deg. For instance, with NExctnHdg = 36, HydroDyn will precomupte the wave excitation for 0, 10, 20, …, 350 deg platform heading. The instantaneous wave excitation applied on the platform during the time-domain simulation is interpolated from this data based on the instantaneous PtfmRefY. NExctnHdg should be set appropriately to ensure adequate angular resolution in platform heading. However, a high NExctnHdg can increase memory use by OpenFAST substantially.

Additional constraints on HydroDyn inputs apply when PtfmYMod = 1. The strip-theory hydrodynamic load must be evaluated using the wave kinematics and dynamic pressure at the displaced structure position by setting WaveDisp = 1. State-space wave excitation cannot be used. ExctnMod must be either 0 (no wave excitation) or 1 (frequency-to-time domain transform using inverse discrete Fourier transform). Lastly, full difference- and sum-frequency QTFs are not supported, requiring both DiffQTF and SumQTF to be set to 0. However, mean- or slow-drift loads based on Newman’s approximation can be included through the MnDrift or NewmanApp inputs explained below.

Note that the inputs PtfmYMod and PtfmRefY also affect the strip-theory hydrodynamic load. This is because the orientation of the strip-theory members is updated based on PtfmRefY instead of the instantaneous platform yaw rotation. Behavior of previous versions of HydroDyn can be approximately recovered by setting PtfmYMod = 0 and PtfmRefY = 0 deg, in which case, the inputs PtfmYCutoff and NExctnHdg are not used.

HydroDyn has two methods for calculating the radiation memory effect. Set RdtnMod to 1 for the convolution method, 2 for the linear state-space model, or 0 to disable the memory effect calculation. For the convolution method, RdtnTMax determines how long to track the memory effect (truncating the convolutions at t – RdtnTMax, where t is the current simulation time), but it also determines the frequency step used in the cosine transform, from which the time-domain radiation kernel (radiation impulse-response function) is derived. A RdtnTMax of 60 s is usually more than sufficient because the radiation kernel decays to zero after a short amount of time; setting RdtnTMax much greater than this will cause HydroDyn to run significantly slower. (RdtnTMax does not need to match or exceed the total simulation length.) Setting RdtnTMax to 0 s disables the memory effect, akin to setting RdtnMod to 0. For the convolution method, RdtnDT is the time step for the radiation calculations (numerical convolutions), but also determines the maximum frequency in the cosine transform. For the state-space model, RdtnDT is the time step to use for time integration of the linear state-space model. In this version of HydroDyn, RdtnDT must match the glue code (OpenFAST/driver program) simulation time step; the DEFAULT keyword can be used for this. Depending on the choice of RdtnMod, suitable hydrodynamic input files must be provided through the PotFile input. More information below.

HydroDyn supports the inclusion of multiple potential-flow bodies. NBody specifies the number of potential-flow bodies present. NBodyMod controls how multiple potential-flow bodies should be modeled. HydroDyn will retain the full hydrodynamic coupling among the potential-flow bodies if NBodyMod = 1. For this option, all bodies should be present in the same WAMIT run with NBody in HydroDyn being equal to NBODY in the WAMIT input file. The WAMIT output files should contain results for 6·NBody modes. HydroDyn will neglect hydrodynamic coupling among the potential-flow bodies if NBodyMod = 2 or 3. In either case, WAMIT should be run for each body separately one at a time. If the WAMIT computation is run with each body centered at the origin (XBODY=0 in WAMIT), NBodyMod = 2 should be used in HydroDyn. In this case, HydroDyn will process the WAMIT outputs to account for the shift in wave phase due to any offset of each potential-flow body from the origin/PRP. HydroDyn will also rotate the WAMIT outputs according to the heading of each body in HydroDyn. NBodyMod = 2 is convenient when, e.g., multiple identical potential-flow bodies are present in the structure. If the hydrodynamic coupling among the bodies can be neglected, the same set of WAMIT output files can be used for each body by setting NBodyMod = 2. On the other hand, NBodyMod = 3 should be used if each body is already positioned and oriented correctly relative to the origin/PRP in WAMIT by setting XBODY in the WAMIT input file. In this case, HydroDyn will use the provided WAMIT output as is.

