Prediction for the Sulfur Content in Pig Iron of Blast Furnace by Combining Artificial Neural Network with Genetic Algorithm

Abstract

Sulphur content in pig iron is one of the most important indices to represent the quality of liquid iron in blast furnace. In order to timely control sulfur content, a mathematical model is developed to predict the sulfur content in pig iron. Compared with the conventional artificial neural network model, a new method is developed to integrate artificial neural network with genetic algorithm. The genetic algorithm with the binary coded chromosome is used to optimize initial neural network weights. In this project, the first aim is to train the model by integrating artificial neural network with genetic algorithm according to the past processing variables. The second is to predict the next sulfur content based on the training result in the first step and according to the current processing variables. Compared with only using artificial neural network model, this developed method can improve the predicted accuracy. From practical applications, it can be found the model is exact and reliable to predict the sulfur content in the pig iron of blast furnace.