Abstract:We consider a non-overlapping domain decomposition method for optimal control problems of the tracking type governed by time-fractional diffusion equations in one space dimension, where the fractional time derivative is considered in the Caputo sense. We concentrate on a transmission problem defined on two adjacent intervals, where at the interface we introduce an iterative non-overlapping domain decomposition in the spirit of P.L. Lions for the corresponding first-order optimality system, such that the optimality system corresponding to the optimal control problem on the entire domain is iteratively decomposed into two systems on the respective sub-domains; this approach can be framed as first optimize, then decompose. We show that the iteration involving the states and adjoint states converges in the appropriate spaces. Moreover, we show that the decomposed systems on the sub-domain can in turn be interpreted as optimality systems of so-called virtual control problems on the sub-domains. Using this property, we are able to solve the original optimal control problem by an iterative solution of optimal control problems on the sub-domains. This approach can be framed as first decompose, then optimize. We provide a mathematical analysis of the problems as well as a numerical finite difference discretization using the L1-method with respect to the Caputo derivative, along with two examples in order to verify the method.
What problem does this paper attempt to address?
### Problems the paper attempts to solve
This paper aims to solve the non - overlapping domain decomposition method in one - dimensional optimal control problems (OCPs) controlled by time - fractional diffusion equations (TFDEs). Specifically, the paper focuses on the transmission problems defined on two adjacent intervals, where an iterative non - overlapping domain decomposition method is introduced at the interval interface to handle the corresponding optimality system.
### Main research contents
1. **Problem background**:
- Time - fractional diffusion equations are widely discussed in many applications, such as anomalous diffusion, viscoelastic fluid models with weakly singular kernels, the flow of nanofluids in vertical channels, sub - diffusion models in neuronal dendrites, etc.
- Fractional optimal control problems (FOCPs) arise naturally in the applications of time - fractional diffusion equations.
2. **Research methods**:
- Introduce the non - overlapping domain decomposition method to decompose the optimal control problem on the whole domain into two systems on sub - domains.
- Prove that the iterations of the state and the adjoint state converge in the appropriate spaces.
- Interpret the decomposed sub - domain systems as the optimality systems of virtual control problems.
- Propose two methods: first optimize, then decompose and first decompose, then optimize.
3. **Mathematical analysis**:
- Conduct a mathematical analysis of the problem, including existence and uniqueness results.
- Use the L1 method to perform numerical finite - difference discretization of the Caputo derivative and verify the effectiveness of the method through two examples.
### Specific problem description
Consider the following optimal control problem:
\[
\text{minimize } J(y, u)=\frac{\kappa}{2} \sum_{i = 1}^{2} \int_{0}^{T} \int_{0}^{\ell_{i}}|y_{i}(x, t)-z_{d}^{i}(x, t)|^{2} \, dx \, dt+\frac{\nu}{2} \sum_{i = 1}^{2} \int_{0}^{T} \int_{0}^{\ell_{i}}|u_{d}^{i}(x, t)|^{2} \, dx \, dt
\]
Subject to the following system constraints:
\[
\begin{cases}
{}^{C}D_{t}^{\alpha}y_{i}(x, t)=\frac{\partial^{2}y_{i}(x, t)}{\partial x^{2}}+f_{i}(x, t)+u_{d}^{i}(x, t), & (x, t)\in(0, \ell_{i})\times(0, T), \, 0 < \alpha < 1,\\
y_{i}(x, 0)=y_{0}^{i}(x), & x\in(0, \ell_{i}), \, i = 1, 2,\\
y_{i}(0, t)=y_{j}(0, t), & i\neq j, \, i, j = 1, 2, \, t\in(0, T),\\
\frac{\partial y_{1}(0, t)}{\partial x}+\frac{\partial y_{2}(0, t)}{\partial x}=0, & t\in(0, T),\\
y_{i}(\ell_{i}, t)=0, & t\in(0, T), \, i = 1, 2.
\end{cases}
\]
### Solutions
- **Non - overlapping