Estimation of coal quality variables by covariance matching constrained kriging Kömür kalite deǧişkenlerinin kovaryans eşlemeli krigleme yöntemi ile kestirimi


ERTUNÇ G., HİNDİSTAN M. A., TERCAN A. E., ÜNVER B.

23rd International Mining Congress and Exhibition of Turkey, IMCET 2013, Antalya, Turkey, 16 - 19 April 2013, vol.3, pp.1785-1792 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 3
  • City: Antalya
  • Country: Turkey
  • Page Numbers: pp.1785-1792
  • Hacettepe University Affiliated: Yes

Abstract

Covariance matching constrained kriging (CMCK) can be described as an hybrid method that combines local accuracy propety of ordinary kriging (OK) and reproduction of spatial variability property of geostatistical simulation. CMCK estimator is built under constraints such as unbiasedness and vanance-covanance matching of predictor and corresponding predictants. In this study, mean lower heating values of lignite seam blocks are estimated by using CMCK and OK. The seam has an average dip of 7°. When estimating faulted and folded blocks, the estimates follow a "zebraic pattern" which is undoubtedly unrealistic. In order to prevent this phenomenon, blocks and composites are transformed onto the horizontal plane by using rotation equations. Estimations are for unfolded blocks and in the end blocks are back transformed into their original position. Results of CMCK and OK are compared by means of image maps, histograms, vanograms and quality-tonnage curves.