3
$\begingroup$

Let $p(x)$ be a positive measurable function on $(-1,1)$. Consider the Prandtl equation $$ u(x)-\frac{p(x)}\pi \int_{-1}^1 \frac{u'(t)}{t-x}dt=p(x)h_0(x),\quad u(1)=u(-1)=0.\quad\quad(\star) $$ What is state of the art of the existsence and regularity theory for $(\star)$, what is a standard reference? Recall that $(\star)$ may be rewritten as $u+p(x)\sqrt{-d^2/dx^2}u=p(x)h_0(x)$, where $-d^2/dx^2$ is Dirichlet Laplacian on $[-1,1]$, that relates this to the regularity theory for fractional Laplacians.

$\endgroup$
1

1 Answer 1

1
$\begingroup$

With $q=1/p$, let me write your equation as $$ \vert D\vert u+ q u= h_0, \quad u(\pm 1)=0. $$ Multiplying the equation by $u$, we get $$\Vert{u}\Vert_{H^{1/2}_0}^2\le \Vert{u}\Vert_{H^{1/2}_0}^2+\underbrace{\langle qu, u\rangle_{L^2}}_{\ge 0}=\langle h_0, u\rangle_{L^2}\le \Vert{u}\Vert_{H^{1/2}_0}\Vert{h_0}\Vert_{H^{-1/2}}, $$ entailing $ \Vert{u}\Vert_{H^{1/2}_0}\le \Vert{h_0}\Vert_{H^{-1/2}}. $ Assuming $h_0\in H^{-1/2}$ will give $u$ in $H^{1/2}_0$ (completion of smooth compactly functions supported in $(-1,1)$ for the $H^{1/2}$ norm; here $H^{-1/2}$ stands for the dual space of $H^{1/2}_0$). With no more information on the regularity of $1/p$, it seems questionable to improve the regularity $H^{1/2}$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.