地质学报
地質學報
지질학보
ACTA GEOLOGICA SINICA
2014年
4期
644-657
,共14页
李晓晖%袁峰%张明明%贾蔡%周涛发%张淑虹%郑通科%高道明%洪东良%刘晓明
李曉暉%袁峰%張明明%賈蔡%週濤髮%張淑虹%鄭通科%高道明%洪東良%劉曉明
리효휘%원봉%장명명%가채%주도발%장숙홍%정통과%고도명%홍동량%류효명
三维地质建模%人工神经网络%成矿预测%空间分析%白象山%宁芜盆地
三維地質建模%人工神經網絡%成礦預測%空間分析%白象山%寧蕪盆地
삼유지질건모%인공신경망락%성광예측%공간분석%백상산%저무분지
3D geological modelling%Artificial Neural Network%prospectivity mapping%3D spatial analysis%Baixiangshan,Ningwu Basin
本文基于三维地质环境,综合白象山矿区积累的地质资料和物探成果,首先开展三维地质建模工作,详细刻画了白象山矿区的三维地质结构;在三维地质模型基础上,利用三维空间分析手段对三维控矿因素进行定量挖掘,提取了多种三维控矿因素;最后采用人工神经网络方法进行三维成矿定位预测。预测结果显示,人工神经网络三维成矿定位预测能很好的定位出已知矿体,同时显示,在已知矿体北部及东部的深边部具有较高的成矿概率,可作为开展进一步找矿勘探的靶区。因此,人工神经网络三维成矿定位预测对于白象山矿区的应用是有效的,可服务于新老矿区的深边部三维成矿定位预测,同时可为隐伏矿、盲矿的成矿预测和优选靶区提供定量、定位新的方法和途径。
本文基于三維地質環境,綜閤白象山礦區積纍的地質資料和物探成果,首先開展三維地質建模工作,詳細刻畫瞭白象山礦區的三維地質結構;在三維地質模型基礎上,利用三維空間分析手段對三維控礦因素進行定量挖掘,提取瞭多種三維控礦因素;最後採用人工神經網絡方法進行三維成礦定位預測。預測結果顯示,人工神經網絡三維成礦定位預測能很好的定位齣已知礦體,同時顯示,在已知礦體北部及東部的深邊部具有較高的成礦概率,可作為開展進一步找礦勘探的靶區。因此,人工神經網絡三維成礦定位預測對于白象山礦區的應用是有效的,可服務于新老礦區的深邊部三維成礦定位預測,同時可為隱伏礦、盲礦的成礦預測和優選靶區提供定量、定位新的方法和途徑。
본문기우삼유지질배경,종합백상산광구적루적지질자료화물탐성과,수선개전삼유지질건모공작,상세각화료백상산광구적삼유지질결구;재삼유지질모형기출상,이용삼유공간분석수단대삼유공광인소진행정량알굴,제취료다충삼유공광인소;최후채용인공신경망락방법진행삼유성광정위예측。예측결과현시,인공신경망락삼유성광정위예측능흔호적정위출이지광체,동시현시,재이지광체북부급동부적심변부구유교고적성광개솔,가작위개전진일보조광감탐적파구。인차,인공신경망락삼유성광정위예측대우백상산광구적응용시유효적,가복무우신로광구적심변부삼유성광정위예측,동시가위은복광、맹광적성광예측화우선파구제공정량、정위신적방법화도경。
This study integrates geological and geophysical data for the Baixiangshan iron deposit within the Ningwu Basin of the Middle and Lower Yangtze Metallogenic Belt,China,creates a three-dimensional geological model that can describes three-dimensional geological structure of Baixiangshan mining area in detail.Based on three-dimensional geological models,several quantitative exploration criteria have been extracted by using 3D spatial analysis,which were then used for the prospectivity mapping by using Artificial Neural Networks.The result shows that the prospectivity map can identifies the known orebody very well and highlights some high prospectivity areas in the north and east of known orebody that are favorable for exploration in the Baixiangshan mining area;these areas should be considered as high priority targets for future exploration. In addition, the result shows that 3D localization and quantitative prospectivity mapping by using Artificial Neural Networks is effective for the Baixiangshan Mining area.It can be used for the future mineral exploration in deep and peripheral areas of the new and old mining area and also provide a new localization and quantitative method for predicting the concealed mineralization and optimizing the prospecting target.