Artificial Neural Network for Predicting Silicon Content in the Hot Metal Produced in a Blast Furnace Fueled by Metallurgical Coke

The main production route for cast iron and steel is through the blast furnace. silicon content in an important indicator of thermal condition a High contents indicate increase furnace's input and, some cases, may excess coke reactor. As costs predominate iron, tighter control therefore has economic advantages. objective this article was to design artificial neural network predict hot metal, varying number neurons hidden layer by 10, 20, 25, 30, 40, 50, 75, 100, 125 , 150, 170 200 neurons. In general, all networks showed excellent results, with 30 showing best results among 12 modeled networks. validation models confirmed using Mean Square Error (MSE) Pearson's correlation coefficient. cross-validation technique used re-evaluate performance short, can be practical operations due correlations between real values ​​and those calculated network.