Results of the LUBBY2 problem

Usability
1

Hallo everyone,

after solving the LUBBY2 problem, i plotted the results (displacement versus time) usng ParaView. I noticed that the results i got show a linear relation between the two aforementioned variables. Whereas the results on the OGS website show a nonlinear relation. Can somebody help me?

Thank you in advance

image
image2353×819 150 KB

image
image1119×493 70.5 KB

2

Hi Mohamad,

I guess you are referring to this test case with this prj file

In the test OGS first makes one time step of 0.0001 s and afterwards several 0.01 s timesteps:

                <time_stepping>
                    <type>FixedTimeStepping</type>
                    <t_initial>0</t_initial>
                    <t_end>1</t_end>
                    <timesteps>
                        <pair>
                            <repeat>1</repeat>
                            <delta_t>0.0001</delta_t>
                        </pair>
                        <pair>
                            <repeat>100</repeat>
                            <delta_t>0.01</delta_t>
                        </pair>
                    </timesteps>
                </time_stepping>

However, OGS only outputs the initial condition, the first timestep and the last timestep:

        <output>
            <type>VTK</type>
            <prefix>cube_1e3</prefix>
            <timesteps>
                <pair>
                    <repeat>1</repeat>
                    <each_steps>1</each_steps>
                </pair>
                <pair>
                    <repeat>1</repeat>
                    <each_steps>10000000</each_steps>
                </pair>
            </timesteps>
            ...
        </output>

If you change the output configuration to the following you should see the desired nonlinear curve:

        <output>
            <type>VTK</type>
            <prefix>cube_1e3</prefix>
            <timesteps>
                <pair>
                    <repeat>1000</repeat>
                    <each_steps>1</each_steps>
                </pair>
            </timesteps>
            ...
        </output>

Since our test cases are run very often during the OGS development we typically reduce them to the bare minimum, e.g. by restriciting output or limiting the simulation time.