Change in solution with time step

7 weeks ago by
Hello, I'm using the advection-diffusion-reaction code from the fenics tutorial as an example, but I'm finding my solution to change with the time step without converging to constant value even at really small time steps (I've made some changes to the code). I'm using the formula (F = ((u_1 - u_n1) / k)*v_1*dx(0) + eps*dot(grad(u_1), grad(v_1))*dx(0)). The top curve in the figure below is for 50000 time steps and the bottom curve is for 10000 time steps (both between 0 and 5 seconds). I can post the code if that would make it easier

Community: FEniCS Project
Please login to add an answer/comment or follow this question.

Similar posts:
Search »