Application of Taguchi method and artificial neural network model for the prediction of reductive leaching of cobalt(III) from oxidised low-grade ores

Abstract

The leaching process of cobalt using a wide range of experimental variables is described. The treated cobalt samples were from the Kalumbwe Mine in the south of the Democratic Republic of Congo. In this study, a predictive model of cobalt recovery using both the Taguchi statistical method and an artificial neural network (ANN) algorithm was proposed. The Taguchi method utilising a L25 (55) orthogonal array and an ANN multi-layer, feed-forward, back-propagation learning algorithm were adopted to optimise the process parameters (acid concentration, leaching time, temperature, percentage solid, and sodium metabisulfite concentration) responsible for the high recovery of cobalt by reducing sulfuric acid leaching. The ANN was built with a neuron in the output layer corresponding to the cobalt leaching recovery, 10 hidden layers, and 5 input variables. The validation of the ANN model was performed with the results of the Taguchi method. The optimised trained neural network depicts the testing data and validation data with R2 equal to 1 and 0.5676, respectively.

Recovery of Cu, Co, and Fe from Pyrite Cinder Based on Mineral Phase Reconstruction

“…In addition, in ref , Co(II) has four peaks attributed to Co 2p 3/2 , the binding energy at 2p 3/2 of high-valence cobalt (e.g., Co 3 O 4 ) was less than 780 eV, and there was also a fifth peak, which proved that Co in the samples was bivalent cobalt, consistent with the results of the cobalt chemical phase analysis. Usually, the high-valence cobalt needs to be reduced to divalent cobalt; then, it can be easily extracted by sulfuric acid solution. − The reduction roasting proved to be effective for the reduction of cobalt, and it transformed the high-valence phase of cobalt into the low-valence phase, thereby increasing the leaching rate of cobalt. Nevertheless, Co(II) still exists in the leaching residue, indicating that cobalt has not been fully extracted, which was because some cobalt may be wrapped by hematite.…”

Bayesian statistics study of a sustainable dissolution of cobalt‐bearing minerals from Cu‐Co ores

“…These include purely mathematical approaches such as the response surface methodology, [ 17 ] statistical approaches such as that of Taguchi, [ 9 ] principal component analysis [ 4 ] or artificial intelligence approaches such as artificial neural networks. [ 10,18 ] In the field of mineral processing, many mathematical or statistical models and algorithms have been used to predict and/or simulate the dissolution of minerals containing cobalt, however to date, no study has been devoted to probabilistic modelling, particularly with the Bayesian approach. This approach has evolved a lot over time under different names: belief networks, probabilistic‐oriented graphical models, probabilistic networks.…”