Three-Dimensional Distribution Analysis of Phosphorus Content of Limestone Through a Combination of Geostatistics and Artificial Neural Network

摘要: One of the factors that determines suitability limestone for industrial use and its commercial value is phosphorus (P) content, i.e., weight percentage contained in small quantities limestone. Because P content changes locally, geostatistical techniques including semivariogram, ordinary kriging, conditional indicator sequential simulation were used this study to identify spatial correlation estimate three-dimensional distribution an open-pit mine. The data at 43,000 points five different bench levels analyzed. It was found horizontal semivariograms produced by using same level show anisotropic behavior are represented sum two spherical models with ranges sills. twelve vertical also constructed from boring cores. After these classified into four types, a multilayered neural network applied clarify each one. assigned arbitrary grid point area criterion type as one estimated borehole site producing semivariogram nearest point. With technique, kriging combined provided proper estimation depth direction.