1
$\begingroup$

I work on a bounded domain in $\mathbb{R}^n$. Let $u \in H^1(0,T;H^{-1})\cap L^2(0,T;H^1)$ be a solution of the heat equation: $$\langle u', v \rangle + \int \nabla u \nabla v = 0$$ for each test function $v$. The solution has the property $$\int_\Omega u(t) = 0$$ for each $t$.

Is it possible to conclude from this information that $u \equiv 0$?

$\endgroup$
2
  • $\begingroup$ This might be more appropriate at math.stackexchange. For an answer, what about unit ball as $\Omega$ and $u(x,t)=x_1$? This solves the heat equation and averages to zero. You specified no boundary conditions. $\endgroup$ Aug 15, 2014 at 11:37
  • $\begingroup$ @JoonasIlmavirta I think use of $H^{-1}$ as a function space in the formulation of the question means that the boundary condition, which isn't written is Dirichlet $0$ (but it doesn't change the answer to the question asked). $\endgroup$
    – username
    Feb 28, 2018 at 9:13

1 Answer 1

4
$\begingroup$

I think the answer to your question is 'no' since the function $e^{-t}\sin (x)$ solves the heat equation on $\Omega=(-1,1)\times \mathbb R_+$ but the integral is zero by symmetry. If you want an example where $n>1$, just add a dummy variable.

$\endgroup$

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.