摘要
木吉-乌孜别里山口一带位于印度-欧亚大陆碰撞造山带西段的帕米尔构造结,发育多处金铜等矿床(点),砂金矿床大量发育,显示巨大的金矿找矿潜力。在总结木吉一带金矿区地质特征基础上,剖析区域Au元素异常,提取区内各类遥感蚀变异常,构建地质-地球化学-遥感综合预测模型,利用随机森林算法,开展多信息集成的定量预测。基于随机森林算法定量预测找矿概率,结合研究区目前的研究现状及已知矿床点的类型、数量、代表性、规律等信息,对找矿靶区进行优选,圈定A级找矿靶区3个,B级找矿靶区2个,C级找矿靶区1个。基于随机森林算法的机器学习,在地物化遥感多数据区域中,预测精度较好,提高了预测效率,为该区域实现高效定量预测提供了依据。
The Pamir tectonic junction in the western section of the Indian-Eurasian continental collision orogenic belt is located in the muji-Wuzibieli mountain pass.It is found that there are many gold-copper deposits(points),and a large number of gold deposits are developed,showing great gold prospecting potential.With the deepening of geological prospecting work in the area,a large amount of geological prospecting information has been accumulated.Mature prediction theory and method are needed to obtain the distribution location,output probability and resource potential of mineral resources in the area,so as to achieve efficient metallogenic prediction.On the basis of summarizing the geological characteristics of the gold mining area in Muji area,the regional Au element anomaly is analyzed,and various remote sensing alteration anomalies in the area are extracted.The geological-geochemical-remote sensing comprehensive prediction model is constructed,and the quantitative prediction of multi-information integration is carried out by using the random forest algorithm.Based on the random forest algorithm to quantitatively predict the probability of prospecting,combined with the current research status of the study area and the type,quantity,representativeness and regularity of known ore deposits,the prospecting target areas are optimized,and three A-level prospecting target areas are delineated.Two B-level prospecting target areas and one C-level prospecting target area.The machine learning based on random forest algorithm has better prediction accuracy and improves the efficiency of prediction in the multi-data area of geophysical and geochemical remote sensing,which provides a basis for efficient quantitative prediction in this area.
作者
呼冬强
何福宝
李辉
郝延海
张强
冯昌荣
廖风云
Hu Dongqiang;He Fubao;Li Hui;Hao Yanhai;Zhang qiang;Feng Changrong;Liao Fengyun(School of Mining Engineering and Geology,Xinjiang Institute of Engineering,Urumqi,Xinjiang,830023,China;The Second Geological Brigade of Xinjiang Bureau of Geology and mineral resources,Kashi,Xinjiang,844000,China)
出处
《新疆地质》
CAS
CSCD
2024年第1期158-163,共6页
Xinjiang Geology
基金
新疆地质局自筹资金项目
克州战略性矿产资源成矿规律与找矿靶区优选(XGMB202363)资助。
关键词
成矿预测
金矿床
随机森林
机器学习
Metallogenic prediction
Gold deposits
Random forest
Machine learning
