Hi all, I want to know how to add initial stress under the consideration of a THM process. I search all the benchmarks about ThermoHydroMechanics project files but I can’t see any useful codes to imitate. I do notice that there is a child parameter named initial_stress that can be added, but there is no end-to-end test cases. I read some other codes which add initial stress in other process like mechanic and Thermomechanics. I use the codes in THM process but it doesn’t work.Can anyone provide me some examples or the correct codes?

Hi @tjlyh777

it should work as in this file for the TRM process:

So inital_stress defines a parameter that is define in line 203 ff.

Hi, @joergbuchwald

I have just found the code today,too and I use ogs6.4.0 to run the project. it works fine. Thank you!

Hi, @joergbuchwald

I have another question about the initial stress. I think the four expression of sigma0 sigma1 sigma2 and sigma3 represent the stress_xx, stress_yy, stress_xy and maximum principle stress respectively, but the results of my simulation is not pleasant. Am I wrong with the definition of the initial stress? What is the meaning of the sigma 0,1,2 and 3 in my result vtu file?

If you are in 2D, I guess it is sigma_xx, sigma_yy, sigma_zz, sigma_xy.

In 3D, (and this is very confusing as not standard notation): sigma_xx, sigma_yy, sigma_zz, sigma_xy, sigma_yz, sigma_zx.

But I have to check that every time my self again. This should be better documented, I agree. Another thing to keep in mind is that the stress always refers to the effective stress.

I got it. Thank you! By the way, I am a little confused about the sigma_xy and sigma_zz in 2D model. If my model is bulit in xy plane and I consider the gravity in the y direction, I guess this is correct, right? In this case, why does there occurs sigma_zz? Can I bulid the model in xz plane and consider gravity in z direction so that the initial vertical stress is written in the third term as an express ion of Z? If this is feasible, is the last sigma still sigma_xy? Maybe iy should be sigma_zx?

The z-direction is the direction with infinite extension. So, everything has to be symmetric in that sense. (g along z would break that). However, it is important for the calculation of pressure/stress. E.g., if you have a strain somewhere in the xy-plane, you’ll also have also a coupling along the z-direction due to the poisson’s ratio. That’s why there is also a z-component. However, the non-diagonal elements related to the z-direction vanish.