The PotFile input should contain the path and root name (without extensions) for the WAMIT output files enclosed in quotation marks. These files consist of the .1, .3, .hst, and second-order files. The .hst file contains the hydrostatic restoring (stiffness) matrix. The .1 file contains the frequency-dependent hydrodynamic added-mass and damping matrix from the wave radiation problem. The .3 file contains the frequency- and direction-dependent first-order wave-excitation vector from the linear wave diffraction problem. These are written by the WAMIT program and should not include any file headers. When the linear state-space model is used in place of frequency-to-time domain transformation for wave excitation or in place of convolution for radiation, the .ssexctn file for wave excitation (more information to be provided in the future) and/or the .ss file for radiation generated by SS_Fitting must have the same root name as the other WAMIT-related files.

When NBodyMod = 1, PotFile should only contain one entry irrespective of NBody because the hydrodynamic coefficients for all bodies with hydrodynamic coupling should be contained within a single set of files. When NBodyMod = 2 or 3, PotFile should contain NBody entries, each enclosed in quotes and separated from each other with commas or spaces. Each entry of PotFile corresponds to a single potential-flow body.

In the reminder of this section, each input should contain NBody entries separated by commas or spaces, irrespective of NBodyMod.

The output files from WAMIT are in a standard nondimensional form that HydroDyn will dimensionalize internally upon input. WAMITULEN is the characteristic body length (in m) used to redimensionalize the WAMIT output. The body motion and force/moment in these WAMIT files are always resolved in the body-local frame of reference given by XBODY in the WAMIT input file. To correctly interpret the WAMIT outputs, the position and heading of each potential-flow body relative to the origin/PRP must be specified using PtfmRefxt, PtfmRefyt, PtfmRefzt, and PtfmRefztRot (in m or deg). With the exception of NBodyMod = 2, these inputs must match XBODY in the WAMIT input file. When NBodyMod = 2, these inputs can be set freely except for PtfmRefzt, which must always be zero.

While HydroDyn expects hydrodynamic coefficients derived from WAMIT, if you are not using WAMIT, it is recommended that you reformat your data according to the WAMIT format (including nondimensionalization) before inputting them to HydroDyn. Information on the WAMIT format is available from Chapter 4 of the WAMIT User’s Guide [LN06].

PtfmVol0 is the displaced volume of water when the potential-flow body is in its undisplaced position (in m3). This value should be set equal to the value computed by WAMIT as output in the WAMIT .out file. PtfmCOBxt and PtfmCOByt are the X and Y offsets (in m) of the center of buoyancy of each body from the origin/PRP, NOT from PtfmRefxt and PtfmRefyt.

4.10.2.3.2. 2nd-Order Floating Platform Forces

The 2ND-ORDER FLOATING PLATFORM FORCES section of the input file allows the option of adding second-order contributions to the potential-flow solution. When second-order terms are optionally enabled, the second-order terms are calculated using the first-order wave-component amplitudes and added to the first-order wave excitation at the difference and/or sum frequencies. The second-order terms cannot be computed without also including the first-order terms from the FLOATING PLATFORM section above (PotMod = 1). Enabling the second-order terms allows one to capture some of the nonlinearities in the wave loads, permitting more accurate modeling at the expense of greater computational effort (mostly at HydroDyn initialization).

While the cut-off frequencies in the 2nd-Order Waves section of the SeaState module apply to both the second-order wave kinematics used by strip theory and the second-order diffraction loads in potential-flow theory, the second-order terms themselves are enabled separately. The second-order wave kinematics used by strip theory are enabled in the 2nd-Order Waves section while the second-order diffraction loads in potential-flow theory are enabled in this section.

