DocumentCode :
653041
Title :
Iron and steel process energy consumption prediction model based on selective ensemble
Author :
Na Wei ; Li Li ; Jun Zhu ; Na Li
Author_Institution :
Coll. of Electron. & Inf. Eng., Tongji Univ., Shanghai, China
fYear :
2013
fDate :
25-27 Sept. 2013
Firstpage :
203
Lastpage :
207
Abstract :
Aiming at solving the energy consumption and pollution problem of China´s steel industry, this paper proposed an energy consumption prediction model by using intelligence algorithms based on selective ensemble theory. The main work of this paper is realizing the prediction model through BP artificial neural network and genetic algorithm, moreover, combining the 2 methods mentioned above to improve accuracy. Based on the same dataset, this paper compared the experimental result with single intelligence algorithm, and it indicates that selective ensemble could enhance the weak learning algorithm to a strong one, so that to improve the prediction accuracy. This method has achieved a relatively good effect in energy prediction by adjusting coking charge ratio, meanwhile provided guidance for iron enterprise in energy saving and emission reduction.
Keywords :
backpropagation; coke; energy consumption; genetic algorithms; iron; neural nets; production engineering computing; steel; steel industry; BP artificial neural network; coking charge ratio; emission reduction; genetic algorithm; intelligence algorithm; iron enterprise; iron-steel process energy consumption prediction model; pollution problem; prediction accuracy; selective ensemble theory; steel industry; Artificial neural networks; Genetic algorithms; Steel; Training; coking process; energy-consumption prediction; intellectual algorithm; selective ensemble;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Advanced Mechatronic Systems (ICAMechS), 2013 International Conference on
Conference_Location :
Luoyang
Print_ISBN :
978-1-4799-2518-6
Type :
conf
DOI :
10.1109/ICAMechS.2013.6681778
Filename :
6681778
Link To Document :
بازگشت