[Binoy]

Hello Everyone,

I am working on modeling offshore submarine slope in CFX. I am trying to validate a model by using CFX which is previously done by finite element method. A soft soil block sliding along an inclined rigid seabed. The rigid base is fixed with an inclination of 5 degree. The soil block is initially stationary and begins to slide due to self weight. I am trying to compare the soil block's front toe velocity and run-out distance of front toe with time.

For my case in CFX I gave no slip boundary condition at bottom layer of domain to represent the rigid seabed. After simulation I got the results that are not consistent with the FE model.

In FE model they found maximum runout distance is 22 m at 10s, maximum velocity is 4 m/s and slide becomes stationary at 10s.
But in CFX I got maximum runout distance is 20 to 22m at 20 s, maximum velocity is 1.5 m/s. And slide becomes stationary after 25s.
It seems runout distance is almost close to the FE model but velocity remains lower than FE model. I also tried with fine mesh at the bottom boundary, but it changes max velocity up to 1.9 m/s.

I tried by many ways but every-time i got lower velocity than FE model.

Here I have attached an image and output file of the model.

I am sorry for my grammatical and spelling mistakes.

I hope someone can help me to figure out the problem.

Thanks in Advance.

Best regards,
Binoy

[ghorrocks]

* You appear to be using a continuous fluid model for the clay in a multiphase model.
* You have no model for clay maximum packing, siltation, etc
* The clay strain rate is set to a discontinuous and apparent arbitrary function
* You have defined an apparent arbitrary time step size, and presumably mesh size and convergence tolerance

All these points are assumptions you have made which appear dubious to me.

[Binoy]

Thanks Glenn for your reply.

Can't I use a continuous fluid model for the clay block in a multiphase model?
For the modelling of clay maximum packing I have to use dispersed solid or dispersed fluid to model the clay block. I will try to do by this way.

Clay shear strain rate:
One of my previous model I faced this problem also and you suggested to me to use tanh function to define the clay shear strain rate. I am still struggling with this and couldn't figure out how to use the function. Also I don't know the limit of minimum and maximum shear strain rate of clay for this particular problem, it varies with different problem, I guess. Please give me suggestion.

Time step size, mesh size and convergence tolerance:
Please give me suggestion about the using of time step size. Using which relation I need to use time step size. I know one relation which is related with maximum node velocity and minimum element size. But for this model I know only minimum element size. And also give me suggestion about the mesh size and convergence tolerance.

Once again, thank you for your time.
Binoy

[ghorrocks]

Can't I use a continuous fluid model for the clay block in a multiphase model?

You can use anything you like. The question is - is it accurate enough for your purposes? Only you can answer that question. But I would have thought that using a fluid model for something which is not a fluid is unlikely to give useful results. I am no expert on the movement of clay in water, but a eularian multiphase model does not sound right to me.

Please give me suggestion about the using of time step size.

Unless you have a good reason not to, use adaptive time steps, homing in on 3-5 coefficient loops per iteration. Then CFX will reliably find its own time step size.

Mesh Size: Do a sensitivity analysis
Convergence: Do a sensitivity analysis.