1

I am now solving a Schrodinger equation with a magnetic field. I don't know its exact solution. I am trying to test the convergence of the scheme by comparing the difference of numerical solution when using different mesh-sizes.

However, in the 2D case, the solution is stored in a matrix; when I change the number of nodal points $n$ ($dx=(x_{max}-x_{min})/(n-1)$), the size of the matrix changes.

Both the $L^2$ norm and $\sup{}$-norm require the same dimensions.

In this case, how can I measure the difference of the solutions?

ekkilop
  • 2,278
Haiyi Tan
  • 85

2 Answers2

3

Your simplest course of action is to directly obtain approximations for the solution of the PDE at exactly the same grid points.

This can be done be repeatedly refining an existing grid. The simplest method is to systematically cut the step size in half.

To be specific, suppose your 2D domain is the unit square and the initial step size is $h_0 = \frac{1}{2}$ in either direction. Your approximation can be stored in a matrix $A_0$ of dimension 3. Now reduce the step size to $h_1 = \frac{1}{2}h_0 = \frac{1}{4}$. You now require a matrix $A_1$ of dimension $5$. However, the submatrix $B_1$ of dimension $3$ obtained by deleting the second and fourth columns of $A_1$ contains approximations for the same grid points, as $A_0$. In short, it is meaningful to examine $B_1 - A_0$.

If you are working with MATLAB, then you can construct $B_1$ using as $$B_1 = A_1(1:2:5,1:2:5).$$

Carl Christian
  • 12,583
  • 1
  • 14
  • 37
  • I see. But when I calculate the error of B1 and A0, (I used L2 norm), what meshsize should I take? h1 or h0? – Haiyi Tan Jul 21 '17 at 03:18
  • @HaiyiTan: $h_0$. When discarding elements of $A_1$ you are restricting your fine grid function to the coarse grid with stepsize $h_0$. – Carl Christian Jul 21 '17 at 06:10
  • ok. thank you so much! – Haiyi Tan Jul 21 '17 at 12:34
  • 1
    @HaiyiTan: You are very welcome. Ekkilop describes how to construct test problems with known solutions. When you want to compute reliable error estimates for an unknown solution, then Richardson extrapolation using a sequence of grids is often a good option. – Carl Christian Jul 21 '17 at 13:30
  • what would you do in the case where you cannot restrict your fine grid to the coarser one? For example, I have two solutions obtained from different methods that are vectors of size $1609$ and $257$. Does it make sense to multiply the norm of the finer grid by $\sqrt(257)/\sqrt(1609)$? – Mjoseph May 09 '22 at 20:41
  • 1
    @Mjoseph I would interpolate on the fine grid using an interpolation method that has the exactly the same order of accuracy as the method used to generate the approximations in the first place. In this manner, I could obtain different approximations for the target value at each of the coarse grid points and Richardson extrapolation would have a chance. – Carl Christian May 10 '22 at 06:18
  • @Mjoseph I did not understand you second question about the norm of the grid. Please add a few more words. – Carl Christian May 10 '22 at 06:20
  • Thank you for responding so far after the initial question! At risk of over explaining in the comments section, I have posted a new question for my specific problem. If you have the time, the link is right below - thanks! https://math.stackexchange.com/questions/4448178/convergence-check-for-2d-problem-solved-numerically-using-different-methods – Mjoseph May 11 '22 at 14:54
2

When we want to test the convergence of a numerical scheme without knowledge of an exact solution, we can use the method of manufactured solutions.

Say that the PDE we want to solve can be written as follows: $$ u_t + P(u) = F(x,t), \\ u(x,0) = f(x), \\ u(0,t) = g(t), $$ where $P$ is some differential operator, and $F(x,t)$, $f(x)$ and $g(t)$ are known functions (data). For the purpose of testing convergence of a numerical scheme, simply choose any (appropriately differentiable) function $v$ that you want to be your solution, and operate on it with $\partial_t + P$. In general, the result will be $$ v_t + P(v) = G(x,t) \neq F(x,t). $$ However, this doesn't matter if the goal is to study the convergence of the numerical scheme, which will be the same whether we use $F$ or $G$ (so long as both $F$ and $G$ are sufficiently smooth). We may therefore replace our original PDE with $$ u_t + P(u) = G(x,t), \\ u(x,0) = \tilde{f}(x) ,\\ u(0,t) = \tilde{g}(t), $$ where $\tilde{f}(x) = v(x,0)$ and $\tilde{g}(t) = v(0,t)$. Since $v$ is known (after all, we have chosen it to be anything we want), we have access to the exact solution to this new PDE, hence we may test the convergence of the numerical scheme as usual.

ekkilop
  • 2,278