Hi dears

I am simulating the interaction of water with a concrete pool in Eulerian Abaqus (v-2021).

-Mechanical behaviour of concrete is CDP and was used Eos (Us-Up) for water.

-The solver is the Explicit type and the acceleration of the Northridghe earthquake is applied to the bottom of the pool.

Even though the weight of water and pool is not taken into account, in the process of solving the problem, there are severe disturbances in the boundry force in the vertical direction, which will cause the concrete to crack, while the acceleration of the earthquake is in the horizontal direction. If weight is considered, this situation is more critical.

Also, the acceleration at the pool wall reaches more than 10,000 m/s2, although the applied value at the foundation of the pool is 6 m/s2.

But when I remove the Euler elements and attach the water mass to the pool (non-structural mass), the result makes sense.

Does this method in Abaqus have the ability to correctly predict the interaction of water and structure, or does the simulated model need to be modified?

Thank you