摘要
内蒙古深部矿体受构造改造与覆盖层影响显著,传统找矿易受多解性制约。本文基于成矿地质规律,构建融合地质、物化探与遥感数据的三维特征体系,引入信息增益赋权与支持向量机模型,实现深部成矿概率定量表征。通过典型靶区钻探验证,模型在预测准确率与靶区圈定效率方面均显著优于传统方法,为复杂成矿区深部勘查提供有效技术路径。
Deep ore bodies in Inner Mongolia are strongly affected by structural overprinting and cover,making traditional prospecting highly ambiguous.This study builds a 3D feature system integrating geological,geophysical-geochemical,and remote-sensing data,and applies information-gain weighting with a support vector machine to quantify deep metallogenic probability.Drilling results show higher prediction accuracy and target delineation efficiency than traditional methods,offering an effective approach for deep exploration in complex metallogenic zones.
作者
其格勒
张欣
张虎
王奇
孙雪敏
Qi Gele;Zhang Xin;Zhang Hu;Wang Qi;Sun Xuemin(Inner Mongolia Land Resources Exploration and Development Co.,Ltd.,Hohhot,China)
出处
《科学技术创新》
2026年第3期85-88,共4页
Scientific and Technological Innovation
关键词
内蒙古地区地质构造
深部找矿
预测技术
geological structure of Inner Mongolia
deep prospecting
prediction technology
