2
$\begingroup$

My question is regarding the Kirchhoff plate pde which is the so-called biharmonic operator. Let $\Omega=[a,b]\times[a,b]\subset \mathbb{R}^2$, and $Z$ be the Hilbert space $Z=L^2[\Omega]$. The operator is given by the following:
\begin{equation} \mathcal{A}: D(A) \rightarrow Z \end{equation} \begin{equation} \mathcal{A}h= \frac{\partial^4}{\partial x^4}+2\frac{\partial^4}{\partial x^2 y^2}+\frac{\partial^4}{\partial y^4}, \end{equation} with domain \begin{equation} \begin{split} D(\mathcal{A})=\{h\in Z | h\in \mathcal{H}^4 \text{with} \\ M_{x}(0,.)={Q}_{x}(0,.)=0, M_{x}(1,.)={Q}_{x}(1,.)=0,\\ M_{y}(.,0)={Q}_{y}(.,0)=0, M_{y}(.,1)={Q}_{y}(.,1)=0 \}. \end{split} \end{equation} where \begin{equation} Q_{x}=-\frac{\partial M_{x}}{\partial x}-2\frac{\partial M_{xy}}{\partial y}, \end{equation} \begin{equation} Q_{y}=-\frac{\partial M_{y}}{\partial y}-2\frac{\partial M_{xy}}{\partial x}, \end{equation} \begin{equation} M_{x}=\frac{\partial^2 h }{\partial x^2}+\nu\frac{\partial^2 h}{\partial y^2}, \end{equation} \begin{equation} M_{y}=\nu\frac{\partial^2 h }{\partial x^2}+\frac{\partial^2 h}{\partial y^2}, \end{equation} Boundary components are the be nding moment $M_{x}$, $M_{y}$ and shear forces ${Q}_{x}$ and ${Q}_{y}$ and all equal to zero.
Note that $\mathcal{H}^4(\Omega)$ is the Sobolev space and consists of all functions in $L^2(\Omega)$ which are four times differentiable and whose derivatives up to the 4th order also lies in $L^2(\Omega)$.

In order to prove the self-adjointness of the operator, I have to show that it's symmetric and (surjective)onto. I've shown the symmetry by using inner product which is given by the following; \begin{equation} \langle \mathcal{A}x,y \rangle= \langle x, \mathcal{A}y \rangle \end{equation} My problem is about the surjectivity. How can I show that this operator is surjective (in other words the inverse exists)?

Thanks in advance Fatemeh

$\endgroup$
3
  • 2
    $\begingroup$ Welcome to MathOverflow! First a remark concerning the symmetry: The domain of your operator is too large, i.e., it does not encode any boundary conditions; this implies that the operator is actually not symmetric (in order to prove symmetry, you need to integrate by parts, and there you need certain $0$-boundary conditions which ensure that the occurring boundary terms vanish). $\endgroup$ Commented Dec 15, 2020 at 9:57
  • 4
    $\begingroup$ Crossposted from Mathematics Stack Exchange. Please explicitly mention and link the crosspost within your question in such a case. $\endgroup$ Commented Dec 15, 2020 at 10:12
  • $\begingroup$ Dear Jochen, thanks for your comment. I've added the boundary conditions to the question's post. Since they are all equal to zero, after integration by parts they disappeared. Please note that I used the notation 'bending moment' $M_{x}, M_{y}$ and shear force $Q_{x},Q_{y}$ instead of 'displacement field' $h$, which have second and third order derivative of the displacement field. For more details about the notation, please see the paper with DOI:10.1016/j.apm.2019.04.036. I used integration by parts and showed the symmetry. But regarding the surjectivity I don't have any clue in my mind! $\endgroup$ Commented Dec 15, 2020 at 15:36

0

You must log in to answer this question.

Browse other questions tagged .