基于神经网络模型的金矿成矿远景预测——以白马山-龙山地区为例

PREDICTION OF GOLD METALLOGENIC PROSPECT BASED ON THE NEURAL NETWORK MODEL:A Case Study of the Baimashan-Longshan Area in Hunan Province

摘要: 白马山-龙山地区是湘中重要的金锑成矿带.以1：20万水系沉积物检测数据为基础，结合区域地层、岩性、构造、岩浆岩等地质信息，在全区19 100 km²范围内以4 km²为评价单元，应用神经网络模型对区域金矿成矿远景进行预测，共划分了4个成矿远景区.

Abstract: The Baimashan-Longshan area is the significant metallogenic zone of gold and antimony in central Hunan Province. Based on the data of 1:200 000 stream sediment survey, in combination with the geological information of stratigraphy, lithology, structure and magmatite in this area, the neural network model is adopted to predict the regional gold metallogenic prospect. The whole area of 19 100 km² is assessed by 4 km² as evaluation unit. Then four metallogenic prospect areas are delineated.