基于BP神经网络的湘西金矿成矿预测  

Metallogenic prediction of Xiangxi gold deposit based on BP neural networks

作　　者：邵拥军[1] 贺辉[1] 张贻舟[1] 梁恩云[1] 丁宗炜[1] 陈星霖[1] 刘忠法[1] 

机构地区：[1]中南大学地学与环境工程学院,湖南长沙410083

出　　处：《中南大学学报（自然科学版）》2007年第6期1192-1198,共7页Journal of Central South University:Science and Technology

基　　金：国家科技攻关项目(2001BA609A-06);国家"十一五"科技支撑计划项目(2006BAB01B07)

摘　　要：从变异函数特征和分形维数特征方面,对在湘西金矿深部新发现的V7和V8矿脉与原有的V1,V2,V3和V4矿脉的成矿特征进行对比研究。采用BP神经网络技术对成矿特征图形进行模糊识别并进行定量化的相似类比研究,最后依据类比的结果对深部盲矿体的成矿富集规律进行预测。研究结果表明:新发现的V7矿脉具有广阔的找矿前景,而V8矿化不稳定。From the characteristics of the variation function and fractal dimension, the comparative research on the old veins V1, V2, V3 and V4 with new veins V7 and V8 in the deep of Xiangxi gold deposit was carried out. By the BP neural network technology, the fuzzy recognitions on metallogenic characteristic pattern and the quantitatively similar analogy were investigated. According to the similar analogy result the prediction on the metallogenic enrichment rule of the deep blind ore bodies was done. The predictive result shows that the vein V7 has a broader exploring foreground than the vein V8.

关 键 词：湘西金矿 变异函数 分形维数 BP神经网络 成矿预测 

分 类 号：B659[哲学宗教—外国哲学]