An Exponentially Convergent Nonpolynomial Finite Element Method for Time-Harmonic Scattering from Polygons

Abstract

Abstract. In recent years nonpolynomial finite element methods have received increasing attention for the efficient solution of wave problems. As with their close cousin the method of particular solutions, high efficiency comes from using solutions to the Helmholtz equation as basis functions. We present and analyze such a method for the scattering of two-dimensional scalar waves from a polygonal domain that achieves exponential convergence purely by increasing the number of basis functions in each element. Key ingredients are the use of basis functions that capture the singularities at corners, and representing the scattered field towards infinity by a combination of fundamental solutions. The solution is obtained by minimizing a least-squares functional, which we discretize in such a way that a matrix least-squares problem is obtained. We give computable exponential bounds on the rate of convergence of the least-squares functional that are in very good agreement with the observed numerical convergence. Challenging numerical examples, including a nonconvex polygon with several corner singularities, and a cavity domain, are solved to around 10 digits of accuracy with a few seconds of CPU time. The examples are implemented concisely with MPSpack, a MATLAB toolbox for wave computations with nonpolynomial basis functions, developed by the authors. A code example is included.

Learning dominant wave directions for plane wave methods for high-frequency Helmholtz equations

“…where N (x) is the number of fronts/rays passing through x, and the phases φ n and amplitudes A n are single valued functions satisfying the eikonal/transport equations (8), each defined in a suitable domain with suitable boundary conditions [9]. Based on the above geometric optics ansatz, one can derive a local plane wave approximation at any point where φ n and A n are smooth with variations on a O(1) scale.…”

“…2 Adaptive methods, on the other hand, aim to leverage à priori knowledge of the solution of the Helmholtz equation, such as its known oscillatory behavior. In practice, adaptive methods have mostly focused on adaptivity to the medium, such as polynomial Galerkin methods with hp refinement [3,70,73,96,107,111], specially optimized finite differences [23,45,92,93,102] and finite elements [4,99], enriched finite elements [30][31][32][33], plane wave methods [5,21,42,43,46,69,74], generalized plane wave methods [54,55], locally corrected finite elements [17,38,82], and discretizations with specially chosen basis functions [7,8,76], among many others. They have been especially successful on reducing the pollution effect by accurately capturing the dispersion relation.…”

A new integral representation for quasi-periodic scattering problems in two dimensions

“…MPSpack is an object-oriented MATLAB toolbox for solving 2D Helmholtz PDE problems recently developed by Betcke and one of the authors [6]. The methods of…”

Error estimates for the ultra weak variational formulation in linear elasticity

“…In addition to the UWVF, DGM and PUFEM, other non-polynomial methods that can be applied to linear elasticity problems include, for example, the least-squares method (LSM) [27] and the non-polynomial finite element method [2]. To date, the LSM and non-polynomial finite element method have been used in 2D acoustic problems.…”