The Use of Neural Network BP Algorithm in Magnesium Smelting Process Parameter Optimization

Because artificial neural networks discard the traditional modeling methods, it can extract domain knowledge from a large number of discrete experimental data via study and training, and express these knowledge as network connection weights, so as to establish the corresponding relation model. In this paper, based on neural network BP algorithm, we built a relation model that shows how various process parameters affect the magnesium output rate in Pidgeon magnesium reduction process. This laid a foundation for process parameters optimization.