The Indosinian and Yanshanian orogenic movements are both important Mesozoic orogenies in eastern China. The resulted tectonic belts are neither products of the third stage of crustal evolution, as proposed by Chen Guoda, nor intra-continental(or intraplate) orogenic belts generated by intraplate dynamics, as argued by some scholars —rather, they are superposed orogenic belts formed on the pre-existing continental crust in eastern China due to Mesozoic Paleo-Pacific dynamic system. In the past, these orogenic belts were called the marginal Pacific epicontinental activation belts of eastern China. In the Mesozoic, under the effect of Paleo-Pacific dynamic system, the East Asia margin orogenic system formed along Northeast Russia-Sikhote Alin(Russia)–Japan-Ryukyu-Taiwan(China)-Palawan(Philippines)regions, while simultaneously the Mesozoic superposed orogenic system formed in the pre-existing continental crust in eastern China adjacent to the East Asia continental margin. The two orogenic systems,both driven by Mesozoic Paleo-Pacific dynamic system, developed synchronously to form the giant Mesozoic orogenic system in the Pacific tectonic domain in eastern Asia, radically changing the pre-Indosian tectonic framework of the area.
Mineral knowledge acquisition is indispensable in geological research. Currently, mineral knowledge can be acquired though a search engine like Google, or though some mineral databases. But usually, the answers from the search engines are not professional enough, and the mineral databases cannot answer the questions input in natural language. In this paper, a Chinese mineral question and answering system based on knowledge graph is proposed with the aim of answering the mineral questions in Chinese natural language professionally, accurately and efficiently. Two deep learning models based on BERT are built to obtain mineral entities and relationships to construct the mineral knowledge graph. After the mineral knowledge graph is constructed, accordingly question templates are defined and another deep learning model based on BERT is built to recognize the intent and the entity/attribute of the input question. The query statement is generated according to the question template pre-defined and is input to the mineral knowledge graph constructed. Then the answer to the question is returned by the knowledge graph. The knowledge graph we constructed has 22,568 entities and 91,699 relationships, which is richer than most of the geological knowledge graph. The question and answering system achieve a 91.2% accuracy on the 2,000 test questions.
With an aim at analyzing the suitability of development of urban underground space in Shenzhen， this paper first studies its relationship with various factors， such as geological environment， surface environment， economic development and geological disasters. A suitability evaluation system for the development and utilization of shallow underground space was established， and the index weight was determined by the index scaling method and the analytic hierarchy process （AHP） method. It is shown that the evaluation indicators of the suitability of development and utilization of the shallow underground space in Shenzhen include six categories and 14 indicators. The suitability evaluation results are divided into four grades: suitable， sub-suitable， average and poor. For Shenzhen， the suitable and sub-suitable zones are concentrated in Futian District， Nanshan District and Bao’an District. The total area of suitable and sub-suitable zones accounts for about 80% of the whole city， indicating that the overall suitability of underground space development in Shenzhen is good. The evaluation results can provide a theoretical basis for the development planning of urban underground space in Shenzhen， and the evaluation approach can be applied to other cities with similar geological conditions to Shenzhen.
扇研究一直是沉积学中充满活力的研究主题,河流扇是扇研究的重要组成部分.近年来,河流扇研究引起了诸多学者关注,多种相关术语体系并行.针对上述问题,我们建立了一个融合383个全球现代河流扇沉积数据集,统计分析各类数据达8 400条.基于数据集,我们首次掌握了现代河流扇的全球分布,划分了河流扇的主要类型,明确了扇体形成、发育的控制因素.其中,河流扇的全球分布具备明显的纬度分带和地域分区特征:从纬度上看,主要发育在北纬30°～50°(49.5％);从地域上看,主要分布在亚洲中部及中国西部(50.1％).根据地形限制和河道摆动情况,可以把河流扇分为非限制迭进型、非限制摆动型、限制迭进型和限制摆动型4种形态类型.河流扇主要在干旱、半干旱条件下形成,受局部构造条件、气候以及物源供给控制.扇体内部建造以河流作用为主,从扇体顶点到远端呈现出明显的异质性.扇体的面积直接受控于纬度、温度、坡度、集水区面积、区域年降雨量和山前距离等因素.随着研究的进一步深入,河流扇将在沉积学、地质灾害防治、石油勘探,甚至行星际沉积学等方面发挥重要作用.
