Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

My question is about an article by J.L. Lions 1, where he introduced the "Hilbert Uniqueness Method" (HUM) for finding a boundary control function (dirichlet action) to bring the system to rest within a finite time T.

Regard the initial value problem to the wave equation,

$$\partial_{tt}u(x,t)-\Delta u(x,t)=0\hspace{.2in}\text{in } \Omega\times(0,T)$$ with initial values $$u(x,0)=u_0,~ \partial_t u(x,0)=u_1$$ and boundary term $$u=v\hspace{.2in}\text{on }\partial\Omega\times (0,T).$$ The task is to find a control function $v$ on (parts of) the boundary $\partial \Omega$ so that the system stops until $t=T$, i.e. $u(x,T)=\partial_t u(x,T)=0$.

The HUM starts with writing down the 2 wave systems

$$\phi''-\Delta\phi=0\hspace{.2in}\text{in } \Omega\times(0,T),$$ $$\phi(0)=\phi_0,~\phi'(0)=\phi_1\hspace{.2in}\text{in } \Omega,$$ $$\phi=0 \hspace{.2in}\text{on } \partial\Omega\times(0,T),$$ and $$\psi''-\Delta\psi=0\hspace{.2in}\text{in } \Omega\times(0,T),$$ $$\psi(T)=\psi'(T)=0\hspace{.2in}\text{in } \Omega,$$ $$\psi=\partial \phi/\partial \nu\hspace{.2in}\text{on } \partial\Omega\times(0,T).$$ Now, Lions considers the map $$\Lambda\{\phi_0,\phi_1\}=\{\phi'(0),-\phi(0)\}.$$

The following argumentation is that if $\Lambda$ is invertible, the whole problem is solveable by considering $\Lambda\{\phi_0,\phi_1\}=\{u_1,-u_0\}$, since both systems obtain a unique solution. You then just solve for $\phi_0,\phi_1$, solve the first system, calculate the normal derivative of $\phi$ at the boundary for all times $t\in(0,T)$, insert in the second system and obtain a solution for a system that comes to rest at time $T$, so that the evaluation of $\psi$ at the boundary delivers the desired control.

My question: Why the choice $\psi|_{\partial\Omega}=\partial \phi/\partial \nu$? I mean, through a rather technical proof, he was able to show that with that choice, $\Lambda$ is invertible for sufficently large $T$. But I don't get the idea, intuitively.

share|cite|improve this question

There's no intuition here! It is a classical Neumann Boundary value problem. Basically, this condition gives you information about the "angle" with which your solution hits the boundary. Recall that $\nu$ is the normal to the boundary itself.

The map $\Lambda$ is the Dirichlet to Neumann map.

Hope this help!

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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