Questions about proof of theorem

In summary, we discussed the exterior sphere condition of a space $\Omega$ at a point $x_0 \in \partial{\Omega}$, which is satisfied if there is a point $y \notin \overline{\Omega}$ and a number $R>0$ such that $\overline{\Omega} \cap \overline{B_y(R)}=\{ x_0 \}$. We also defined the set $F_{\phi}$ containing subharmonic functions in $\Omega$ with boundary values less than or equal to $\phi \in C^0(\partial{\Omega})$. Then, we proved a theorem stating that if $u(x)=\sup_{v \in F_{\phi}} v(x
  • #1
evinda
Gold Member
MHB
3,836
0
Hello! (Wave)

We say that the space $\Omega$ satisfies the exterior sphere condition at the point $x_0 \in \partial{\Omega}$ if there is a $y \notin \overline{\Omega}$ and a number $R>0$ such that $\overline{\Omega} \cap \overline{B_y(R)}=\{ x_0 \}$.

Let the function $\phi \in C^0(\partial{\Omega})$. Let $F_{\phi}$ the set that contains the " under functions of $\phi$ in $\Omega$ ", i.e. the functions $v \in C^0(\overline{\Omega})$ that are subharmonic in $\Omega$ and in $\partial{\Omega}$ it holds that $v|_{\partial{\Omega}} \leq \phi$. (We say that the function $v$ is subharmonic in $\Omega$ if for each ball $B \subset \Omega$ it holds that $v \leq H_B[v]$, where $H_B[v](x)=\left\{\begin{matrix}
\text{ harmonic for } x \in B \\
v(x) \text{ for } x \in \Omega \setminus{B}
\end{matrix}\right.$)
Theorem: If $u(x)=\sup_{v \in F_{\phi}} v(x)$ and at the point $x_0$ the space $\Omega$ satisfies the exterior sphere condition, then $\lim_{x \to x_0} u(x)=\phi(x_0), x \in \Omega$.

Proof: Let $n \geq 3$. We define the function $h(x)=\frac{1}{R^{n-2}}-\frac{1}{|x-y|^{n-2}}$.

Obviously $h(x_0)=0$ and $h(x)>0$ in $\overline{\Omega} \setminus{\{ x_0\}}$. Since $\phi$ is continuous we have that $\forall \epsilon>0 \ \exists \delta>0$ such that

$|\phi(x)-\phi(x_0)|< \epsilon$ for $|x-x_0|< \delta$.

Also there is a constant $C>0$ such that $Ch(x)>2M$ for $|x-x_0| \geq \delta$ since $h(x)>0$ in $\overline{\Omega} \setminus{ \{ x_0 \}}$, here $M=\sup \phi$.

  • Could you explain to me why there is a $C$ such that $Ch(x)>2M$ ?

In order to prove the theorem it suffices to show that $|u(x)-\phi(x_0)| \leq \epsilon+ Ch(x)$ since $Ch(x) \to 0$ while $x \to x_0$.

So we want the folllowing inequality to hold.

$\phi(x_0)-\epsilon-Ch(x) \leq u(x) \leq \phi(x_0)+ \epsilon+ Ch(x), \forall x \in \Omega$The first inequality follows from the fact that $\phi(x_0)- \epsilon-Ch(x) $ is an under function.
Indeed, this function is harmonic and so also subharmonic and $\phi(x_0)-\epsilon-C h(x)|_{\partial{\Omega}} \leq \phi(x), \forall x \in \partial{\Omega}$.

  • I tried to show that the function is harmonic, but I got that $\Delta h=2(n-2) \frac{n-2(n-3) |x-y|}{|x-y|^{n-3}}$. Have I done something wrong? Also how do we get the inequality?
    Don't we have for $|x-x_0| \leq \delta$ that $\phi(x_0)- \epsilon> \phi(x)$ ? Or am I wrong? (Sweating)
The second inequality holds since $\phi(x_0)+ \epsilon+C h(x)$ is harmonic and $u|_{\partial{\Omega}} \leq \sup \phi \leq \phi(x_0)+ Ch(x)|_{\partial{\Omega}}$.
 
Physics news on Phys.org
  • #2
evinda said:
Also there is a constant $C>0$ such that $Ch(x)>2M$ for $|x-x_0| \geq \delta$ since $h(x)>0$ in $\overline{\Omega} \setminus{ \{ x_0 \}}$, here $M=\sup \phi$.

  • Could you explain to me why there is a $C$ such that $Ch(x)>2M$ ?

Hey evinda! (Smile)

Can't we pick for instance:
$$C = \frac{2M + 1}{\inf_{\xi \in\overline{\Omega} \setminus{ \{ x_0 \}}} h(\xi)}$$
(Wondering)
 

FAQ: Questions about proof of theorem

What is a theorem?

A theorem is a statement that has been proven to be true using logical reasoning and previously established facts or axioms. It is a fundamental concept in mathematics and science, and it is used to explain and predict various phenomena.

What is the process of proving a theorem?

The process of proving a theorem involves using logical reasoning and mathematical techniques to demonstrate that the statement is true. This often includes breaking down the statement into smaller, more manageable parts and using previously established facts or axioms to build a logical argument.

How do you know when a theorem has been proven?

A theorem is considered to be proven when a logical argument has been constructed that demonstrates that the statement is true using previously established facts or axioms. This proof should be clear, concise, and logically sound.

Can a theorem be proven wrong?

No, a theorem cannot be proven wrong. This is because the process of proving a theorem involves using logical reasoning and previously established facts or axioms, which are considered to be indisputable. However, a theorem can be disproven if a logical argument is constructed that shows it to be false.

Why is it important to prove theorems?

Proving theorems is important because it allows us to establish the truth of mathematical and scientific statements. This not only helps us understand and explain the world around us, but it also provides a foundation for further research and development in various fields.

Similar threads

Replies
16
Views
4K
Replies
18
Views
4K
Replies
4
Views
2K
Replies
5
Views
2K
Replies
1
Views
1K
Replies
3
Views
2K
Back
Top