The second-order difference-frequency potential-flow terms can be enabled in one of three ways. To compute only the mean-drift term, set MnDrift to a nonzero value; to estimate the mean- and slow-drift terms using Standing et al.’s extension to Newman’s approximation, based only on first-order effects, set NewmanApp to a nonzero value; or to compute the mean- and slow-drift terms using the full difference-frequency QTF set DiffQTF to a nonzero value. Valid values of MnDrift are 0, 7, 8, 9, 10, 11, or 12 corresponding to which WAMIT output file the mean-drift terms will be calculated from. Valid values of NewmanApp are 0, 7, 8, 9, 10, 11, or 12 corresponding to which WAMIT output file the Newman’s approximation will be calculated from. Newman’s approximation cannot be used in conjunction with directional spreading (WaveDirMod must be 0) and the second-order cut-off frequencies do not apply to Newman’s approximation. Valid values of DiffQTF are 0, 10, 11, or 12 corresponding to which WAMIT output file the full difference-frequency potential-flow solution will be calculated from. Only one of MnDrift, NewmanApp, and DiffQTF can be nonzero; a setting of 0 disregards the second-order difference-frequency contributions to the potential-flow solution.

The .7 WAMIT file refers to the mean-drift loads (diagonal of the difference-frequency QTF) in all DOFs derived from the control-surface integration method based on the first-order solution. The .8 WAMIT file refers to the mean-drift loads (diagonal of the difference-frequency QTF) only in surge, sway, and yaw derived from the momentum conservation principle based on the first-order solution. The .9 WAMIT file refers to the mean-drift loads (diagonal of the difference-frequency QTF) in all DOFs derived from the pressure integration method based on the first-order solution. For the difference-frequency terms, 10, 11, and 12 refer to the WAMIT .10d, .11d, and .12d files, corresponding to the full QTF of (.*10d*) loads in all DOFs associated with the quadratic interaction of first-order quantities, (.*11d*) total (quadratic plus second-order potential) loads in all DOFs derived by the indirect method, and (.*12d*) total (quadratic plus second-order potential) loads in all DOFs derived by the direct method, respectively.

The second-order sum-frequency potential-flow terms can only be enabled using the full sum-frequency QTF, by setting SumQTF to a nonzero value. Valid values of SumQTF are 0, 10, 11, or 12 corresponding to which WAMIT output file the full sum-frequency potential-flow solution will be calculated from; a setting of 0 disregards the second-order sum-frequency contributions to the potential-flow solution. For the sum-frequency terms, 10, 11, and 12 refer to the WAMIT .10s, .11s, and .12s files, corresponding to the full QTF of (.*10s*) loads in all 6 DOFs associated with the quadratic interaction of first-order quantities, (.*11s*) total (quadratic plus second-order potential) loads in all DOFs derived by the indirect method, and (.*12s*) total (quadratic plus second-order potential) loads in all DOFs derived by the direct method, respectively.

Note that also apply here are the various considerations associated with running WAMIT for multiple potential-flow bodies discussed in the FLOATING PLATFORM section for first-order loads.

4.10.2.3.3. Platform Additional Stiffness and Damping

The vectors and matrices of this section are used to generate additional loads on the platform (in addition to other hydrodynamic terms calculated by HydroDyn), per the following equation.

(4.195)\[\overrightarrow{F}_{Add} = \overrightarrow{F}_{0} - [C] \overrightarrow{q} - [B] \dot{\overrightarrow{q}} - [B_{quad}] ABS \left(\dot{\overrightarrow{q}}\right) \dot{\overrightarrow{q}}\]

where \(\overrightarrow{F}_{0}\) corresponds to the AddF0 static load (preload) vector, \([C]\) corresponds to the AddCLin linear restoring (stiffness) matrix, \([B]\) corresponds to the AddBLin linear damping matrix, \([B_{quad}]\) corresponds to the AddBQuad quadratic drag matrix, and \(\overrightarrow{q}\) corresponds to the displacement vector of the potential-flow bodies (translation and rotation), where the overdot refers to the first time-derivative.

