DBPapers
DOI: 10.5593/sgem2012/s03.v1043

POWDER FACTOR PREDICTION IN URMIA CEMENT MINE UTILISING NEURAL NETWORK

DA. K. AHANGARAN, M. NIKZAD, A. ZOMORODIAN, A. WETHERELT, P. J FOSTER, A. BIJAN YASREBI, P. AFZAL
Wednesday 1 August 2012 by Libadmin2012

References: 12th International Multidisciplinary Scientific GeoConference, www.sgem.org, SGEM2012 Conference Proceedings/ ISSN 1314-2704, June 17-23, 2012, Vol. 1, 729 - 736 pp

ABSTRACT

Powder factor is one of the fundamentally crucial parameters in a blasting design. It is
believed that any inaccurate blasting pattern has the consequences of back-break,
fragmentation, fly rock and excess vibration which all result in negative impacts. The
absolute value of powder factor has to be determined with respect to a specific rock type
and rock mass which are generally considered the most important elements among other
technical issues in a blast design. To date, extensive research and associated
examination of this problematic area has focused on achieving a technological solution
and methodology to enable correct mine blasting, hence, to address this many
techniques have come forward and are now widely utilised such as Neural Network,
Fuzzy Logic and Expert Systems. However, within those techniques mentioned above,
Artificial Neural Network is deemed an excellent approach.
The aim of this study is to determine an optimal powder factor for future blasts at the
Urmia Cement Mine located in the north west of Iran which is a major cement producer
in the western part of the country. In order to do this, Artificial Neural Network and
MATLAB software were employed in terms of 120 blasts being carefully analysed at
this operation. The Results of this research indicate that the developed Neural Network
model can be applied to generate future designs and plans considering blasting
operations in Urmia Cement Mine.

Keywords: Powder factor; blasting; Joint; Artificial Neural Network (ANN); Iran