Title :
Predictive model of Mn-Si Alloy Smelting Energy Consumption based on Double Wavelet Neural Network
Author :
Yang, Hong-Tao ; Li, Xiu-Lan ; Zhang, Niao-na
Author_Institution :
Inst. of Electr. & Electron. Eng., Changchun Univ. of Technol., Changchun, China
Abstract :
Unit electricity consumption is important indicator on production of Mn-Si alloy. There exists a serious nonlinear relationship among unit electricity consumption and the ferromanganese´s grade and furnace average power and the amount of ferrosilicon powder and the amount of coke and daily average output etc. The predictive model of Mn-Si Alloy Smelting Energy Consumption based on Double Wavelet Neural Network was put forward, and the research of verifying the model was made by comparing the predictive value with the practical data of a Ferroalloy Company. The results show that the mean absolute relative forecasting error of unit electricity consumption was 0.9%, while the mean absolute relative forecasting errors of regression wavelet neural network and time-delay wavelet neural network were 2.1% and 1.3% respectively. It was proved that the double wavelet neural network model had preferable forecasting accuracy.
Keywords :
blast furnaces; energy consumption; manganese alloys; neural nets; power engineering computing; production engineering computing; regression analysis; silicon alloys; smelting; wavelet transforms; MnSi; double wavelet neural network; ferroalloy company; ferrosilicon powder; furnace average power; predictive model; regression wavelet neural network; relative forecasting error; smelting energy consumption; time delay wavelet neural network; unit electricity consumption; Accuracy; Double Wavelet Neural Network; Mn-Si Alloy Smelting; Predictive Model;
Conference_Titel :
Computer, Mechatronics, Control and Electronic Engineering (CMCE), 2010 International Conference on
Conference_Location :
Changchun
Print_ISBN :
978-1-4244-7957-3
DOI :
10.1109/CMCE.2010.5610323