AddF0 is either a column vector with 6NBody entries if NBodyMod = 1 or NBody column vectors with six entries each if NBodyMod = 2 or 3. In the former case, AddF0 will span 6NBody lines with each line containing a single number in the input file. In the latter case, AddF0 will span six lines with each line containing NBody numbers in the input file.

AddCLin, AddBLin, and AddBQuad are either a single 6NBody-by-6NBody matrix if NBodyMod = 1 or six 6-by-6 matrices if NBodyMod = 2 or 3. In the former case, each matrix spans 6NBody lines in the input file with each line containing 6NBody numbers. In the latter case, each matrix spans six lines in the input file, with each line containing 6NBody numbers.

These terms can be used, e.g., to model a linearized mooring system, to augment strip-theory members with a linear hydrostatic restoring matrix (see Section 4.10.4.8.3), or to “tune” HydroDyn to match damping to experimental results, such as free-decay tests. While likely most useful for floating systems, these matrices can also be used for fixed-bottom systems; in both cases, the resulting load is applied at the reference point of each potential-flow body given by PtfmRefxt, PtfmRefyt, and PtfmRefzt.

4.10.2.3.4. Strip theory options

WaveDisp can be set to 0 to compute the strip-theory loads using the wave kinematics and dynamic pressure at the undisplaced position of the structure. If set to 1, the loads will be computed using the wave kinematics and dynamic pressure at the instantaneous displaced positions of the strip-theory members. Note that when wave stretching is not used (WaveStMod = 0 in SeaState), only the X- and Y-displacements of the strip-theory member nodes are considered when WaveDisp = 1, while the vertical Z-displacement is ignored. This is done to avoid discontinuous nodal loads that can result in unphysical structural vibration with a SubDyn substructure model. When WaveStMod > 0 and WaveDisp = 1, displacements of strip-theory members in all three directions are considered when computing the wave kinematics. A load smoothing procedure is performed to avoid discontinuous nodal loads in this case.

AMMod controls the computation of distributed strip-theory added-mass force. If AMMod = 0, the strip-theory added-mass force is always evaluated up to the SWL while neglecting the vertical displacement of the strip-theory member nodes, even if wave stretching is enabled. With AMMod = 1, the strip-theory added-mass force is evaluated up to the instantaneous free surface if WaveStMod > 0. The vertical displacement of strip-theory members will also be accounted for if WaveDisp = 1. AMMod should only be set to 0 if wave stretching is causing numerical instabilities with flexible fixed-bottom support structures modeled in SubDyn.

4.10.2.3.5. Axial Coefficients

This and the next several sections of the input file control the strip-theory model for both fixed-bottom and floating substructures.

HydroDyn computes lumped viscous-drag, added-mass, fluid-inertia, and static pressure loads at member ends (joints). The hydrodynamic coefficients for the lumped loads at joints are referred to as “axial coefficients” and include viscous-drag coefficients, AxCd, added-mass coefficients, AxCa, and dynamic-pressure coefficients, AxCp. AxCa influences both the added-mass loads and the scattering component of the fluid-inertia loads. Any number of separate axial coefficient sets, distinguished by AxCoefID, may be specified by setting NAxCoef > 1.

There are three optional inputs that affect the viscous drag force on endplates. These are AxFDMod, AxVnCOff, and AxFDLoFSc.

AxFDMod can be either 0 or 1. When set to 0, the drag force on endplates will be computed as in previous versions of OpenFAST. When set to 1, drag force will only be applied when the relative flow is directed away from the endplate where flow separation is expected, not when the relative flow is impinging on the endplate where flow separation is unlikely. Option 0 is suitable for strip-theory-only members, whereas option 1 might be better suited for hybrid potential-flow members with drag force. Note that option 1 uses a leading coefficient of 1/4 when computing the drag force, while option 2 uses the more common leading coefficient of 1/2 since drag is usually only applied to one of the two endplates of the member instead of on both.

