摘要
在分析松辽盆地区域航放、航磁特征的基础上,研究了钱家店铀矿床航放、航磁特征,建立了基于该矿床航放特征的神经网络预测模型。提取了全区航放铀含量、活性铀、铀增量、相对变迁系数、铀迁移富集系数、地球化学活动性指数等6种航放多元参量,能够反映盆内铀成矿的地质、地球化学背景及水文动力环境,可指示找铀有利部位。据此,圈定了15个砂岩型铀矿找矿预测区,指出除钱家店-宝龙山铀成矿带外,盆地西南部开鲁坳陷和西南隆起区的东来-大林、巴颜塔拉西南、阿鲁科尔沁旗东、奈曼旗西南、新开河北、詹榆-新华-太平川,以及盆地西部斜坡区的双岗、白城东-舍力、泰来北东等地区均具有较好的找矿潜力;盆地西南部伊胡塔镇、林海-小城子,西北部七棵树东等地区也应给予一定的关注。
In this paper,the characteristics of the regional aerogeophysical field and the mineralization background of the sandstone type uranium was analyzed in detail for Songliao basin,The aero-radiometric and aeromagnetic characteristics of Qianjiadian uranium deposit were studied,and the neural network prediction model was established with the aero-radiometric data. Six parameters were extracted for the basin in the active uranium,uranium increment,relative change coefficient,migrationenrichment coefficient of uranium,geochemical activity index,which are important criterions to reflect the uranium metallogenic geology-geochemistry- hydrological dynamic environment and can be used to indicate the favorable uranium prospecting areas. Accordingly,15 sandstone-type prospecting sectors were delineated. It was also pointed out that Donglai-Dalin,southwestern Bayantala,eastern Alukeerqin,southwestern Naiman,northern Xinkaihe,Zhanyu-Xinhua-Taipingchuan,and Shuanggang,eastern Baicheng-Sheli,northeastern Tailai have better uranium prospecting potential. Besides Qianjiadian-Baolongshan uranium metallogenic belt,Yihuta,Linhai-Xiaochengzi,Qikeshu should also be given some attention to in the future exploration.
作者
张翔
张仁红
江民忠
李怀渊
宁媛丽
段晨宇
卢亚运
胡国民
ZHANG Xiang;ZHANG Renhong;JIANG Minzhong;LI Huaiyuan(Airborne Survey and Remote Sensing Center of Nuclear Industry,Shijiazhuang,Hebei 050002,China;Key Laboratory of Uranium Resources Geophysical Exploration Technology,CNNC,Shijiazhuang ,Hebei 050002,China)
出处
《铀矿地质》
CAS
CSCD
2019年第4期229-240,共12页
Uranium Geology
基金
中国核工业地质局项目“松辽盆地砂岩型铀矿磁放重综合预测评价”(编号:201732-1)资助
关键词
松辽盆地
航放航磁
航放多元信息
神经网络预测
找矿潜力
Songliao basin
aeroradiometric and aeromagnetic survey
aeroradiometric multiple information
neural network predicting
prospecting potential