The application of improved back propagation neural network model
Fang Li1, Changze Wu2
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12C) 34-39
1Chongqing City Management College, Chongqing 400031, China
2College of Computer Science, Chongqing University, Chongqing, China
During the granulation process of Iron ore sinter mixture, there are many factors affecting the granulation result such as chemical composition, size distribution, surface feature of particle, and so on. Some researchers use traditional fitting calculation methods like least square method and regression analysis method to predict granulation result, where exists big error. In order to provide better performance in prediction, we use improved BP (Back propagation) neural network model to do data analysis and processing. Granulating effect neural network model with a shooting rate of 92%, has a good prediction accuracy, robust, and the high ability of recognition to new sample, which can give a good guidance to granulation process. It obtains better effect than traditional fitting calculation methods.