Carbonate rocks account for about 20% of phanerozoic sedimentary rocks， and carbonate reservoirs hold at least 60% of the world’s proven hydrocarbon reserves. Calcite is a common carbonate mineral， and surface reactions on calcite are important for the formation of secondary pores in the shallow part of the crust. This paper made a necessary summary of the previous researches in related fields， and the results show that the determining factors affecting the growth of calcite include temperature， pressure， degree of supersaturation as well asthe type and radius of ions and so on. The decrease of the pressure， the increase of ionic strength and supersaturation are beneficial for the growth of calcite. The determining factors affecting the dissolution of calcite include temperature， pressure， pCO 2 ， ionic strength， pH value， medium fluid compositions， Zeta potential， and the degree of saturation of solution relative to calcite etc. The Zeta potential of calcite is the main influencing factor to determine whether the ions in the medium fluids can promote or inhibit the dissolution of calcite. Additionally， the concentrations and types of foreign ions have significantly different micromorphologies during the growth and dissolution of calcite， which makes it possible to recover the components of palaeo-fluid by observing the mineral surface morphology. In this paper， the inconsistent results of the previous experiments are analyzed and we proposed a new viewpoint for interpreting the inconsistency. The factors affecting the dissolution of calcite are investigated to find the optimal solution for carbonate dissolution， so as to provide a reliable theoretical basis for delineating high-quality reservoirs. Another significance of this paper is convenient for the following researchers to acquire the research hot-spot and the latest progresses in this field.
我国深层海相碳酸盐具有埋藏深度大、地质时代老、地质演化过程复杂等特点,其储层形成和分布受到众多因素制约,是海相油气勘探面临的关键问题之一.本文总结了团队在塔里木盆地长期研究成果,从深层碳酸盐岩建造和改造作用出发,揭示塔里木盆地多类型碳酸盐岩储层的主要发育条件和有序性分布.(1)塔里木盆地台地格局、地层发育演化及后期的埋藏与改造,均是盆缘构造事件在盆内的构造-沉积作用响应.早古生代盆地经历了地貌张裂分化与小型裂陷群形成、地貌挤压分化与层间不整合发育、古隆起与大型不整合作用、多期不整合叠加与古隆起埋藏改造等演化过程.(2)碳酸盐岩台地沉积过程中存在着不同尺度、不同类型的分异演化作用,可分为雏形期、建造期、鼎盛期和分解-消亡期,相应的台地边缘演化经历了缓斜坡、进积镶边、加积镶边和退积陡坡等建造过程,控制着台缘礁滩、台内滩、早期白云岩化潮坪等有利于规模化储集岩发育的沉积相带分布.(3)多级次不整合和断裂-流体改造是优质储层形成的重要机制.多期区域构造活动和周期性海平面变化控制了不同级别的不整合面发育,海平面下降主导的层序界面导致碳酸盐岩周期性暴露,形成了不同规模的层间岩溶和同沉积岩溶,区域构造运动所对应的大型不整合控制了大规模表生岩溶储层发育,不同规模的断裂和裂缝既是储层发育带,也是流体活动和改造的主要区带.(4)塔北-顺北地区不整合发育强度和断裂活动强度由北向南有序变化,控制了碳酸盐岩储层类型与分布的有序性变化,自北向南依次发育雅克拉断凸潜山型白云岩储层、塔河主体区古风化壳型岩溶储层、塔河斜坡区层控型岩溶储层、塔河覆盖区断溶型储层和顺北断控型储层,反映出由剥蚀区至覆盖区不整合控储作用逐渐变弱,而断裂控储作用增强.