AxVnCOff is the cutoff frequency in Hz for high-pass filtering the relative normal flow velocity used to compute the endplate drag force. This input parameter should be used together with the weighting factor AxFDLoFSc (between 0 and 1). When AxFDLoFSc = 0, the endplate drag force is computed purely based on the high-pass filtered relative normal velocity. When AxFDLoFSc = 1, the endplate drag force is computed purely based on the unfiltered relative normal velocity. This formulation is added to allow the user to attenuate the drag force in response to lower-frequency motion. In some cases, this approach can help address the underprediction of low-frequency resonance motion.

Users can opt to omit all three optional inputs. In this case, HydroDyn will compute the endplate drag force as in previous versions of OpenFAST. Alternatively, users can include only the optional parameter AxFDMod. No velocity filtering will be applied in this case. Lastly, users can include all three optional parameters to control the behavior of endplate drag force as explained above.

Axial viscous-drag loads will be calculated for all specified member joints. Axial added-mass, fluid-inertia, and static-pressure loads will only be calculated for member joints of members not modeled with potential flow (PropPot = FALSE). Axial loads are only calculated at user-specified joints. Axial loads are not calculated at joints HydroDyn may automatically create as part its solution process. For example, if you want axial effects at a marine-growth boundary (where HydroDyn automatically adds a joint), you must explicitly set a joint at that location.

4.10.2.3.6. Member Joints

The strip-theory model is based on a substructure composed of joints interconnected by members. NJoints is the user-specified number of joints and determines the number of rows in the subsequent table. Because a member connects two nodes, NJoints must be exactly zero or greater than or equal to two. Each joint listed in the table is identified by a unique integer, JointID. The (X,Y,Z) coordinate of each joint is specified in the global inertial-frame coordinate system via Jointxi, Jointyi, and Jointzi, respectively. JointAxID corresponds to an entry in the AXIAL COEFFICIENTS table and sets the axial coefficients for a joint. This version of HydroDyn cannot calculate joint overlap when multiple members meet at a common joint; therefore JointOvrlp must be set to 0. Future releases will enable joint overlap calculations.

Modeling a fixed-bottom substructure embedded into the seabed (e.g., through piles or suction buckets) requires that the lowest member joint(s) lie below the water depth. Placing a joint at or above the water depth results in static pressure loads being applied.

4.10.2.3.7. Member Cross-Sections

Members in HydroDyn are assumed to be straight circular (and possibly tapered) cylinders. Apart from the hydrodynamic coefficients, the circular cross-section properties needed for the hydrodynamic load calculations are member outer diameter, PropD, and member thickness, PropThck. You will need to create an entry in this table, distinguished by PropSetID, for each unique combination of these two properties. The member property-set table contains NPropSets rows. The member property sets are referred to by their PropSetID in the MEMBERS table, as described in Section 4.10.2.3.9 below. PropD determines the static buoyancy loads exterior to a member, as well as the area used in the viscous-drag calculation and the volume used in the added-mass and fluid-inertia calculations. PropThck determines the interior volume for fluid-filled (flooded/ballasted) members.

4.10.2.3.8. Hydrodynamic Coefficients

HydroDyn computes distributed viscous-drag, added-mass, fluid-inertia, and static buoyancy loads along members.

The hydrodynamic coefficients for the distributed strip-theory loads are specified using any of three models, which we refer to as the simple model, a depth-based model, and a member-based model. All of these models require the specification of both transverse and axial hydrodynamic coefficients for viscous drag, added mass, and dynamic pressure. The added-mass coefficient influences both the added-mass loads and the scattering component of the fluid-inertia loads. There are separate set of hydrodynamic coefficients both with and without marine growth. A given element will either use the marine growth or the standard version of a coefficient, but never both. Note that input members are split into elements, one of the splitting rules guarantees the previous statement is true. Which members have marine growth is defined by the MARINE GROWTH table of Section 4.10.2.3.11. You can specify only one model type, MCoefMod, for any given member in the MEMBERS table. However, different members can specify different coefficient models.

