Multiple Neural Networks Based on Factor Analysis and its Application in Soft Sensor

Considering the fact that a large number of factors influence the industry process and their influence varies, a multiple neural networks (FA-MNN) model based on factor analysis is proposed. The structure of FA-MNN is introduced and its learning algorithm is improved. The FA-MNN model is applied in the process of high pressure digestion of alumina. Simulation results show that the FA-MNN model can sense the ratio of soda to aluminate and leaching rate online effectively.