羌塘盆地晚白垩世广泛发育阿布山组陆相粗碎屑岩沉积,然而当前对其沉积环境、物源特征、构造背景尚不清楚,这制约着我们对羌塘-拉萨地体碰撞后羌塘盆地的沉积演化和高原早期隆升的准确认识.为了解决上述问题,本文对安多114道班地区阿布山组的沉积时代、沉积环境、物源特征进行了详细研究.野外观察发现,阿布山组与下伏安山岩和上覆牛堡组均呈角度不整合接触,结合安山岩喷出时代和牛堡组的沉积年龄,限定其沉积时代为晚白垩世;阿布山组中砾石成分和形态结果显示,这些砾石主要为灰岩,且经历的搬运距离较短,为近源堆积.砂岩碎屑成分、重矿物特征、碎屑锆石年龄谱表明,阿布山组的物源主要来自南羌塘坳陷内部和中央隆升带的晚三叠世—侏罗纪地层.对比邻区阿布山组特征,本文认为羌塘盆地晚白垩世陆相红色碎屑岩的沉积与羌塘-拉萨地体持续汇聚作用形成的一系列逆冲推覆活动有关.
根据塔北地区海相碳酸盐岩油藏勘探开发实践,基于对断裂破碎作用、岩溶作用及溶蚀断裂带的控储、控藏特征研究,创新提出的"断溶体"圈闭(油藏)概念,业已成为碳酸盐岩油气勘探开发的新目标、新类型,丰富了海相碳酸盐岩油气藏理论认识.断溶体油藏宏观上具有"断溶控储、深断控藏、物性圈闭、垂向疏导、复式聚集、分段成藏"的成藏特征.断溶体油气藏沿断裂破碎带呈条带状分布,其宏观展布不受局部构造控制,油气藏无统一油水界面.基于以上地质条件的耦合,受断裂带结构特征差异的影响,同一断裂带的不同段,其富集、含水及连通特征具有明显差异,具有"一断裂段一单元,一单元一油藏"的特征,与传统的碳酸盐岩构造型、复合型等油藏类型具有显著差别.针对不同的断溶体油藏特征,需有针对性地制定开发治理对策.国内外油气田开发实践证实,包括碳酸盐岩、白云岩和砂岩在内的多种岩性地层中,均发育规模不等的断裂带、裂缝带,重视和深化研究深层海相碳酸盐岩断溶体圈闭(油藏)对油气勘探开发具有重要意义.
塔河油田奥陶系碳酸盐岩储层的开发方向已逐渐从地震反射的"串珠型"缝洞储层转移到"非串珠"型的小缝洞储层,确定小尺度缝洞的形成机制和发育规律是提高此类储量动用率的关键.本次研究对塔河油田主体区岩心和薄片开展了缝洞描述,确定了小尺度缝洞的裂缝成因类型、缝洞参数和充填性质等特征.研究区裂缝划分为构造裂缝和非构造成因裂缝,呈高角度或垂直产状,构造裂缝充填程度弱于风化裂缝;孔洞的充填程度相对较高,但仍存在残留的孔隙空间.基于古岩溶地貌、局部残丘构造幅度的刻画、断裂特征分析和构造应力场模拟,确定了小尺度缝洞体的形成机制.小尺度缝洞的形成受古岩溶地貌、局部残丘构造幅度、断裂和构造应力场综合控制.地貌相对高部位利于岩溶的发生,主体东区地貌整体较主体西区高,缝洞相对西区更发育.残丘构造幅度也影响了岩溶储层的分布差异,残丘缓翼构造坡度相对较缓,利于水岩充分反应,更利于岩溶储层形成.主大断裂控制大尺度溶洞的形成,同时大溶洞和断裂又控制着小尺度缝洞的发育,深部小尺度缝洞受断裂控制的发育深度可达T74界面以下200 m深度.构造应力场影响着构造裂缝的分布,构造裂缝的形成又为孔洞的形成提供了溶蚀通道,间接控制着小尺度缝洞体的分布.在小尺度缝洞形成机制分析的基础上,建立了基于非线性神经网络深度学习算法的小尺度缝洞预测方法,确定了小尺度缝洞平面分布,预测结果与单井缝洞发育情况和初期产液能力相匹配.研究成果为油田整体动用小尺度缝洞储量奠定了技术基础.
