| Popis: |
In some cases, the mathematical expressions describing chemical problems are actually very difficult to handle. We are studying one of these systems, the non isothermal combustion of hydrogen-oxygen-nitrogen mixtures over Pd catalysts. Data obtained during stationary and dynamic experiments are analyzed by using neural networks, and the results are compared to results obtained on the same basis with orthogonal collocation. Neural networks analyze experimental data, in order to determine the behaviour of the system. Thanks to this tool, we are able to model the behaviour of such a difficult reactional system in a very good way. New improvements in the use of neural networks for the modelling of chemical phenomena are also showed. |