In the hydrodynamic coefficient input parameters, Cd, Ca, and Cp refer to the viscous-drag, added-mass, and dynamic-pressure coefficients, respectively. MG identifies the coefficients to be applied for members with marine growth (the standard values are identified without MG), and Ax identifies the axial coefficients to be applied for tapered members (the transverse coefficients are identified without Ax). The Cb coefficients allow the user to scale the hydrostatic load for, e.g., non-circular member cross sections. To avoid unphysical hydrostatic loads, the Cb coefficients are not used to directly scale the distributed hydrostatic load. Instead, the local member diameter (with marine growth if specified) is scaled by the square root of Cb when computing the hydrostatic load. This scaling also affects the hydrostatic load on member endplates for consistency.

While the strip-theory solution assumes circular cross sections, the hydrodynamic coefficients can include shape corrections; however, there is no distinction made in HydroDyn between different transverse directions.

4.10.2.3.8.1. Simple Model

This table consists of a single complete set of hydrodynamic coefficients as follows: SimplCd, SimplCdMG, SimplCa, SimplCaMG, SimplCp, SimplCpMG, SimplAxCa, SimplAxCaMG, SimplAxCp, and SimplAxCpMG. These hydrodynamic coefficients are referenced in the members table of Section 4.10.2.3.9 by selecting MCoefMod = 1.

4.10.2.3.8.2. Depth-Based Model

The depth-based coefficient model allows you to specify a series of depth-dependent coefficients. NCoefDpth is the user-specified number of depths and determines the number of rows in the subsequent table. Currently, this table requires that the rows are ordered by increasing depth, Dpth; this is equivalent to a decreasing global Z-coordinate. The hydrodynamic coefficients at each depth are as follows: DpthCd, DpthCdMG, DpthCa, DpthCaMG, DpthCp, DpthCpMG, DpthAxCa, DpthAxCaMG, DpthAxCp, and DpthAxCpMG. Members use these hydrodynamic coefficients by setting MCoefMod = 2. The HydroDyn module will interpolate coefficients for a node whose Z-coordinate lies between table Z-coordinates.

4.10.2.3.8.3. Member-Based Model

The member-based coefficient model allows you to specify a hydrodynamic coefficients for each particular member. NCoefMembers is the user-specified number of members with member-based coefficients and determines the number of rows in the subsequent table. The hydrodynamic coefficients for a member distinguished by MemberID are as follows: MemberCd1, MemberCd2, MemberCdMG1, MemberCdMG2, MemberCa1, MemberCa2, MemberCaMG1, MemberCaMG2, MemberCp1, MemberCp2, MemberCpMG1, MemberCpMG2, MemberAxCa1, MemberAxCa2, MemberAxCaMG1, MemberAxCaMG2, MemberAxCp1, MemberAxCp2, MemberAxCpMG1, and MemberAxCpMG2, where 1 and 2 identify the starting and ending joint of the member, respectively. Members use these hydrodynamic coefficients by setting MCoefMod = 3.

4.10.2.3.8.4. MacCamy-Fuchs diffraction load model

The MacCamy-Fuchs diffraction load model can be enabled for strip-theory members using any of the three coefficient models listed above. To enable the MacCamy-Fuchs model, all transverse Cp and CpMG coefficients should be replaced with the keyword MCF instead of a numeric value. For the simple model, this includes SimplCp and SimplCpMG. With the depth-based model, DpthCp and DpthCpMG on all lines should have the keyword MCF. Finally, for the member-based model, MemberCp1, MemberCp2, MemberCpMG1, and MemberCpMG2 should all have the keyword MCF only for the members to use the MacCamy-Fuchs model. All other coefficients can be specified as usual, including the added-mass coefficients. With this configuration, the distributed transverse fluid-inertia force on the members will simply follow the MacCamy-Fuchs diffraction load, irrespective of the added-mass coefficient set by the user. In this case, the added-mass coefficient only affects the force component proportional to the structure acceleration, not the force component proportional to the fluid acceleration.