21世纪工业4.0促进了智能矿山诞生,智能矿山的地质、勘探、采矿、选矿、环境、测绘等多学科数据集构成了矿山大数据,从而促进了地球科学的数字化、信息化和智能化迅速发展.本文以河南上房沟钼(铁)矿床5G+智能矿山为例,开展矿山大数据的地学信息挖掘,以凸显新工科多学科融合研究,取得了创新性成果与一系列新的地学知识发现.具体内容概括如下:(1)根据"斑岩-夕卡岩"矿床理论和找矿矿物学方法,利用钻孔数据集和露采场大比例尺填图及镜下鉴定分析,查明并构建了矿区主矿种与潜在矿种的时空三维模型,新发现了 NE向构造赋矿地段、贯入式赋矿地段;(2)利用无人机遥感与地面高光谱短波红外和长波红外技术,识别并建立了矿区20多种主要蚀变矿物并构建了三维多参数矿物模型;(3)运用地球化学XRF测量和微区原位测量技术,建立了高光谱匹配的样品数据集,研发了矿区岩矿石有用和有害元素双矩阵制图软件,实现了传统地质统计学(高斯模拟、克里格插值)与机器学习(深度学习)关联的数学建模,复原并查明了 2021年3-4月选矿回收率偏低的配矿矿石矿物组合及其缘由;(4)运用矿区选矿工艺矿物学的生产与实验采选矿的多期次多类型数据集(季-月-日的岩粉、泥粉、精矿、尾矿等,＞1 800),研发岩粉与矿粉测试技术与分析方法,查明了上房沟钼矿的难选矿石类型及其有害矿物种类.研究结果表明,矿山多元多类型的数据集具有大数据"5V"(volume,variety,velocity,veracity,value)特征,准确管控矿山大数据的动态关联测量、分析与快速、高效评价有利于矿山智能决策和经济效益提高(回收率).其中,高精度的多参数三维建模不仅能够深层次挖掘岩矿石的地质、构造、蚀变、矿化等信息模型,核实储量/资源量,还能满足第四代工业5G+矿山的实时矿业(real-time-mining)发展的四维管控需求,例如三维空间可视化的地质矿产资源预测与评价及增储、虚拟仿真式的"年-季-月-日"动态配矿采矿、数字孪生的实时选矿等.本项研究成果为智能矿山的地质矿产深层次地学研究提供了借鉴和参考.
地球演化过程中存在着不同尺度旋回性,其时间跨度可以由一天至数亿年.地球轨道周期驱动的轨道尺度的沉积旋回,是目前定量化地层划分对比研究最好的时间尺度选择.富有机质页岩的连续性使得轨道尺度的天文旋回更容易保存,而米氏旋回的"时间内涵"是进行地层等时对比的有效手段,可以对富有机质页岩进行高精度的地层划分和对比.本文以中国东部渤海湾盆地东营凹陷页岩油勘探的主力烃源岩层段为例,利用地球轨道周期实现陆相富有机质页岩多尺度地层等时对比,相关研究具有重要的科学和页岩油勘探指导意义.以斜率作为地层划分与对比的基本单元,实现东营凹陷富有机质页岩六级层序等时对比,误差小于4万年.通过天文年代标尺的区域对比,明确东营凹陷页岩油高产层段时间为42.85～41.85 Ma,在E2-E4这3个长偏心率周期范围内,与斜率周期的高振幅段恰巧吻合,推测其与斜率周期驱动的纹层状页岩发育所导致的页岩油高产有关.地球轨道周期具有较高的时间分辨率且较容易识别,万年级的年代框架为陆相页岩高精度等时对比提供了可能,可为页岩油勘探中优质烃源岩层段预测提供精细年代框架.
