EGSIM
BMBF-Project EGSIM "Efficient groundwater simulation for the production of energy resources"
Project data:
Duration: 01.10.2010 – 31.03.2013
Coordinator TUD: Prof. Dr.-Ing. habil. Peter-Wolfgang Gräber
Partners :
Ingenieurbüro für Grundwasser und Boden Leipzig (project
supervisor)
DUALIS GmbH IT Solution, Dresden
TU Dresden, IAA
Funding: BMBF
The aim of the project is the development of software to
simulate flow and reactive transport processes in the soil and
the groundwater (porous media). Simulation of such processes is
a basis for the planning of environmentally sound and
energy-efficient lignite mining, with lignite being one of the
most important energy sources in Germany. The planning engineer
shall be enabled to develop and monitor sustainable water
management projects by using an environment based on GIS
software. Exemplary applications could be in the field of
groundwater management in open-cast mining, or in renaturation
of post-mining landscapes. By using optimized simulations
significant amounts of water resources can be protected (f.e.
environmental protection to sustain natural wetlands), and the
amount groundwater pumping can be lowered, thus saving energy.
Until now planning and simulation were carried out using
individual calculations and static graphic analysis, sometimes
requiring days or weeks to simulate reactive transport
processes. The parameter spaces resulting from these
simulations can be reduced by modern methods of sensitivity
analysis. Especially in water management projects often several
goals need to be achieved at the same time which can only be
solved using intelligent multicriterial optimization
algorithms. Based on existing source codes a scalable
simulation software will be developed, using modern software
algorithms like efficient parallel equation solvers and
database structures capable of handling large numerical
datasets, and by utilizing massive parallel computer hardware.
The work schedule comprises seven tasks:
- direct software coupling and management of four individual
components
- data and project management with database structures
- acceleration of simulations by using HPC
- integration of conceptional object-oriented modelling
elements
- data input and output by software coupling to GIS
- development of a sensitivity module for calibration
- development of a multicriterial optimizer for parameter
identification