Speaker
Description
As part of our research, we cooperated with GLASS SERVICE, a. s., with which we had an economic contract. The parameters of the forehearth were optimized for this company using multilayer artificial neural networks. We used the lib4neuro and neuron4dyn libraries developed by our team colleagues. In the future, it is planned to use these libraries also for many other purposes.
The main aim of this research was to analyze the possibilities of optimization of the forehearth by parameters adjusted at the inlet based on temperature measurement at the outlet, using artificial neural networks (ANN). First, was done a generalization of the existing linear regression model used by company GLASS SERVICE, a.s. by replacing empirical 1D transmission functions with artificial neural networks.
Our research has not been published in any magazine yet, but it is planned to be published soon. The project resulted in a contract successfully fulfilled.