TY - JOUR
T1 - Studies of the relationship between petrography and grindability for Kentucky coals using artificial neural network
AU - Bagherieh, A. H.
AU - Hower, James C.
AU - Bagherieh, A. R.
AU - Jorjani, E.
PY - 2008/1/21
Y1 - 2008/1/21
N2 - Although there are several formulas available for predicting Hardgrove grindability of coal, most of them are linear and do not simultaneously take into consideration most of the relevant factors. The artificial neural network is an information processing tool that is capable of establishing an input-output relationship by extracting controlling features from a database presented to the network. In this paper, a neural network approach was proposed to deal with the grindability behavior of coal. 195 sets of experimental data were evaluated with artificial neural network to predict the HGI of Kentucky coals. Two different kinds of the trained artificial neural network were undertaken using the database created in this study. It is shown from the examples that the artificial neural network adequately recognized the characteristics of the coal experimental data sets, retaining a generality for further prediction. It is believed that an artificial neural network based prediction procedure shown in this paper can be further employed for Hardgrove grindability index prediction. The influence of liptinite, vitrinite, ash, and sulfur content on HGI was studied by a parametric study.
AB - Although there are several formulas available for predicting Hardgrove grindability of coal, most of them are linear and do not simultaneously take into consideration most of the relevant factors. The artificial neural network is an information processing tool that is capable of establishing an input-output relationship by extracting controlling features from a database presented to the network. In this paper, a neural network approach was proposed to deal with the grindability behavior of coal. 195 sets of experimental data were evaluated with artificial neural network to predict the HGI of Kentucky coals. Two different kinds of the trained artificial neural network were undertaken using the database created in this study. It is shown from the examples that the artificial neural network adequately recognized the characteristics of the coal experimental data sets, retaining a generality for further prediction. It is believed that an artificial neural network based prediction procedure shown in this paper can be further employed for Hardgrove grindability index prediction. The influence of liptinite, vitrinite, ash, and sulfur content on HGI was studied by a parametric study.
KW - Artificial neural network (ANN)
KW - Coal
KW - Coal petrography
KW - Hardgrove grindability index (HGI)
UR - http://www.scopus.com/inward/record.url?scp=36249003916&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=36249003916&partnerID=8YFLogxK
U2 - 10.1016/j.coal.2007.04.002
DO - 10.1016/j.coal.2007.04.002
M3 - Article
AN - SCOPUS:36249003916
SN - 0166-5162
VL - 73
SP - 130
EP - 138
JO - International Journal of Coal Geology
JF - International Journal of Coal Geology
IS - 2
ER -