Özet:

In this study, the approach of the artificial nerve network (YSA) was applied to the elimination of S. cerevisiae and Remazol Yellow (RR) in the cutting system. In this scope, the system input data for the artificial nerve network is determined as the pH, the initial paint concentration and the initial biosorbent concentration, while the output layer is estimated as the percentage of the paint extraction value.  The artificial nerve network (YSA) training is done using the Levenberg-Marquardt advanced nutrition algorithm and the experimental data is divided into 60% training, 20% validation and 20% test. The maximum turn value (epoch) is 12000 iterations.  The system’s R2 values are 98% for training, 96% for validation and 98% for the entire biosorption system. In the continuation of the study, separate modeling studies for the pH, initial paint and biosorbent concentrations, which are the system variables within the framework of the biosorbent system modeling, have been carried out. The study resulted in comparing the experimental and model estimate percentage deduction values, the YSA and the system were well modeled and the model had a good predictive ability.

Anahtar Kelimeler: