Investigation of the dehydration of ulexite ore with different parameters and modeling with artificial neural network (ANN) method

Study experiments were conducted considering the temperature, time, and sample weight parameters in order to model dehydration by applying processes ulexite ores. Data obtained from of ore compared with TG analyzes. It was observed as result heat treatment that fastest water removal provided temperature range 150-250 °C it very low 400-750 °C. In design ANN method, 4 different models proposed same network structure determined. The performance assessed means error measurements i.e. absolute (AE), relative (ARE) coefficient determination (R2). mean value R2 99%. found independent variables explained dependent variable efficiently successful. shown new can be created using genetic algorithms or hybrid methods future studies requiring fewer following process present study.