Pore structure is the key factor affecting the distribution and flow of shale gas, and fractal analysis can be used to quantitatively characterize the complex pore structures in shale. To better understand the porous structure of the organic-rich shale from the Lower Cambrian Niutitang Formation in northern Guizhou Province, southeastern China, shale samples from Well ZK, Songtao area, are analyzed to determine the pore parameters using high-pressure mercury injection and low-temperature nitrogen adsorption methods, combined with scanning electron microscope observation, shale geochemistry and mineral composition analysis. The pore fractal dimensions are calculated by FHH model, and the influencing factors for pore structure are discussed. To summarize:（1） the quartz content in the organic-rich shale ranges between 39%-68.4%; clay content, 11.5%-28.2%; organic carbon content, 2.77%-5.81%（average 3.81%）; while kerogen is mainly of type I, with high thermal maturity.（2） The BET specific surface area ranges between 11.954-21.744 m~2/g（average 14.572 m~2/g）; total pore volume, 0.0186-0.0259 cm~3/g（average 0.0214 cm~3/g）; and average pore size, 4.773-7.025 nm（average 5.967 nm）. Micropores contribute largely to the total specific surface area; while mesopores and macropores account for a large proportion of pore volume.（3） The organic-rich shale has complex pore structure dominated by micropores, with high heterogeneity and multifractal characteristics. The pore fractal dimensions D 1 and D 2 obtained from low-temperature nitrogen adsorption data have narrow pore-size distributions（D 1 , 2.65-2.71; D 2 , 2.79-2.85）, while mercury injection data yield a wider pore-size distribution for macropores(D Hg between 2.21-2.81).（4） D 2 is positively correlated with TOC content and micropore volume, and mineral composition has no significant effect on pore fractal dimensions. The fractal dimension of the Niutitang shale in the study area is similar to that of the gas-producing Longmaxi shale, indicating shales of this area have good pore structure.
The Shengyi District is located in the western Shengtuo Oilfield, where rich oil/gas deposits are found in reverse drag anticline traps formed under the control of the Shengbei Fault. In this study, sedimentary microfacies of the 1st sand group, 2nd member, Shahejie Formation（or ‘Sha’er 1 sand formation’, for short）, Shengyi District, was investigated through core analysis, grain size analysis, and sedimentary facies analysis. The sand formation is mainly made up of shallow lake beach-bar deposits, with five sedimentary microfacies: main bar; bar edge; beach ridge; beach sheet; and inter-beach depression. Facies characteristics and depositional environments of the Sha’er 1 sand formation, revealed by facies analysis on typical cores from single and continuous wells, show that the deposition of the sand formation is mainly controlled by provenance, landform, wind field, and water depth. This research can provide strong technical support for sustainable and stable oil field development in the later stage.
Among the five branches of basic science（mathematics, physics, chemistry, biology and geology）, quantum physics, quantum mathematics（including quantum computing and quantum astronomy） and quantum chemistry are currently driving the rapid development of modern science, and remarkable progress has been made in quantum biology. By comparison, quantum geology is lagging behind. To address this issue, we suggest that both quantum science and Big Data should be emphasized in geological research. However, the scientific community has yet to reach a consensus on the following two questions:（1） Is the “quantum geology” idea reasonable?（2） Can quantum approach be adopted in geological research? Here we summarize the development, application and future of quantum technology, discuss the urgency and feasibility of conducting quantum geology research in China, and suggest several potential research areas. As Big Data technology has been widely used in geological research, we point out several challenging issues, which include Big Data and paradigm revolution, Big Data thinking, Big Data and artificial intelligence, Big Data and pragmatism, as well as the prospect of and challenges in geological Big Data research. Finally, we suggest that a forward-looking approach should be taken in promoting quantum geology research, and that geological Big Data research should be greatly strengthened and the data-sharing problem in geological research should be addressed.
Shale gas exploration and development is relatively low level in the Lower Cambrian Niutitang Formation in the Upper Yangtze where organic matter-hosted pores（OMPs） developed unevenly across the region. In this paper, shale samples from three wells with different depositional settings—namely well EYY1 in the western Hubei rift trough of eastern Upper Yangtze, well W001-4 in the shallow-water shelf of northern Yangtze, and well SNY1 in the Hannan ancient land of northern Yangtze—were collected and analyzed, and shales were classified according to mineral compositions. Organic matter types and OMP development characteristics were investigated by scanning electron microscope method, and the main controlling factors for OMP development were discussed. The results show that（1） mineral compositions and types of the Niutitang shales vary between different regions. Shale samples from EYY1 and W001-4 were classified into siliceous, calcareous and mixed shales; while in SNY1 only siliceous shale was found.（2） Organic matters in the Niutitang shales were divided into depositional and migrated organic matters. In the three wells, the carbon-to-oxygen elemental ratio（C∶O） in the depositional organic matter（averaging 6.74） was higher than that in the migrated organic matter（averaging 2.71）.（3） OMPs were well developed in EYY1 and SNY1, with relatively high surface porosities of 6%-28%; while in W001-4 they were poorly developed, with surface porosities ranging between 3%-10%. The OMP diameter in the three wells showed a unimodal distribution, centering between 2-50 nm in EYY1 and SNY1 and generally between 5-25 nm in W001-4. Organic matter in siliceous shales had the highest surface porosity, higher than in calcareous shales.（4） Migrated organic matter showed higher level pore development compared to depositional organic matter. OMP development was affected by counteraction between pore production from hydrocarbon generation from organic matter and pore reduction from compaction. Clay mineral is another important factor hindering micro-scale OMP development in shales, and siliceous shales rich in clay minerals may be an important clue for the determination of exploration “sweet spots” for Cambrian shale gas resource.
