Perdition of Semi-autogenous mill Power Using Radial Artificial Neural Network Based on Principal Component

Document Type : Research Article


1 Department of Mining & Metallurgical Engineering, Amirkabir University of Technology, Tehran, Iran

2 Department of Mining Engineering, University of Kashan, Kashan, Iran


Providing of semi-autogenous (SAG) mill models for prediction of its effectiveness is one of the most useful tools for better design of grinding circuit. Many SAG mill models have been presented in the literature, but in most of them have not been predicted the mill performance in industrial scale. Semi-autogenous mill power has an effective impact on the mill performance. So in this study, a new model based on combination of radial artificial neural network and principal component is presented to predict semi-autogenous mill power. The feed moisture, mass flowrate, mill load cell weight, SAG mill solid percent, inlet and outlet water to the SAG mill and work index selected as input variables and evaluated the effect of them on the mill power. The results showed that the trained hybrid model of artificial neural network and principal component with R=0.8456 and RMSE= 68.0752 can be used to predict the semi-autogenous mill power in industrial scale. The sensitivity analysis results showed that all model input parameters had a significant effect on the output.


Main Subjects

Principal Component

[1] M. Silva, A. Casali, Modelling SAG milling power and specific energy consumption including the feed percentage of intermediate size particles, Minerals Engineering, 70 (2015) 156-161.
[2] J. Salazar, L. Magne, G. Acuna, F. Cubillos, Dynamic modelling and simulation of semi-autogenous mills, Minerals Engineering, 22(1) (2009) 70-77.
[3] S. Morrell, A new autogenous and semi-autogenous mill model for scale-up, design and optimisation, Minerals Engineering, 17(3) (2004) 437-445.
[4] S. Morrell, A method for predicting the specific energy requirement of comminution circuits and assessing their energy utilisation efficiency, Minerals Engineering, 21(3) (2008) 224-233.
[5] M. Van Nierop, The effect of overloading and premature centrifuging on the power of an autogenous mill, Journal of the Southern African Institute of Mining and Metallurgy, 97(7) (1997) 313-317.
[6] J. Herbst, W. Pate, Object components for comminution system softsensor design, Powder Technology, 105(1) (1999) 424-429.
[7] T.J. Napier-Munn, S. Morrell, R.D. Morrison, T. Kojovic, Mineral comminution circuits: their operation and optimisation, Julius Kruttschnitt Mineral Research Centre, University of Queensland, 1996.
[8] T. Apelt, S. Asprey, N. Thornhill, Inferential measurement of SAG mill parameters, Minerals engineering, 14(6) (2001) 575-591.
[9] F.S. Hoseinian, R.S. Faradonbeh, A. Abdollahzadeh, B. Rezai, S. Soltani-Mohammadi, Semi-autogenous mill power model development using gene expression programming, Powder Technology, 308 (2017) 61-69.
[10] D.W. Patterson, Artificial neural networks: theory and applications, Prentice Hall PTR, 1998.
[11] S.C. Chelgani, B. Shahbazi, B. Rezai, Estimation of froth flotation recovery and collision probability based on operational parameters using an artificial neural network, International Journal of Minerals, Metallurgy, and Materials, 17(5) (2010) 526-534.
[12] M. Curilem, G. Acuña, F. Cubillos, E. Vyhmeister, Neural networks and support vector machine models applied to energy consumption optimization in semiautogeneous grinding, Chemical Engineering Transactions, 25 (2011) 761-766.
[13] T. Valenzuela, K. Carvajal, G. Acuña, M. Chacón, L. Magne, Neural grey box model for power estimation in semiautogenous mill, Advances in Neural Networks–ISNN 2005, (2005) 973-973.
[14] H. Çamdevýren, N. Demýr, A. Kanik, S. Keskýn, Use of principal component scores in multiple linear regression models for prediction of Chlorophyll-a in reservoirs, Ecological Modelling, 181(4) (2005) 581-589.
[15] S. Sousa, F. Martins, M. Alvim-Ferraz, M.C. Pereira, Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations, Environmental Modelling & Software, 22(1) (2007) 97-103.
[16] P.R. Peres-Neto, D.A. Jackson, K.M. Somers, How many principal components? Stopping rules for determining the number of non-trivial axes revisited, Computational Statistics & Data Analysis, 49(4) (2005) 974-997.
[17] J.S. Torrecilla, J. García, E. Rojo, F. Rodríguez, Estimation of toxicity of ionic liquids in Leukemia Rat Cell Line and Acetylcholinesterase enzyme by principal component analysis, neural networks and multiple lineal regressions, Journal of hazardous materials, 164(1) (2009) 182-194.
[18] Y.B. Dibike, D. Solomatine, M.B. Abbott, On the encapsulation of numerical-hydraulic models in artificial neural network, Journal of Hydraulic research, 37(2) (1999) 147-161.
[19] J. Mason, R. Price, A. Tem'Me, A neural network model of rainfall-runoff using radial basis functions, Journal of Hydraulic Research, 34(4) (1996) 537-548.
[20] R.B. Santos, M. Ruppb, S.J. Bonzi, A.M.F. Filetia, Comparison between multilayer feedforward neural networks and a radial basis function network to detect and locate leaks in pipelines transporting gas, Chem. Eng. Trans, 32(1375) (2013) e1380.
[21] M. Khandelwal, D.J. Armaghani, R.S. Faradonbeh, P. Ranjith, S. Ghoraba, A new model based on gene expression programming to estimate air flow in a single rock joint, Environmental Earth Sciences, 75(9) (2016) 1-13.
[22] R.S. Faradonbeh, M. Monjezi, D.J. Armaghani, Genetic programing and non-linear multiple regression techniques to predict backbreak in blasting operation, Engineering with Computers, 32(1) (2016) 123-133.