Title: Modelling and simulation of the cyanidation process of Aghdareh gold ore using artificial neural network and multiple linear regression

Authors: Asghar Azizi; Reza Ghaedrahmati; Nader Ghahramani; Reza Rooki

Addresses: Department of Mining, Petroleum and Geophysics, Shahrood University of Technology, Shahrood, 3619995161, Iran ' Mining Faculty, Engineering Department, Lorestan University, Lorestan, 681514416, Iran ' Mining Faculty, Engineering Department, Lorestan University, Lorestan, 681514416, Iran ' Mining Faculty, Engineering Department, Birjand University of Technology, Birjand, 9719866981, Iran

Abstract: This paper describes a simple and more reliable artificial neural network (ANN) method and multiple linear regression (MLR) to predict the gold recovery and to simulate the effects of pH, solid percentage, NaCN concentration, particle size and leaching time during cyanidation process. Feed-forward ANN with back-propagation learning algorithm with 5-9-1 arrangement was found capable to predict the recovery of Au from cyanide leaching solution. The results showed that simulated values obtained by the network were very close to the experimental results. The coefficient of determination value (R2) was 0.9803 for training set, and in testing stage the R2 value was 0.8213. On the contrary, the correlation coefficients were low for the results predicted by MLR method. R2 values obtained 0.5561 and 0.6705 for training and testing data, respectively. The results obtained from this paper can be considered as an easy and cost-effective method to simulate the cyanide leaching process of gold.

Keywords: cyanide leaching; gold recovery; ANNs; artificial neural networks; MLR; multiple linear regression; modelling; simulation; cyanidation; gold ore; pH; solid percentage; NaCN concentration; sodium cyanide; particle size; leaching time.

DOI: 10.1504/IJMME.2016.076497

International Journal of Mining and Mineral Engineering, 2016 Vol.7 No.2, pp.139 - 154

Received: 26 Oct 2015
Accepted: 12 Jan 2016

Published online: 10 May 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article