IEA 15 MW RWT on UMaine semi-sub floater-Total Mass Increase

Wind & Water Computer-Aided Engineering Software Tools
1

Dear all,

I am trying to model a localised H2 production plant on the UMaine Volturn US-S semi-sub floater, coupled with the IEA 15 MW RWT. All components have an extra mass that needs to be added on the floater, which is approximately 220 t.

However, the change in the total floater mass will result in changing various parameters of the model, including: (i) Platform mass and roll,pitch and yaw inertias, (ii) Platform centre of gravity and centre of buoyancy, (iii) Volume of water displaced by the platform and (iv) eventually the WAMIT potential flow solution.

I was wondering whether such a mass increase of the IEA 15 MW RWT/UMaine semi-sub floater can be indeed modelled in OpenFAST and if there is a way of simplyfing the modifications needed for this action.

Perhaps, could I consider the added mass as an additional platform ballast?In this case, what are the modifications that I should perform on the existing model?

Best regards,
Ioannis Voultsos.

2

Dear @Ioannis.Voultsos,

Certainly such a mass increase can be modeled in OpenFAST. With some work, you can certainly change all of the parameters you mention (items (i) through (iv)). I’m sure there are simpler ways too. The best approach will depend on what level of accuracy you want. But I can’t really provide specific recommendations without knowing more.

Best regards,

3

Dear Jason,

Thank you for your prompt answer.

In my thesis project, I am mainly interested on the effect of the floater motion to the power production of the turbine.

When the added mass is considered, my goal is to identify changes in the power production characteristics of the IEA 15 MW RWT due to the increased mass. I am thinking of comparing the power curves of the FOWT with the added weight against the original FOWT design. I will not be interested in the dynamic behavior of the floater by means of studying e.g. the natural frequency change of the platform 6 DoFs.

Therefore, I am inclined to believe that the mass increase should be done in a way, that would require as less modifications in the floater design as possible. This is due to the fact that, not only I am not studying the floater behavior itself, but also, considerable modifications should be done (e.g. recalculating the potential flow solution around the platform), which may be impractical to do, owing mainly to time constraints.

I hope that my response clarifies your thought about the issue. I am available for further explanations on my project scope.

Best regards,
Ioannis Voultsos.

4

Dear @Ioannis.Voultsos,

Well, the simplest approach for accounting for the addition mass in the platform is to modify the platform mass in ElastoDyn (PtfmMass), including any relevant changes to the platform center of mass and inertias. To ensure that the floater remains in equilibrium in heave when you change PtfmMass, you should also change the undisplaced volume in HydroDyn (PtfmVol0) accordingly, i.e.:

PtfmMass = PtfmMass + deltaMass
PtfmVol0 = PtfmVol0 + deltaVol0

where,

deltaVol0 = deltaMass/WtrDens

Please note that this change will only account for the change in mass; the volume change is needed to ensure heave equilibrium. In reality, the volume change would mean a geometry change that should involve changes to the potential-flow and strip-theory solutions of HydroDyn. But changing those requires more work. It is difficult to say up front which changes will most influence the turbine power production.

Best regards,

5

Dear Jason,

Thank you for your answer. Steps to increase the floater mass are clear.

However, I have trouble in understanding how to incorporate the change of the platform inertias. From the documentation of the UMaine Volturn US-S semi-sub floater, the total FOWT inertia in respect to roll,pitch and yaw axes are presented, without an explanation of the way it are calculated. In addition, since the semi-sub floater is a complicated structure, it is difficult to determine how moments of inertia are calculated.

Therefore, due to the floater added mass, not only the floater inertia at all three axes will increase, but since the floater centre of mass will change, it will involve a change in the default floater inertia.

Are there any recommendations on this issue?

Best regards,
Ioannis Voultsos.

6

Dear @Ioannis.Voultsos.

I’m not familiar with how these properties were calculated for the UMaine VolturnUS semisubmersible. Regardless, if you know the center of mass location of your additional mass, you should be able to calculate the new total mass and new center of mass of the augmented floater. You can then use the parallel axis theorem to transfer the inertias to the new center of mass location.

Best regards,

7

Dear Jason,

Thank you for your answer. I have some follow-up questions.

More specifically, in the process of finding the new floater center of mass (CM), my idea is to use the original total floater mass (17,839 t) and the original CM, since the report does not provide detailed information on the CM of each component, comprising the floater. Then, the new CM will be defined as:

CM,new= M,tot,old/M,tot,new * z,cm,old + Σ_i M,added(i)/Mtot,new *z(i) .

where M,tot,old and M,tot,new is the old and new total floater mass respectively, z,cm,old the vertical position of the old CM and z(i) the vertical position of the added mass.

However, due to the extra mass, the floater draft is going to increase from its original value of 20 m, therefore moving the floater initial CM to a more negative value. Therefore, I have to calculate the new draft. For this reason, I am trying to replicate the UMaine Volturn US-S semisubmersible platform numerical results, presented in its documentation.

According to my calculations, in case that the whole turbine+floater mass is considered (20,093 t), plus the weight of the mooring system (or vertical pretension force, according to the floater documentation), then the calculated displaced water volume is the one stated in the paper (20,206 m3). So, this is the process for calculating the new displaced water volume

However, then, I have trouble in calculating the draft. When I divide this value with the total area of the floater (3 pontoons+4 columns), I have a draft of 10.3 m, which is smaller than 20 m. Then, since the pontoons are flooded with water and thus,they do not contribute to buoyancy, in case I use only the area of the 4 columns, using a similar process as before, the calculated draft is about 50 m, which is higher than 20 m.

To conclude:

1.) Do you think that my rationale is correct?
2.) If yes, do you have any indication of which factor I do not consider in my calculations?
3.) Do you have a hint on what does it happen in the case I substitute the extra platform weight with less water ballast?

Best regards,
Ioannis Voultsos.

8

Dear @Ioannis.Voultsos,

I agree with your calculation of the new floater center of mass. And I agree that if you don’t compensate this additional mass by increasing the floater volume or decreasing the ballast, that the system will heave.

I do not understand how you are calculating draft.

When pontoons are flooded, we normally consider the pontoons has having both an external buoyancy (based on the displaced volume) and an internal mass, which adds weight and will counteract the external buoyancy. (The latter can be modeled in HydroDyn via flooded members, but I don’t see that this modeling approach was used in the UMaine semisubmersible.)

Best regards,

9

Dear Jason,

Thank you for your answer. I was able to replicate the draft results of the UMaine floater using some assumptions, occuring from the model nature of the floater.

When adding the extra mass on the floater without reconfiguring it (eg. its ballast weight etc.), the augmented floater mass increases by 1.05 %, the platform center of mass by 0.93% (more negative), whereas the augmented inertias in respect to roll and pitch axes by 1.63%. Therefore, the mass/inertia variation of the augmented floater can be thought as comparatively small.

From your experience, do you think that the credibility of the simlulations will be compromised when using the original WAMIT and Cummins equation matrices for the augmented floater?

Best regards,
Ioannis Voultsos.

10

Dear @Ioannis.Voultsos,

The first-order solution from WAMIT needed by OpenFAST (hydrostatics, wave radiation, and wave excitation expressed as matrices/vectors) only depends on the external shape (wetted surface) of the floater, not on its mass distribution, so, I would not expect any compromise with the first-order solution.

Please note, though, that the second-order WAMIT solution (QTFs), if used, does depend on the first-order motion response, and so, depends on the mass distribution.

Best regards,