Multidirectional waves in Hydrodyn

Wind & Water Computer-Aided Engineering Software Tools
1

Hello @Jason.Jonkman,

I am trying to simulate the 5 MW barge-type FOWT in the case of multidirectional waves. Please I would like whether you can validate the following statements to see if i understand the functionality of multidirectional waves in OpenFAST.

1- I found that WaveDirRange in HydrDyn input cannot support angle greater than 360 degrees (a
full circle).

2- The directional spectrum used in HydroDyn module of OpenFAST is the same stated in the following
paper:
https://www.nrel.gov/docs/fy14osti/61161.pdf

3- The value of wave spreading parameter s always describes short-crested wave except for s=0 ? In other words, could you please provide any paper which gives the meaning of the values that s
can take ?

4- I simulate the 5 MW barge-type FOWT for WaveDirRange equals 360 degrees and a number of wave directions equals 9. I notice that HydroDyn increases the number of wave direction to 11. It said that the number of wave direction was increased to 11 to obtain an equal number of frequencies in each direction which is normal. Right ? The WaveNDir should be always an odd number. Right ? That’s why the FOWT when subjected to WaveDirRange = 360 degrees, the total hydrodynamic load is not zero, in contrast for the case where WaveNDir is even which in this case, the total hydrodynamic load is zero (except for load in heave direction). Right ?

Thank you in advance

Best Regards,

Riad

2

Dear @Riad.Elhamoud,

Here are my responses:

  1. Correct.
  2. Correct.
  3. I’m not aware of an NREL publication stating physically relevant ranges for s, but a cosine spreading function is quite common in the offshore industry, so, I would hope you could find typical values of s in the literature.
  4. WaveNDir must be an odd number. HydroDyn may internally increase WaveNDir by a bit to ensure that the FFTs used within HydroDyn are computationally efficient. But I’m not sure I understand the second-half of your question.

Best regards,

1 Like
3

Dear @Jason.Jonkman,

Thank you very for your reply. I can now say that i understand the multidirectional waves functionality in HydroDyn :grin:

Concerning my last question, i was saying that if OpenFAST allows an even number of wave direction and i set WaveDirRange to 360 degrees, this will not induce a net hydrodynamic load equals zero due to the fact that the wave directions are equally spaced and the loads induced by two opposite directions cancel out ? What do you think ?

Thank you in advance.

Best Regards,

Riad

4

Dear @Riad.Elhamoud

I suppose you are assuming a uniform distribution of wave directions rather than a cosine distribution, where the energy is more concentrated around WaveDir. Regardless, HydroDyn does not permit WaveNDir to be an even number. Moreover, the opposing waves from opposite directions will not cancel out regardless because the frequency components are distributed between the wave direction bins rather than having each direction bin using the same frequency components. That is, a wave propagating along a given directly will never be identical to a wave propagating from the opposite direction.

Best regards,

2 Likes
5

Dear @Jason.Jonkman

Thank you so much for your explanation.

I was computing the value of the normalizing constant called C used in the directional spectrum in the paper:

I found that C is greater than 1 when theta is set to the mean wave direction (i.e. theta = theta_bar), theta_max is set to pi over 2 and the spreading coefficient s is set to 10. For me, it’s not a logic value because the directional spectrum computed at mean wave direction should be a fraction of the frequency spectrum and i expect that C should be lower than 1.

The calculation of C was done in MATLAB. Here is a screenshot of the command window.

Capture_1

Do you think that i did a mistake ?

Thank you in advance.

Best Regards,

Riad

6

Dear @Riad.Elhamoud,

That looks correct to me. The normalizing parameter (C) does not have to be less than unity; it is the cosine spreading function (D) that integrates to unity when integrating over wave directions.

Best regards,

1 Like