Exploratory analysis of multivariate drill core time series measurements

Abstract

Demand for mineral resources is increasing, necessitating exploitation of lower grade and more heterogeneous orebodies. The high variability inherent in such orebodies leads to an increase in the cost, complexity and environmental footprint associated with mining and mineral processing. Enhanced knowledge of orebody characteristics is thus vital for mining companies to optimize profitability. We present a pilot study to investigate prediction of geometallurgical variables from drill sensor data. A comparison is made of the performance of multilayer perceptron (MLP) and multiple linear regression models (MLR) for predicting a geometallurgical variable. This comparison is based on simulated data that are physically realistic, having been derived from models fitted to the one available drill core. The comparison is made in terms of the mean and standard deviation (over repeated samples from the population) of the mean absolute error, root mean square error, and coefficient of determination. The best performing model depends on the form of the response variable and the sample size. The standard deviation of performance measures tends to be higher for the MLP, and MLR appears to offer a more consistent performance for the test cases considered.

References

• R. M. Balabin and S. V. Smirnov. Interpolation and extrapolation problems of multivariate regression in analytical chemistry: Benchmarking the robustness on near-infrared (NIR) spectroscopy data”. Analyst 137.7 (2012), pp. 1604–1610. doi: 10.1039/c2an15972d
• C. M. Bishop. Pattern recognition and machine learning. Springer, 2006. url: https://link.springer.com/book/9780387310732
• J. B. Boisvert, M. E. Rossi, K. Ehrig, and C. V. Deutsch. Geometallurgical modeling at Olympic dam mine, South Australia”. Math. Geosci. 45 (2013), pp. 901–925. doi: 10.1007/s11004-013-9462-5
• T. Bollerslev. Generalized autoregressive conditional heteroskedasticity”. J. Economet. 31.3 (1986), pp. 307–327. doi: 10.1016/0304-4076(86)90063-1
• C. Both and R. Dimitrakopoulos. Applied machine learning for geometallurgical throughput prediction—A case study using production data at the Tropicana Gold Mining Complex”. Minerals 11.11 (2021), p. 1257. doi: 10.3390/min11111257
• J. Chen and G. Li. Tsallis wavelet entropy and its application in power signal analysis”. Entropy 16.6 (2014), pp. 3009–3025. doi: 10.3390/e16063009
• S. Coward, J. Vann, S. Dunham, and M. Stewart. The primary-response framework for geometallurgical variables”. Seventh international mining geology conference. 2009, pp. 109–113. https://www.ausimm.com/publications/conference->url: https://www.ausimm.com/publications/conference- proceedings/seventh-international-mining-geology- conference-2009/the-primary-response-framework-for- geometallurgical-variables/
• A. C. Davis and N. B. Christensen. Derivative analysis for layer selection of geophysical borehole logs”. Comput. Geosci. 60 (2013), pp. 34–40. doi: 10.1016/j.cageo.2013.06.015
• C. Dritsaki. An empirical evaluation in GARCH volatility modeling: Evidence from the Stockholm stock exchange”. J. Math. Fin. 7.2 (2017), pp. 366–390. doi: 10.4236/jmf.2017.72020
• R. F. Engle and T. Bollerslev. Modelling the persistence of conditional variances”. Econ. Rev. 5.1 (1986), pp. 1–50. doi: 10.1080/07474938608800095
• A. S. Hadi and R. F. Ling. Some cautionary notes on the use of principal components regression”. Am. Statistician 52.4 (1998), pp. 15–19. doi: 10.2307/2685559
• J. Hunt, T. Kojovic, and R. Berry. Estimating comminution indices from ore mineralogy, chemistry and drill core logging”. The Second AusIMM International Geometallurgy Conference (GeoMet) 2013. 2013, pp. 173–176. http://ecite.utas.edu.au/89773>url: http://ecite.utas.edu.au/89773 on p. C210).
• R. Hyndman, Y. Kang, P. Montero-Manso, T. Talagala, E. Wang, Y. Yang, M. O’Hara-Wild, S. Ben Taieb, H. Cao, D. K. Lake, N. Laptev, and J. R. Moorman. tsfeatures: Time series feature extraction. R package version 1.0.2. 2020. https://CRAN.R-project.org/package=tsfeatures>url: https://CRAN.R-project.org/package=tsfeatures on p. C222).
• C. L. Johnson, D. A. Browning, and N. E. Pendock. Hyperspectral imaging applications to geometallurgy: Utilizing blast hole mineralogy to predict Au-Cu recovery and throughput at the Phoenix mine, Nevada”. Econ. Geol. 114.8 (2019), pp. 1481–1494. doi: 10.5382/econgeo.4684
• E. B. Martin and A. J. Morris. An overview of multivariate statistical process control in continuous and batch process performance monitoring”. Trans. Inst. Meas. Control 18.1 (1996), pp. 51–60. doi: 10.1177/014233129601800107
• E. Sepulveda, P. A. Dowd, C. Xu, and E. Addo. Multivariate modelling of geometallurgical variables by projection pursuit”. Math. Geosci. 49.1 (2017), pp. 121–143. doi: 10.1007/s11004-016-9660-z
• S. J. Webb, G. R. J. Cooper, and L. D. Ashwal. Wavelet and statistical investigation of density and susceptibility data from the Bellevue drill core and Moordkopje borehole, Bushveld Complex, South Africa”. SEG Technical Program Expanded Abstracts 2008. Society of Exploration Geophysicists, 2008, pp. 1167–1171. doi: 10.1190/1.3059129
• R. Zuo. Identifying geochemical anomalies associated with Cu and Pb–Zn skarn mineralization using principal component analysis and spectrum–area fractal modeling in the Gangdese Belt, Tibet (China)”. J. Geochem. Explor. 111.1-2 (2011), pp. 13–22. doi: 10.1016/J.GEXPLO.2011.06.012