Current cyclostratigraphic researches mainly focus on the Phanerozoic strata whilst the Precambrian era—the longest span of geological time in Earth’s history—is less studied. The Officer Basin in western Australia possesses well-preserved mid-Neoproterozoic（～800 Ma） successions. The Lancer 1 well, located in the western Officer Basin, has recovered ～750 m of continuous sedimentary strata of the Neoproterozoic Browne-Hussar formations, mainly interbeded with sandstone and mudstone and occasional dolomite, showing great promise for the cyclostratigraphic analysis. In this study, the gamma-ray logging data of the Browne-Hussar formations are used as paleoclimate proxy. Spectral analysis shows that the wavelength ratios of the cycles in the 940-1 200 m mudstone interval of the Hussar Formation correspond to the hierarchies of the Milankovitch cycles, where sedimentary cycles, defined as 34 m, 8-10 m, 1.8-2.7 m or 1.2-1.4 m thick cycles, can be interpreted to correspond respectively to the Milankovitch long-eccentricity, short-eccentricity, obliquity and precession cycles. After calibrating to the 405-ka long-eccentricity cycle, a floating astronomical timescale is established for the mudstone interval of the Hussar Formation and the periods of the four Milankovitch cycles are estimated to be 405 ka, 127 ka, 27-34 ka and 15.2-18.1 ka, respectively. Based on the obliquity period of（27±0.7） ka, the Earth-Moon distance and the length of day（LOD） are calculated to be（362 935±1 495） km and（20.56±0.2） h, respectively. These results represent the first report on climate cycles in the Early-Neoproterozoic（～800 Ma） in response to orbital forcing, which provides a key reference to elucidating the evolution of the Earth-Moon system.
Framboidal pyrite occurs widely in shales of all ages and its morphological and grain size characteristics can be used as a redox indicator in environmental research. In this study, micromorphology of pyrites in shale samples from the Lower Cambrian Shuijingtuo and Yanjiahe Formations in well EYY1, Yangtze area, was investigated in detail using Ar ion milling and scanning electron micrography（SEM） techniques, where relevant parameters for framboidal pyrite indicative of paleoenvironments were obtained through statistical analysis using image processing software（ImageJ）. The results show that various types of pyrites including framboidal, lumpy, euhedral and anhedral pyrites are developed in EYY1. Among them, framboidal pyrite, as the most predominant type, shows obvious changes in gain size（D）, microcrystalline grain shape and size（d） and D/d ratio in the vertical direction. The grain shape of microcrystalline framboidal pyrite in the Yanjiahe Formation is mainly octahedral; while in the Shuijingtuo Formation it is mainly dodecahedral or near-spherical in the lower part, octahedral or cubic in the middle part, and cubic or tetrahedral in the upper part, with decreasing microcrystalline grain roundness upward from the bottom part, indicating weaker reducing conditions in the same trend. The average grain sizes of framboidal pyrite in organic-rich shales and microcrystalline framboidal pyrite in the EYY1 core section are 4.43 and 0.338 μm, respectively, and gradually decrease from bottom to top, which is opposite to the upward reduction-to-oxidation trend revealed by geochemical parameters. The D/d ratio for framboidal pyrite is relatively small in the Yanjiahe Formation and largest at the bottom of the Shuijingtuo Formation then gradually decreases upward. These results suggest that framboidal pyrite developed at the boundary between the Shuijingtuo and Yanjiahe Formations may reflect the redox condition prior to the stratigraphical change. The small grain size of framboidal pyrite in the upper Shuijingtuo Formation may suggest an aqueous redox interface and reduced abundance of iron-bearing minerals in the source supply.
