Recovery of Classical Solution for the Homogeneous Dirichlet Problem

104 Views Asked by At

In section 9.5 of Brezis' text, Brezis goes through the example of the homogeneous Dirichlet problem. My issue is with the final step where he recovers a classical solutions from the weak solutions. The argument is as follows:

Assume that the weak solution $u\in H^{1}_{0}(\Omega)$ belongs to $C^{2}(\overline{\Omega})$, and assume that $\Omega$ is of class $C^{1}$. Then $u=0$ on $\partial\Omega$ (by Theorem 9.17). On the other hand, we have, \begin{align} \int_{\Omega}(-\Delta u+u)v=\int_{\Omega}fv\quad\forall v\in C_{c}^{1}(\Omega), \end{align} and thus $-\Delta u+u=f$ a.e. on $\Omega$. In fact, $-\Delta u+u=f$ everywhere on $\Omega$, since $u\in C^{2}(\Omega)$; thus $u$ is a classical solution.

Can someone explain the reasoning here. What is the logical process from initially assuming $u\in C^{2}(\overline{\Omega})$ and $\Omega$ being of class $C^{1}$? How does this assumption, implying $u=0$ on $\partial\Omega$, allow us to conclude that the weak solution is in fact zero on $\partial\Omega$?

For the second part why do we not require $f$ to be continuous in order to conclude $-\Delta u+u=f$ everywhere on $\Omega$?

Theorem 9.17: Suppose $\Omega$ is of class $C^{1}$. Let, \begin{align} u\in W^{1,\,p}(\Omega)\cap C(\overline{\Omega})\quad\text{with }1\leq p<\infty. \end{align} Then the following properties are equivalent: \begin{align} \text{(i)}&\,u=0\text{ on }\partial\Omega,\\ \text{(ii)}&\, u\in W^{1,\,p}_{0}(\Omega). \end{align}

1

There are 1 best solutions below

0
On

(1)

We are assuming $u$ is a weak solution (of the homogeneous Dirichlet problem), $u\in C^2(\bar{\Omega})$, and $\Omega$ is of $C^1$ class.

Since $u\in H^1_0(\Omega)\cap C(\bar{\Omega})$ and $\Omega$ is of class $C^1$, Theorem 9.17 implies $u$, the weak solution, vanishes on $\partial \Omega$.

(2)

The assumption does not require $f$ to be continuous, but it turns out (in this case) $f$ can be redefined (on a null set) as a continuous function:

$$-\Delta u+u=f$$ a.e., which means $f$ can be redefined (on a null set) so that the equation holds everywhere. Then $f$ has to be continuous since the left hand side is continuous.