Strictly speaking, the MacCamy-Fuchs diffraction solution only applies to fixed-bottom or deep-drafted vertical circular cylinders with a constant diameter. To ensure it is approximately applicable while still allowing for some flexibility, some constraints are placed on members when applying the MacCamy-Fuchs model:

  • The member must be surface-piercing at least when the structure is undisplaced in calm water.

  • The member must be nearly vertical with an inclination from vertical less than 10 deg.

  • The member can be tapered slightly, but the diameter must be within +/-10% of MCFD in the SeaState input file.

  • The member must have a draft at least as large as 0.5MCFD.

Because the MacCamy-Fuchs diffraction solution is based on linear potential-flow theory, second-order contributions to the fluid acceleration are neglected when computing the wave load even if second-order wave kinematics are enabled in SeaState. However, the MacCamy-Fuchs diffraction model can be used in conjunction with any of the available wave-stretching models.

4.10.2.3.9. Members

NMembers is the user-specified number of members and determines the number of rows in the subsequent table. For each member distinguished by MemberID, MJointID1 specifies the starting joint and MJointID2 specifies the ending joint, corresponding to an identifier (JointID) from the MEMBER JOINTS table. Likewise, MPropSetID1 corresponds to the starting cross-section properties and MProSetID2 specify the ending cross-section properties, allowing for tapered members. MDivSize determines the maximum spacing (in meters) between simulation nodes where the distributed loads are actually computed; the smaller the number, the finer the resolution and longer the computational time. Each member in your model will have hydrodynamic coefficients, which are specified using one of the three models (MCoefMod). Model 1 uses a single set of coefficients found in the SIMPLE HYDRODYNAMIC COEFFICIENTS section. Model 2 is depth-based, and is determined via the table found in the DEPTH-BASED HYDRODYNAMIC COEFFICIENTS section. Model 3 specifies coefficients for a particular member, by referring to the MEMBER-BASED HYDRODYNAMIC COEFFICIENTS section. The MHstLMod switch controls the computation of hydrostatic loads on strip-theory members when PropPot = FALSE. Setting MHstLMod to 0 disables hydrostatic load. If set to 1, hydrostatic loads will be computed analytically. This approach is efficient, but it only works for fully submerged or surface-piercing members that are far from horizontal without partially wetted endplates. For nearly horizontal members close to the free surface or members that experience partially wetted endplates, a semi-numerical approach for hydrostatic load can be selected by setting MHstLMod to 2. This approach works with any member positioning in relation to the free surface at the cost of slightly longer computing time. The PropPot flag indicates whether the corresponding member coincides with the body represented by the potential-flow solution. When PropPot = TRUE, only viscous-drag loads and ballasting loads will be computed for that member.

4.10.2.3.10. Filled Members

Members—whether they are also modeled with potential-flow or not—may be fluid-filled, meaning that they are flooded and/or ballasted. Fluid-filled members introduce interior buoyancy that subtracts from the exterior buoyancy and a mass. Both distributed loads along a member and lumped loads at joints are applied. The volume of fluid in the member is derived from the outer diameter and thickness of the member and a fluid-filled free-surface level. The fluid in the member is assumed to be compartmentalized such that it does not slosh. Rotational inertia of the fluid in the member is ignored. A member’s filled configuration is defined by the filled-fluid density and the free-surface level. Filled members that have the same configuration are collected into fill groups.

NFillGroups specifies the number of fluid-filled member groups and determines the number of rows in the subsequent table. FillNumM specifies the number of members in the fill group. FillMList is a list of FillNumM whitespace-separated MemberIDs. FillFSLoc specifies the Z-height of the free-surface (0 for MSL). FillDens is the density of the fluid. If FillDens = DEFAULT, then FillDens = WtrDens.

4.10.2.3.11. Marine Growth

Members not also modeled with potential-flow theory may be modeled with marine growth. Marine growth causes three effects. First, marine growth introduces a static weight and mass to a member, applied as distributed loads along the member. Second, marine growth increases the outer diameter of a member, which impacts the diameter used in the viscous-drag, added-mass, fluid-inertia, and static buoyancy load calculations. Third, the hydrodynamic coefficients for viscous drag, added mass, and dynamic pressure are specified distinctly for marine growth. Rotational inertia of the marine growth is ignored and marine growth is not added to member ends.

