[Splitting methods in Air-Pollution Problems]Project title: Splitting methods in air-pollution problems. Persons to contact: Per Grove Thomsen, building 305, room 117, phone +45 4525 3073, firstname.lastname@example.org
Background: In Air-Pollution problems the computational work is often preventively high when geographically detailed information is wanted. One way of improving the computational models is by using a splitting method for the time-stepping. Several computational schemes based upon splitting have been used and are documented in the litterature. A generalization of the splitting technique can be formulated mathematically whereby it is possible to rank different splitting methods with respect to computational efficiency and accuracy. In this project such a generalization is proposed and a family of methods are investigated for their properties to obtain basic information for this ranking.
Aim: Based upon the generalization of the splitting technique an air-pollution problem is studied and information about the properties.
Prerequisites: Knowledge of complex numbers, partial differential equations, nonlinear dynamics, as well as Fortran, C, or C++ for the modelling and Matlab for the visualization.
Consult DTU campus catalogue: IMM master in engineering courses.