陆相页岩气资源评价中人工智能算法的探索——以苏北盆地溱潼凹陷为例  

The tentative application of artificial intelligence algorithm to evaluating continental shale gas resources:A case study of Qintong sag in Subei(North Jiangsu) Basin

作　　者：谈迎[1] 杨伟松[2] 李振生[3] 

机构地区：[1]江苏省有色金属华东地质勘查局,江苏南京210000 [2]江西科技师范大学通讯与电子学科,江西南昌330013 [3]合肥工业大学资源与环境工程学院,安徽合肥230009

出　　处：《物探与化探》2018年第1期96-103,共8页Geophysical and Geochemical Exploration

基　　金：国家自然科学基金项目(41372194)

摘　　要：将下扬子区苏北盆地溱潼凹陷作为示范区,探讨了人工智能该如何进行页岩气资源评价的问题。给出了可采用的运算方法的组合设计,针对工区陆相页岩气资源评价参数进行了全过程的示范操作:用变形结构分区的方法在平面上将溱潼凹陷分成14个样品区,用模糊数学的方法对定性参数进行定量、半定量取值打分,用因子分析方法对平面上14个样品区进行了单层评价;以平面上14个样品区单层评价为基础,给出了14个样品区5个页岩气目的层在垂向上的叠合评价。为了解决叠合评价时样品区各单层数据的等价性,引入全零值和分级值进行评价、排优。这种针对页岩气资源评价的定性、定量和多层混合问题进行定量综合评价的组合运算方法,可以被开发页岩气资源人工智能评价系统所借鉴。Taking Qintong sag of Subei basin in lower Yangtze area as the demonstration area, the authors investigated the evaluation of shale gas resources with the artificial intelligence method. With the combination algorithm designs, the authors conducted the whole process of demonstration operation based on continental shale gas resources evaluation parameters of the work area. First, Qintong sag was divided into 14 sample areas by the method of deformation structure partition; then, using the fuzzy mathematics method, the authors obtained quantitative and half-quantitative parameter values; after that, using factor analysis method the authors conducted the evaluation of 14 sample areas in a single plane; finally, composite evaluation of five shale gas layers in vertical direction of 14 sample areas was made based on the evaluation of 5 single planes. In order to get the equivalence data of each single layer, the authors introduced the zero value and grading values. The practise of the combination algorithm designs has the reference value for the shale gas resources evaluation system of artificial intelligence.

关 键 词：陆相页岩气 溱潼凹陷 模糊综合预测 人工智能 算法设计 

分 类 号：P631[天文地球—地质矿产勘探]