I'm reading a paper where the following parabolic PDE is considered:

$u_t(x,t)=u_{xx}(x,t)+b(x)u_x(x,t)+\lambda(x)u(x,t)$, with boundary conditions $u_x(0,t)=qu(0,t) \text{ and } u(1,t)=\int_0^1 k(\xi)u(\xi,t)d\xi$ and initial condition $u(x,0)=u_0(x)$.

Here $q \in \mathbb{R}$, $b$ and $\lambda$ are $C^1$ functions and $k$ is infinitely differentiable.

The authors claim that any function $u$ that satisfies the integral equation $u=Fu$, is a solution to the PDE. Here

$Fu=\int_0^1 G(x,\xi,t,\tau)u_0(\xi)d\xi-\int_0^t\int_0^1 G_\xi(x,\xi,t,\tau)b(\xi)u(\xi,\tau)d\xi d\tau \\ + \int_0^t\int_0^1 G(x,\xi,t,\tau)\left[\lambda(\xi)-b_\xi(\xi)\right]u(\xi,\tau)d\xi d\tau \\ +\int_0^t G(x,1,t,\tau)(1+b(1))\int_0^1 k(\xi)u(\xi,t)d\xi d\tau-\int_0^t G(x,0,t,\tau)(b(0)+q)u(0,\tau)d\tau,$

where $G(x,\xi,t,\tau)=2\sum_{n=1}^\infty \cos(\lambda_n x)\cos(\lambda_n \xi) e^{-\lambda_n^2(t-\tau)}$ is the Green's function of the heat operator $\mathcal{L}u=u_t-u_{xx}$ with boundary conditions $u_x(0,t)=u(1,t)=0$ and $\lambda_n=(2n-1)\frac{\pi}{2}$.

I'm really having a hard time understanding that a function $u$ that satisfies $u=Fu$ is a solution to the PDE. For example, I can't even see how such a $u$ even satisfies the boundary conditions.


Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.