Abstract:The semiclassical Schrödinger equation with multiscale and random potentials often appears when studying electron dynamics in heterogeneous quantum systems. As time evolves, the wavefunction develops high-frequency oscillations in both the physical space and the random space, which poses severe challenges for numerical methods. In this paper, we propose a multiscale reduced basis method, where we construct multiscale reduced basis functions using an optimization method and the proper orthogonal decomposition method in the physical space and employ the quasi-Monte Carlo method in the random space. Our method is verified to be efficient: the spatial gridsize is only proportional to the semiclassical parameter and the number of samples in the random space is inversely proportional to the same parameter. Several theoretical aspects of the proposed method, including how to determine the number of samples in the construction of multiscale reduced basis and convergence analysis, are studied with numerical justification. In addition, we investigate the Anderson localization phenomena for Schrödinger equation with correlated random potentials in both 1D and 2D.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the numerical simulation difficulties encountered when dealing with multi - scale and random potential fields in the semi - classical Schrödinger equation. Specifically, when electrons propagate in materials with heterogeneous quantum systems, they will experience the action of multi - scale potential fields. This causes high - frequency oscillations of the electron wave function in physical space and random space, thus posing a severe challenge to numerical methods. Traditional direct methods are very time - consuming, and although methods based on asymptotic analysis have some improvements, they still have difficulties in dealing with multi - scale and random potential fields.
To meet these challenges, this paper proposes a Multiscale Reduced Basis Method (MRBM). This method constructs multi - scale reduced basis functions in physical space through optimization methods and orthogonal decomposition methods, and uses the Quasi - Monte Carlo (qMC) method in random space to efficiently solve this problem. The paper verifies the effectiveness of this method, that is, the spatial grid size is only proportional to the semi - classical parameter, and the number of samples in the random space is inversely proportional to this parameter. In addition, the paper also studies how to determine the number of samples required to construct multi - scale reduced basis functions and convergence analysis, and explores the Anderson localization phenomenon under relevant random potential fields.
### Specific problems solved by the paper:
1. **Handling of high - frequency oscillations**: Under multi - scale and random potential fields, the electron wave function will generate high - frequency oscillations in both physical space and random space, which poses a challenge to numerical simulation. The multi - scale reduced basis method proposed in the paper can effectively handle these high - frequency oscillations and reduce the demand for computational resources.
2. **Improvement of computational efficiency**: Traditional direct methods are very time - consuming when dealing with multi - scale and random potential fields. By constructing multi - scale reduced basis functions and using the Quasi - Monte Carlo method, the paper significantly improves computational efficiency.
3. **Study of Anderson localization**: The paper also explores the existence of Anderson localization under relevant random potential fields, which is of great significance for understanding the transmission behavior of electrons in disordered media.
### Method overview:
- **Construction of multi - scale reduced basis functions**: In physical space, multi - scale reduced basis functions are constructed through optimization methods and orthogonal decomposition methods.
- **Application of the Quasi - Monte Carlo method**: In random space, the Quasi - Monte Carlo method is used to approximate the random part of the solution.
- **Determination of convergence and the number of samples**: The paper studies how to determine the number of samples required to construct multi - scale reduced basis functions and conducts convergence analysis.
### Mathematical model:
The semi - classical Schrödinger equation studied in the paper is as follows:
\[
\begin{cases}
i\epsilon \frac{\partial \psi^\epsilon}{\partial t}=-\frac{\epsilon^2}{2}\Delta \psi^\epsilon + v^\epsilon(x,\omega)\psi^\epsilon, & x\in D,\omega\in\Omega,t\in\mathbb{R}, \\
\psi^\epsilon\in H^1_P(D), & \omega\in\Omega,t\in\mathbb{R}, \\
\psi^\epsilon|_{t = 0}=\psi_{\text{in}}(x), & x\in D,
\end{cases}
\]
where \(0 < \epsilon\ll1\) is an effective Planck constant, which describes the ratio of micro - and macro - scales; \(D = [0,1]^d\) is the spatial domain; \(H^1_P(D)\) is the Sobolev space under periodic boundary conditions; \(v^\epsilon(x,\omega)\) is the given random potential field; \(\psi^\epsilon=\psi^\epsilon(t,x,\omega)\) is the electron wave function; \(\psi_{\text{in}}(x)\) is the initial data.
### Main contributions:
- Proposed an efficient multi - scale reduced basis method, which can effectively solve the semi - classical Schrödinger equation under multi - scale and random potential fields.
- Studied the Anderson localization phenomenon.