I would like to know a hint for this problem. Brezis describes it as it can be "solved explicitly by a very simple calculation". The problem is:

Given $f\in C^1([a,b])$, find a function $u$ satisfying:

\begin{equation}\tag{1} -u''+u=f, \text{ on }[a,b], \end{equation} and $$u(a)=u(b)=0.\tag{2}$$

Thanks Umberto P. I found the solution for (1) is:

$$u(t)=c_1(t)e^t+c_2(t)e^{-t}, \text{ where}$$ $$ c_1'(t)=-\frac{f(t)}{2e^{t}}, \text { and } $$ $$c_2'(t)=\frac{f(t)}{2e^{-t}} .$$

So functions $c_1$ and $c_2$ are determined modulo one constant. However, taking $a=0$ and $b=1$ and writing the boundary condition (2) results in a system that I can't solve.


This is a second order nonhomogeneous ODE with constant coefficients. You can find the method of solution in most calculus books. Look up e.g. "variation of parameters".

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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