Title :
The modeling for pellets induration process based on Bagging method
Author :
Xiaoke Fang ; Jun Peng ; Jianhui Wang ; Xiao Wang
Author_Institution :
Coll. of Inf. Sci. & Eng., Northeastern Univ., Shenyang, China
fDate :
May 31 2014-June 2 2014
Abstract :
The grate-kiln system for iron ore pellet induration is a nonlinear, high coupling and large delay process. Considering the deviation between assumption and actual results, it is hard to build accurate kinetic models. Besides, the pure kinetic model also has the limitations to describe the induration process. In this paper, based on kinetic modeling, a hybrid model is built via neural network ensemble. The Bagging method is applied for training sample set with BP network as its network. The result shows that the hybrid model is more accurate and better than the kinetic model.
Keywords :
backpropagation; kilns; mineral processing; minerals; neural nets; production engineering computing; BP network; bagging method; grate-kiln system; high coupling process; hybrid model; iron ore pellet induration; kinetic models; large delay process; neural network ensemble; nonlinear process; pellet induration process modeling; sample set training; Bagging; Iron; Kinetic theory; Mathematical model; Neural networks; Predictive models; Training; Bagging; Ensemble Learning; Hybrid Modeling; Iron Ore Pellet;
Conference_Titel :
Control and Decision Conference (2014 CCDC), The 26th Chinese
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-3707-3
DOI :
10.1109/CCDC.2014.6852270
