Adapting pattern recognition approach for uncertainty assessment in the geologic resource estimation for Indian iron ore mines

The Geologic resource estimation requires the accurate prediction of the regionalized variables such as ore grade at an un-sampled location with the knowledge of sparse borehole information. It plays prominent role in the decision-making process for investment and development of various mining projects and hence judicious selection of the assessment method is essential for making profitable investment. The grade and quantity of the ore varies with spatial coordinates and directions and thus spatial uncertainty has to be taken into consideration for improvement in estimation of minerals deposits. Traditional geostatistical approaches such as ordinary kriging (OK), Inverse Distance weighing (IDW) and object based approach are still being used for the purpose, but because of its limited capability of honoring the statistics up to the second order, they cannot faithfully represent the complex spatial variability of minerals deposit. The remarkable feature of various pattern recognition techniques to capture the inherent patterns in the complex data made them suitable for the geologic resource estimation. This paper describes the use of two distinct pattern recognition techniques: support vector regression and Gaussian process regression to assess the mineral grade of one of the Indian iron ore mines from eastern region of the country. The spatial coordinates and multiple types of lithology were taken as input variables and iron ore grade as an output variable. The comparative analysis of these models was carried out, and the results obtained were validated with traditional geostatistical method: Ordinary Kriging (OK). The various performance measures such as root mean square error (RMSE), and coefficient of determination (R2) were used to evaluate the performance of the different models. It is found that SVR and GPR provide significant improvement in resource estimation.

[1]  János Fodor,et al.  Traditional and New Ways to Handle Uncertainty in Geology , 2001 .

[2]  Sukumar Bandopadhyay,et al.  Comparative Evaluation of Neural Network Learning Algorithms for Ore Grade Estimation , 2006 .

[3]  R. Olea.  Geostatistics for Natural Resources Evaluation By Pierre Goovaerts, Oxford University Press, Applied Geostatistics Series, 1997, 483 p., hardcover, $65 (U.S.), ISBN 0-19-511538-4 , 1999 .

[4]  Sukumar Bandopadhyay,et al.  Machine Learning Algorithms and Their Application to Ore Reserve Estimation of Sparse and Imprecise Data , 2010, J. Intell. Learn. Syst. Appl..

[5]  Pejman Tahmasebi,et al.  A hybrid neural networks-fuzzy logic-genetic algorithm for grade estimation , 2012, Comput. Geosci..

[6]  Rajive Ganguli,et al.  Sparse Data Division Using Data Segmentation and Kohonen Network for Neural Network and Geostatistical Ore Grade Modeling in Nome Offshore Placer Deposit , 2004 .

[7]  Snehamoy Chatterjee,et al.  General regression neural network residual estimation for ore grade prediction of limestone deposit , 2007 .

[8]  Muhsin Tunay Gençoglu,et al.  Prediction of flashover voltage of insulators using least squares support vector machines , 2009, Expert Syst. Appl..

[9]  Masoud Shariat Panahi,et al.  The application of median indicator kriging and neural network in modeling mixed population in an iron ore deposit , 2011, Comput. Geosci..

[10]  Alexei Pozdnoukhov,et al.  Machine Learning for Spatial Environmental Data: Theory, Applications, and Software , 2009 .

[11]  Jin Li,et al.  Application of machine learning methods to spatial interpolation of environmental variables , 2011, Environ. Model. Softw..

[12]  Sukumar Bandopadhyay,et al.  A hybrid ensemble model of kriging and neural network for ore grade estimation , 2006 .

[13]  J. Chilès,et al.  Geostatistics: Modeling Spatial Uncertainty , 1999 .

[14]  Saro Lee,et al.  Application of Artificial Neural Network for Gold–Silver Deposits Potential Mapping: A Case Study of Korea , 2010 .

[15]  Michael E. Tipping.  Sparse Bayesian Learning and the Relevance Vector Machine , 2001, J. Mach. Learn. Res..

[16]  Michito Ohmi,et al.  Neural Network-Based Estimation of Principal Metal Contents in the Hokuroku District, Northern Japan, for Exploring Kuroko-Type Deposits , 2002 .

[17]  Pejman Tahmasebi,et al.  Application of a Modular Feedforward Neural Network for Grade Estimation , 2011 .

[18]  Shahoo Maleki,et al.  Estimation of Iron concentration by using a support vector machineand an artificial neural network - the case study of the Choghart deposit southeast of Yazd, Yazd, Iran , 2014 .

[19]  Tao Yu,et al.  Adaptive spherical Gaussian kernel in sparse Bayesian learning framework for nonlinear regression , 2009, Expert Syst. Appl..

[20]  Xiao-li Li,et al.  Hybrid self-adaptive learning based particle swarm optimization and support vector regression model for grade estimation , 2013, Neurocomputing.

[21]  Mahmud Güngör,et al.  Generalized Regression Neural Networks and Feed Forward Neural Networks for prediction of scour depth around bridge piers , 2009, Adv. Eng. Softw..

[22]  Cheng Wu,et al.  Robust LS-SVM regression for ore grade estimation in a seafloor hydrothermal sulphide deposit , 2013, Acta Oceanologica Sinica.

[23]  S. Dutta,et al.  Evaluation of artificial neural networks and kriging for the prediction of arsenic in Alaskan bedrock-derived stream sediments using gold concentration data , 2007 .