Marine growth is specified using a depth-based table with NMGDepths rows. This table must have exactly zero or at least 2 rows. The columns in the table include the local depth, MGDpth, the marine growth thickness, MGThck, and marine growth density, MGDens. Marine growth for a particular location in the substructure geometry is added by linearly interpolating between the marine-growth table entries. The smallest and largest values of MGDpth define the marine growth region. Outside this region the marine growth thickness is set to zero. If you want sub-regions of zero marine growth thickness within these bounds, you must generate depth entries which explicitly set MGThck to zero. The hydrodynamic coefficient tables contain coefficients with and without marine growth. If MGThck = 0 for a particular node, the coefficients not associated with marine growth are used.

4.10.2.3.12. Member Output List

HydroDyn can output distributed load and wave kinematic quantities at up to 9 locations on up to 9 different members, for a total of 81 possible local member output locations. NMOutputs specifies the number of members. You must create a table entry for each requested member. Within a table entry, MemberID is the ID specified in the MEMBERS table, and NOutLoc specifies how many output locations are generated for this member. NodeLocs specifies those locations as a normalized distance from the starting joint (0.0) to the ending joint (1.0) of the member. If the chosen location does not align with a calculation node, the results at the two surrounding nodes will be linearly interpolated. The outputs specified in Appendix C. List of Output Channels determines which quantities are actually output at these locations.

4.10.2.3.13. Joint Output List

HydroDyn can output lumped load and wave kinematic quantities at up to 9 different joints. JOutLst contains a list of NJOutputs number of JointIDs. The outputs specified in Appendix C. List of Output Channels determines which quantities are actually output at these joints.

4.10.2.3.14. Output

Specifying HDSum = TRUE causes HydroDyn to generate a summary file with name OutRootname.HD.sum. OutRootName is either specified in the HYDRODYN section of the driver input file when running HydroDyn standalone, or by the OpenFAST program when running a coupled simulation. See Section 4.10.3.2 for summary file details.

For this version, OutAll must be set to FALSE. In future versions, setting OutAll = TRUE will cause HydroDyn to auto-generate outputs for every joint and member in the input file.

If OutSwtch is set to 1, outputs are sent to a file with the name OutRootname.HD.out. If OutSwtch is set to 2, outputs are sent to the calling program (OpenFAST) for writing. If OutSwtch is set to 3, both file outputs occur. In standalone mode, setting OutSwitch to 2 results in no output file being produced.

The OutFmt and OutSFmt parameters control the formatting for the output data and the channel headers, respectively. HydroDyn currently does not check the validity of these format strings. They need to be valid Fortran format strings. Since the OutSFmt is used for the column header and OutFmt is for the channel data, in order for the headers and channel data to align properly, the width specification should match. For example,

"ES11.4" OutFmt
"A11" OutSFmt

4.10.2.3.15. Output Channels

This section controls output quantities generated by HydroDyn. Enter one or more lines containing quoted strings that in turn contain one or more output parameter names. Separate output parameter names by any combination of commas, semicolons, spaces, and/or tabs. If you prefix a parameter name with a minus sign, “-“, underscore, “_”, or the characters “m” or “M”, HydroDyn will multiply the value for that channel by –1 before writing the data. The parameters are not necessarily written in the order they are listed in the input file. HydroDyn allows you to use multiple lines so that you can break your list into meaningful groups and so the lines can be shorter. You may enter comments after the closing quote on any of the lines. Entering a line with the string “END” at the beginning of the line or at the beginning of a quoted string found at the beginning of the line will cause HydroDyn to quit scanning for more lines of channel names. Member- and joint-related quantities are generated for the requested Member Output List and Joint Output List. If HydroDyn encounters an unknown/invalid channel name, it warns the users but will remove the suspect channel from the output file. Please refer to Appendix C for a complete list of possible output parameters.