Loading...
Projects / Programmes source: ARIS

Relating the Physical Properties of Polymeric Materials by Parallel Implementation of Soft Computing Methods

Research activity

Code Science Field Subfield
2.07.00  Engineering sciences and technologies  Computer science and informatics   

Code Science Field
P170  Natural sciences and mathematics  Computer science, numerical analysis, systems, control 
T390  Technological sciences  Polymer technology, biopolymers 
Keywords
physical properties of rubber compounds, soft computing, neural networks, GRID, parallel computers
Evaluation (rules)
source: COBISS
Researchers (9)
no. Code Name and surname Research area Role Period No. of publicationsNo. of publications
1.  13982  PhD Mojca Ciglarič  Computer science and informatics  Researcher  2004 - 2005  368 
2.  17136  Vito Čehovin    Technical associate  2004 - 2005  14 
3.  02137  PhD Ilija Dimitrievski  Materials science and technology  Researcher  2004 - 2007  83 
4.  02272  PhD Andrej Dobnikar  Computer science and informatics  Researcher  2004 - 2007  212 
5.  16109  PhD Uroš Lotrič  Computer science and informatics  Head  2004 - 2007  176 
6.  19365  PhD Matjaž Pančur  Computer science and informatics  Researcher  2004 - 2005  96 
7.  14300  PhD Branko Šter  Computer science and informatics  Researcher  2004 - 2007  151 
8.  02603  PhD Zoran Šušterič  Chemical engineering  Researcher  2004 - 2007  213 
9.  06795  PhD Mira Trebar  Computer science and informatics  Researcher  2004 - 2007  197 
Organisations (2)
no. Code Research organisation City Registration number No. of publicationsNo. of publications
1.  1539  University of Ljubljana, Faculty of Computer and Information Science  Ljubljana  1627023  16,242 
2.  1858  SAVATECH Industrial Rubber and Tyres  KRANJ  1661205  253 
Abstract
For elastomeric materials it stands that the low strain mechanical properties of rubber compounds are related to the elastic properties of their vulcanizates, owing to multitude of secondary intermolecular linkages. These relations, being obvious at low strains and constant temperature, are blurred due to nonlinear response of materials at high strains and at different temperatures. Revealing new relations between materials' properties at high strains and different temperatures would contribute to improvement of laboratory testing quality. The number of lengthy tests on vulcanizates would thus be lowered, and additional information on suitability of test methods would be gained by comparison of the results. Tofind complex relations between the data, techniques based on soft computing, primarily neural networks, evolutionary algorithms and fuzzy logic are used nowadays. Neural networks as general regression models are owing to their capability of generalization, adaptation and learning from examples very popular at solving the foregoing problems. Unfortunately, the learning procedures are time consuming which is particularly concerning when great quantity of data is dealt with. To gain additional processing power resulting in faster procedures of polymer properties relations' establishment, the common desktop computers available in the company (at least at night) are intended to be joined into parallel computer. Possibilities enabled by new technologies will be centered on, in particular joining the computers into clusters and similar architectures.
Views history
Favourite