Research and application of the valuation method for coal and rock digital modeling

Based on deterministic data of drills and other geophysical information, common interpolation methods were studied based on given geological information to speculate unknown spatial information. And summarized the conditions of proper application and advantages and disadvantages of trend surface, weighted least-square fitting, spline function, linear interpolation of triangulation, multi-faceted function, inverse distance square and mobile fitting. It is proposed that the effective optimized methods of parameter selection combined different existing conditions of coal-bed and rock and distribution characteristics of given data. During applying either multi-faceted function or mobile fitting, the article analyzed main factors of causing pathological equation groups and made use of mathematical methods to improve the distortion phenomenon of elevation value. Taking 1# coal of Dalianhe surface coal mine it as an example, we apply the above interpolation methods to analyze and compare their limitations and superiority, and use the cross validation test for verifying. The above research results show that triangulation linear interpolation has strong stability, especially suitable to contain orebody of fault.The root mean square is relatively the least.