使用与自组织神经网聚类相结合的支持向量回归机预测模型对矿体体素品位进行插值,并与多边形法、距离幂次反比法、克里格法进行对比验证.结果表明,该预测模型进行品位插值具备很好的可行性和可靠性.
The method of support vector regression (SVR) in combination with self organization feature mapping (SOFM) network was selected for grade interpolation in orebody, and was compared to the Thiessen polygons method, the distance power inverse ratio method and the Kriging method. The result shows that the prediction model of SVR is feasible and reliable for grade estimation.