Generalized Multiscale Finite Element Methods for Transport Problems with Heterogeneous Media
MetadataShow full item record
Many practical problems are modeled by partial differential equations with highly heterogeneous coefficients. Classical numerical methods for solving these problems typically require very fine computational meshes, and are therefore very expensive to use. In order to solve these problems efficiently, one needs some types of model reduction, which is typically based on upscaling techniques or multiscale methods. In upscaling methods, the heterogeneous coefficient is carefully replaced by an effective medium, so that the system can be solved on a much coarser grid. In multiscale methods, one attempts to represent the solution by some multiscale basis functions. These basis functions are constructed carefully and are usually based on some local cell problems. The purpose is to capture the fine scale properties of the true solution by using a few multiscale basis functions, with the aim of reducing computational costs. There are many existing approaches for multiscale methods, but few works are done about transport problem. And we know there are many applications of transport equation in real life. As such, some efficient model reduction methods are required to handle transport problems. In the dissertation, we consider H(curl)-elliptic problems, transport equations, and Boltzmann equations. We will consider a multiscale method called generalized multiscale finite element method (GMsFEM). We first construct an adaptive multiscale method for solving H(curl)-elliptic problems in highly heterogeneous media. This problem is motivated by electromagnetic wave propagation. And there are few existing works applying upscaling techniques on curl operator. In our method, we will first construct a suitable snapshot space, and a dimensional reduction procedure to identify important modes of the solution. We next develop and analyze an a posteriori error indicator, and the corresponding adaptive algorithm. In addition, we will construct a coupled offline-online adaptive algorithm, which provides an adaptive strategy to the selection of offline and online basis functions. Our theory shows that the convergence is robust with respect to the heterogeneities and contrast of the media. We present several numerical results to illustrate the performance of our method. We next consider solving transport equations. Most of existing multiscale approaches use spatial multiscale basis functions or upscaling, and there are very few works that design spacetime multiscale functions to solve the transport equation on a coarse grid. For the time dependent problems, the use of space-time multiscale basis functions offers several advantages as the spatial and temporal scales are intrinsically coupled. By using the GMsFEM idea with a space-time framework, one obtains a better dimensional reduction taking into account features of the solutions in both space and time. In addition, the time-stepping can be performed using much coarser time step sizes compared to the case when spatial multiscale basis are used. Our scheme is based on space-time snapshot spaces and model reduction using space-time spectral problems derived from the analysis. We give the analysis for the well-posedness and the spectral convergence of our method. We also present some numerical examples to demonstrate the performance of the method. In all examples, we observe a good accuracy with a few basis functions. We finally solve Boltzmann equations, which are used to describe the statistical behavior of a large number of particles driven by the same physics laws. Depending on the media and the particles to be modeled, the equation has slightly different forms. In this dissertation, we investigate a model Boltzmann equation with highly oscillatory media in the small Knudsen number regime, and study the numerical behavior of GMsFEM in the fluid regime when high oscillation in the media presents. The method is divided into the offline and online steps. In the offline step, basis functions are prepared from a snapshot space via a well-designed generalized eigenvalue problem (GEP), and these basis functions are then utilized to patch up for a solution through DG formulation in the online step to incorporate specific boundary and source information. We prove the wellposedness of the method on the Boltzmann equation, and show that the GEP formulation provides a set of optimal basis functions that achieve spectral convergence. Such convergence is independent of the oscillation in the media, or the smallness of the Knudsen number, making it one of the few methods that simultaneously achieve numerical homogenization and asymptotic preserving properties across all scales of oscillations and the Knudsen number.
Li, Yanbo (2019). Generalized Multiscale Finite Element Methods for Transport Problems with Heterogeneous Media. Doctoral dissertation, Texas A&M University. Available electronically from