The introduction of deep learning models can greatly improve the efficiency of geological image analysis and thus increase the level of quantitative research. As an example, the Ar-ion polishing scanning electron microscope（SEM） images of shale samples from the Lower Cambrian Niutitang Formation in western Hubei, Upper Yangtze were analyzed using three deep learning models, Mask-RCNN, FCN and U-Net, to identify the minerals, organic matter and pores（basic tasks） after image pretreatment（binarization, etc.） We compared the running time and identification accuracy between the three models, and discussed the model applicability and model differences in geological image recognition and processing. In addition, we compared the best performance model, U-Net model, with the general image processing softwares（JmicroVision, Adobe Photoshop, etc.） in pore recognition. The FCN model performed well in the basic tasks, but could not distinguish the mineral components and fractures with similar colors; whereas the Mask-RCNN model could identify the main minerals with strong segmentation but not low-resolution pores and fractures. In comparison, the U-Net model greatly improved the efficiency of shale geological image recognition with an 300-fold increase in image recognition speed over the general image processing softwares. Applying the U-Net model, the pore structural types of the Niutitang shale of the study area can be divided into circular intra-granular mineral pores, random irregular inter-granular mineral pores, angular organic matter-hosted pores and dense organic matter-hosted micropores. The pore structural parameters obtained based on SEM image analysis of large enough sample size may be used for reservoir classification and evaluation. The example provided in this study may help improving the efficiency of geological image research as well as promoting artificial intelligence application in oil and gas research.
Hard rock caverns used for underground oil storage are mostly constructed in crystalline rock mass and water sealed, thus the seepage characteristics of the rock mass directly affect the safety of oil storage. The challenge of water sealing in underground caverns can be simplified to the problem of two-phase oil/water seepage through fractures in the surrounding rock mass, so it is necessary to consider both oil and water displacements to addressing this challenge. In this study, seepage visualization experiment and numerical simulation were carried out on fractured natural granite. The fracture roughness was quantified by fractal analysis, and transparent fractures with different roughness values were prepared by 3D printing. Physical model test was carried out by using the home-built single-fracture multiphase-seepage visualization device, and fluid displacement in rough fractures was studied by numerical simulation. Difference in the mobility ratio（water to oil） leads to different two-phase（oil-water） flow dynamics, where the oil-flooding front moves as piston flow while water-flooding front as fingered flow. In oil displacement, the increase of fluid pressure or fracture width can significantly improve oil displacement efficiency, whereas increasing fracture roughness hinders oil flow and impedes oil displacement. Considering the relevance of fracture roughness and oil viscosity, safety requirements for water sealing are not adequate if only fluid pressure is considered. Due to Jamin effect, it is difficult to remove oil completely from rock fractures by water flooding, or to recover leaked oil. The research results provide a reference for understanding two-phase seepage in rough fractures in rocks and water sealing mechanism in underground caverns.
Porphyry copper deposits often experience late-stage uplifting, denudation, destruction and transformation after deposition. The Bainaimiao Cu-Au deposit in Inner Mongolia is one of the few Early-Paleozoic porphyry copper deposits in the orogenic belt of northern China, therefore, it is of both theoretical and practical（in mineral prospecting） significance to study the mineral alteration/preservation processes. According to thermochronological analysis of zircon fission tracks and apatite fission tracks, the total denudation of the Cu-Au mining area since 150 Ma is between 3.11-3.25 km, and the late-stage uplifting and cooling of intrusive rocks in the study area can be characterized by a three-stage process. During the rapid exhumation stage between 150-90 Ma, the tectonic stress is mainly compressional, and the average denudation rate is 0.0457 mm/a; in the stable denudation stage from 90 to 20 Ma, extensional stress dominates, and the average denudation rate is 0.0033 mm/a; and during the rapid exhumation stage from 20 Ma to present, compressional stress again dominates, and the average denudation rate is 0.0388 mm/a. Thus, from the Caledonian to Yanshanian, affected by subduction and collision of the ancient Asian Ocean and intracontinental orogenic activities, the mining area is under tectonic stresses alternating between compressional and extensional, which provides favorable geological conditions for mineral alteration/preservation in the Bainaimiao Cu-Au deposit.
