Modeling and Optimization of Galena Dissolution in Hydrochloric Acid: Comparison of Central Composite Design and Artificial Neural Network

Abstract

Response surface methodology (RSM) and Artificial neural network (ANN) were used for the simulation and optimization of galena dissolution in hydrochloric acid. The galena ore was characterized for structure elucidation using FTIR, SEM and X-ray diffraction spectroscopic techniques and the results indicate that the galena ore exists mainly as lead sulphide (PbS). A feed-forward neural network model with Leverberg-Marquardt back propagating training algorithm was used to predict the response (lead yield). The leaching temperature, acid concentration, solid/liquid ratio, stirring rate and leaching time were defined as input variables, while the percentage yield of lead was labelled as output variable. The multilayer perceptron with architecture of 5-9-1 provided the best performance. All the process variables were found to have significant impact on the response with p-values of <0.0001. The performance of the RSM and ANN model showed adequate prediction of the response, with AAD of 0.750% and 0.295%, and R2 of 0.991 and 1.00, respectively. A non-dominated optimal response of 85.25% yield of lead at 343.96 K leaching temperature, 3.11 M hydrochloric acid concentration, 0.021 g/ml solid/liquid ratio, 362.27 rpm stirring speed and 87.37 min leaching time was established as a viable route for reduced material and operating cost using RSM.