摘要
黏土矿物不仅对寻找黏土类矿物资源具有指示意义,而且对于区域古气候、环境污染、石油勘探等研究都具有重要意义。但黏土矿物空间分布获取较为困难,目前国内积累了覆盖大部分面积的中大比例尺地球化学数据,具有分析元素多、精度高的特征,基于地球化学元素含量开展黏土矿物预测,可以为黏土矿物的空间分布研究提供新思路。如何建立地球化学元素含量和黏土矿物之间的复杂关系,是基于地球化学元素含量开展黏土矿物空间分布预测的关键之一。机器学习和人工智能算法的发展为建立地球化学元素与黏土矿物之间的非线性关系提供了可能,本研究基于河北平原多目标1∶20万地球化学数据和111个表层土壤黏土矿物分析数据,利用BP神经网络建立河北平原区地球化学元素含量与黏土矿物之间的预测模型,对河北平原区高岭石、绿泥石、伊利石的空间分布进行预测。研究发现河北平原区高岭石、绿泥石、伊利石等的黏土矿物分布与研究区地貌特征、土壤类型以及人为因素等都有联系,高岭石、绿泥石、伊利石在整个冀东平原区以及北三县平原区含量普遍偏低,在太行山山前平原区含量普遍偏高;黏土矿物在冲积扇平原、冲积平原含量偏高,在洪积平原含量较低;河北平原区的黏土矿物在保定和唐山等地区富集。
The clay mineral not only has an indication meaning for clay mineral resources prospecting,but also has a great significance for the research of regional paleoclimate,environmental pollution and petroleu exploration.However,it is difficult to obtain the spatial distribution of clay mineral.At present,the medium-large scale geochemical data covering most of the areas are accumulated in the country,which have the characteristics of many analysis elements with high precision.The prediction of clay minerals based on the contents of geochemical elements can provide a new idea for the study of the spatial distribution of clay minerals.How to establish the complex relationship between the content of geochemical element and clay mineral is one of the key to predict the spatial distribution of the clay minerals based on the content of geochemical element.The development of the machine learning and artificial intelligence algorithm provides a possibility for the establishment of the nonlinear relationship between the geochemical elements and the clay minerals.The study is based on the multi-objective 1:200,000 geochemical data of the Hebei Plain and the analysis data of clay minerals of 111 surface soil samples.The spatial distribution of kaolinite,chlorite and illite in the Hebei Plain is predicted by using the BP neural network to establish a prediction model between the contents of geochemical elements and the clay minerals in the Hebei Plain.It is found that the distribution of clay minerals such as kaolinite,chlorite and illite in the Hebei Plain is related to the features of the study area,the soil type and the human factor.The contents of kaolinite,chlorite and illite in the plain areas of the East Hebei and the northern three counties are generally low.The contents of clay minerals are high in the piedmont plain area of Taihang Mountain.The contents of clay minerals in the alluvial fan plain and the alluvial plain are high,and the contents of clay minerals are low in the aggraded flood plain.The clay minerals in th
作者
任钰
汪海城
张生元
宋泽峰
陈文静
REN Yu;WANG Haicheng;ZHANG Shengyuan;SONG Zefeng;CHEN Wenjing(College of Information Engineering,Hebei GEO University,Shijiazhuang 050031,Hebei,China;Institute of Resource and Environmental Engineering,Hebei GEO University,Shijiazhuang 050031,Hebei,China)
出处
《矿产与地质》
2020年第1期81-90,共10页
Mineral Resources and Geology
基金
国家自然科学基金青年项目“基于贝叶斯最大熵的覆盖区地球化学异常信息识别”(No.41802249)
国家重点研发计划项目“深部矿产资源评价理论与方法”课题(No.2016YFC0600500)
“全球主要构造成矿域资源评价与预测”项目共同资助.
