Heat Flux in 1D-Heat Conduction

Usability
1

Dear OGS-Community,

I want to simulate the temperature distribution in the subsurface at different horizontal heights, to get the undisturbed temperature field. For this problem a line in vertical direction should be enough. So I define a temperature curve as Dirichlet BC for the top point and a terrestrial heat flux for the bottom point as Neumann BC. The simulation runs, but the results look not as expected. I want the heat flux only in the z-direction, but it seems that the heat flux go away to the x- and y- direction. Is it possible, to define that the x- and y- component of the heat flux should be zero at every point?

HeatFlow_1D_Heat_Conduction
HeatFlow_1D_Heat_Conduction2307×1514 261 KB

Another way to solve the problem is the usage of a large domain with 2D axially_symmetric mesh, but this is not so elegant and computationally fast as solving only on a line.

The input data for the line example can be download here: GigaMove

Best regards,
Max

2

I found my error in the meantime, I was a bit confused yesterday about the output of heat_flux and HeatFlowRate. Sorry if someone has thought about my problem.

My expected result is to get a geothermal gradient \Delta T according to \dot{q}=\lambda \cdot \Delta T. My initial simulation above runs only for 25 years and the equilibrium is not reached in this time - with a simulation time of 100 years it looks better, but with the currently initial condition its need longer then the 100 years to reach the equilibrium.