Dimensioning a stockpile operation using principal component analysis

Abstract

Mineral processing plants generally have narrow tolerances for the grades of their input raw materials, so stockpiles are often maintained to reduce material variance and ensure consistency. However, designing stockpiles has often proven difficult when the input material consists of multiple sub-materials that have different levels of variances in their grades. In this paper, we address this issue by applying principal component analysis (PCA) to reduce the dimensions of the input data. The study was conducted in three steps. First, we applied PCA to the input data to transform them into a lower-dimension space while retaining 80% of the original variance. Next, we simulated a stockpile operation with various geometric stockpile configurations using a stockpile simulator in MATLAB. We used the variance reduction ratio as the primary criterion for evaluating the efficiency of the stockpiles. Finally, we used multiple regression to identify the relationships between stockpile efficiency and various design parameters and analyzed the regression results based on the original input variables and principal components. The results showed that PCA is indeed useful in solving a stockpile design problem that involves multiple correlated input-material grades.

Keywords

bed-blending / mining / stockpile / principal component analysis / multiple regression

References

[1] Gy PM. Sampling of ores and metallurgical products during continuous transport. Trans. IMM, 1965, 74, 165

[2] Gy PM. A new theory of bed-blending derived from the theory of sampling—Development and full-scale experimental check. Int. J. Miner. Process., 1981, 8, 201.

[3] Gy PM. Optimizing the operational strategy of a mine-metallurgy or quarry-cement works complex. Can. Metall. Q., 1999, 38, 157.

[4] Zhao S, Lu TF, Koch B, Hurdsman A. Automatic quality estimation in blending using a 3D stockpile management model. Adv. Eng. Inf., 2015, 29, 680.

[5] Zhao S, Lu TF, Koch B, Hurdsman A. 3D stockpile modelling and quality calculation for continuous stockpile management. Int. J. Miner. Process., 2015, 140, 32.

[6] Robinson GK. How much would a blending stockpile reduce variation?. Chemom. Intell. Lab. Syst., 2004, 74, 121.

[7] Gy PM. Sampling of Heterogeneous and Dynamic Material Systems: Theories of Heterogeneity, Sampling and Homogenizing, 1992, Amsterdam, Elsevier

[8] Dowd PA. The design of a rock homogenizing stockpile. Miner. Process. UK, 1989 63

[9] Kumral M. Bed blending design incorporating multiple regression modelling and genetic algorithms. J. South Afr. Inst. Min. Metall., 2006, 106, 229

[10] Sreejith J, Ilangovan S. Optimization of wear parameters of binary Al-25Zn and Al-3Cu alloys using design of experiments. Int. J. Miner. Metall. Mater, 2018, 25, 1465.

[11] Hosseini E, Rashchi F, Ataie A. Ti leaching from activated ilmenite-Fe mixture at different milling energy levels. Int. J. Miner. Metall. Mater., 2018, 25, 1263.

[12] De Werk M. Trade-off Between Cost and Performance in Chevron Bed-Blending, 2017, Montreal, McGill University

[13] Paterson DG, Mushia MN, Mkula SD. Effects of stockpiling on selected properties of opencast coal mine soils. S. Afr. J. Plant Soil, 2019, 36, 101.

[14] Li N, Li JX, Long HM, Chun TJ, Mu GT, Yu ZW. Optimization method for iron ore blending based on the sintering basic characteristics of blended ore. TMS Annual Meeting & Exhibition, 2018, Cham, Springer, 455

[15] Singh V, Biswas A, Tripathy SK, Chatterjee SK, Chakerborthy TK. Smart ore blending methodology for ferro-manganese production process. Ironmaking Steelmaking, 2016, 43, 481.

[16] Marques DM, Costa JFCL. An algorithm to simulate ore grade variability in blending and homogenization piles. Int. J. Miner. Process., 2013, 120, 48.

[17] Abdi H, Williams LJ. Principal component analysis. Wiley Interdiscip. Rev. Comput. Stat., 2010, 2, 433.

[18] Agresti A. Foundations of Linear and Generalized Linear Models, 2015, Hoboken, John Wiley & Sons, Inc.