Reduced order models based on pod method for Schrödinger equations

Abstract

Reduced-order models (ROM) are developed using the proper orthogo- nal decomposition (POD) for one dimensional linear and nonlinear Schrödinger equations. The main aim of this paper is to study the accuracy and robustness of the ROM approximations. The sensitivity of generated optimal basis functions on various pa- rameters of the algorithms is discussed. Errors between POD approximate solutions and exact problem solutions are calculated. Results of numerical experiments are presented.