This title appears in the Scientific Report :
2017
Please use the identifier:
http://hdl.handle.net/2128/13912 in citations.
Performance of optimization algorithms for deriving material data from bench scale tests
Performance of optimization algorithms for deriving material data from bench scale tests
In this work the performance of optimization algorithms forinferring material parameters for fire modeling from bench scale tests iscompared to each other.The well known Shuffled Complex Evolution algorithm (SCE) is compared to Artificial Bee Colony algorithm (ABC) and Fitness ScaledChaotic Artifici...
Saved in:
Personal Name(s): | Lauer, Patrick (Corresponding author) |
---|---|
Trettin, Corinna / Wittbecker, Friedrich-Wilhelm / Arnold, Lukas | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
2016
|
Physical Description: |
21 |
Conference: | Fire and Evacuation Modelling Technical Conference 2016, Meliá Costa del Sol (Spain), 2016-11-16 - 2016-11-18 |
Document Type: |
Contribution to a conference proceedings |
Research Program: |
Computational Science and Mathematical Methods |
Link: |
OpenAccess OpenAccess |
Publikationsportal JuSER |
In this work the performance of optimization algorithms forinferring material parameters for fire modeling from bench scale tests iscompared to each other.The well known Shuffled Complex Evolution algorithm (SCE) is compared to Artificial Bee Colony algorithm (ABC) and Fitness ScaledChaotic Artificial Bee Colony algorithm (FSCABC). First, these algorithms are tested with synthetic data, where all the properties are certainin advance. After that, the algorithms are tested with real data gainedfrom bench scale tests, namely thermogravimetric analysis (TGA) andmass loss calorimeter (MLC). Fire Dynamics Simulator (FDS) with itsimplemented pyrolysis model is used to carry out the simulations in anautomated optimization framework on a high performance computingcluster in parallel. The achieved results show which of the compared optimization strategies perform better than SCE related to efficiency and accuracy. |