共查询到20条相似文献，搜索用时 171 毫秒
1.
岩石与矿物的地球化学成分数据具有高维度特征。传统的岩矿地球化学成分研究主要采用二元/三元图解判别法,准确率不高,在数理统计方法上有欠缺。机器学习方法非常适用于对大样本高维度的岩矿成分数据进行数理统计处理。笔者等在介绍机器学习常见算法基本原理的基础上,总结近5年来国内外学者将机器学习方法应用于岩石矿物成分数据研究的实例,包括：(1)根据矿物成分溯源其母岩(源岩)、判别矿床类型,(2)新生代火山岩溯源,(3)判别变质岩原岩,(4)依据岩浆岩成分判别大地构造环境等。已有的研究实例显示,机器学习方法的准确度明显优于传统的低维度判别法。机器学习本质是分析大样本数据的高维度变量之间的相关、归类等多元统计问题。推广机器学习的应用需要建设开放获取(Open Access)的矿物、岩石成分数据库,同时全面实施开放研究(Open Research)的发表策略。 相似文献
2.
花岗岩成矿的地球化学判别标志 总被引：5，自引：4，他引：5
花岗岩具成矿专属性。花岗质岩浆中成矿元素行为受控于源岩成分、氧化状态、结晶分异作用和挥发组份。花岗岩地球化学特征和矿物化学成分一定程度上继承了源岩的特征，反映了岩浆作用的大地构造背景、氧化状态和结晶分异，因此，可使用花岗岩成分判别成矿花岗岩。本文归纳总结了花岗岩成矿的地球化学判别标志。 相似文献
3.
高光谱成像作为目前遥感领域最先进的技术,在地质应用中取得了巨大成功。岩石和矿物由于电子过程和分子振动可以产生特征的光谱吸收,因此可以利用高光谱技术进行岩矿填图,快速且准确地获取区域内岩石和矿物的分布情况,进而圈定有潜力的找矿靶区。岩矿高光谱通过对岩矿信息的提取可获得矿物类别和矿物丰度分布甚至矿物化学成分的填图结果,识别方法包括光谱匹配、模式识别和人工智能方法 3大类。GF-5卫星上搭载的高光谱成像仪覆盖度宽、光谱分辨率和信噪比高,满足矿物精细识别和大比例尺、大面积岩性填图的要求,应用前景广阔。青藏高原地区矿产资源丰富、岩体裸露但地势偏远,有利于高光谱遥感蚀变矿物填图,开展区域找矿工作。同时,高光谱遥感凭借其空间尺度优势,可以方便快捷地获取区域整体信息,结合地球化学的时间信息,可为青藏高原关键地质科学问题的解决提供新的视角。提升岩矿高光谱遥感的定量化和智能化水平,将传统地质学方法获得的地下深部资料与遥感手段获取的地表数据结合,是促进地质遥感和地球系统科学发展的重要途径。 相似文献
4.
依靠岩浆构造环境的地球化学成分认识岩浆形成过程是岩石地球化学中的重要应用。当前利用岩石地球化学成分判别构造环境的工作还不够深入。用4种基于决策树的机器学习方法对来自全球新生代洋岛玄武岩(OIB)、岛弧玄武岩(IAB)及大洋中脊玄武岩(MORB)等镁铁-超镁铁岩中单斜辉石的13种主量元素构成数据集进行了岩浆构造环境判别和主要特征排序。通过对比4种基于决策树的机器学习方法,验证了树类算法对于地球化学成分识别问题的有效性,并总结出4种方法在处理岩浆构造环境判别问题时的优劣:决策树算法判别过程更易于理解,但是其准确率欠佳;boosting算法中的AdaBoost和GBDT对于岩浆构造环境的鉴别准确度较高,但构造过程复杂;bagging集成算法随机森林在权衡性能和模型可理解性时不失为一个良好的选择。此外,还通过4种算法的特征重要性排序得出Cr_2O_3,TFeO,TiO_2,FeO和Al_2O_3是进行岩浆构造环境判别的重要成分。 相似文献
5.
本文应用了洪古勒楞蛇绿岩中铬尖晶石的电子探针分析数据,讨论该蛇绿岩中不同岩石中铬尖晶石化学成分特征,显示出铬尖晶石的成分与寄主岩石有关。基本可将其分为三组:造矿铬尖晶石、堆积岩副矿物铬尖晶石和地幔橄榄岩副矿物铬尖晶石。它们在主要成分和微量成分方面均存在区别,化学成分的相关性也不一样。共生橄榄石-铬尖晶石对的元素分配说明二者基本为平衡矿物。 相似文献
6.
赣中南产铀钨系列花岗岩岩石化学数值多元判别及地质效果 总被引：1，自引：0，他引：1
由于赣中南及邻区地壳初始物质应分的不均匀性而导致的地壳重熔型花岗岩岩石化学成分和成矿元素的差异,以及U、W-Sn、TR、Nb-Ta等与壳型花岗岩有关的成矿元素地球化学性质的差异,决定了这些元素的不共生性,而各自有其成矿专属岩体。按成矿专属性可将本区燕山期花岗岩划分为产U系列花岗岩和产TR、W-Sn、Nb-Ta系列花岗岩。二系列岩体的岩石化学成分和副矿物组合有明显的差别。据此建立的岩石化学数值多元线性判别函数式能判别二系列岩体;对产W-Sn、TR、Nb-Ta系列岩体能进行二级判别,同时能判别壳源型花岗质侵出岩的产矿性,以及区分加里东期交代花岗岩及混熔过渡型Cu花岗若。 相似文献
7.
锆石是在自然界中多种温压条件下能够稳定保存,并记录原岩年龄信息的副矿物。锆石微量元素能完整记录地质演化过程信息。通过微量元素分析锆石成因的研究已久,通常利用Th-U图解和LaN-(Sm/La)N图解等二元图解对锆石进行分类研究。然而,随着锆石研究的深入,以及二元图解无法呈现数据高维度信息的局限性,传统图解已经不能满足对锆石类型进行准确判别,且对已知类型的锆石出现判定偏差。因此,本文将地质大数据与机器学习相结合,训练出高维度锆石成因分类器。文中收集了3 498条不同成因类型的锆石微量元素数据,并通过测试和运用随机森林、支持向量机、人工神经网络和k近邻等4种机器学习算法,最终得出准确率为86.8%的线性支持向量机锆石成因分类器,用于锆石类型的判定与预测。这项工作为锆石分类研究提供了更高维度的判别手段,极大提高了微量元素分析成因结果的精度。将锆石微量元素数据与机器学习方法相结合,是大数据分析与机器学习技术在地球化学研究中的积极探索。 相似文献
8.
金伯利岩微量元素地球化学特征与含金刚石程度的判别 总被引：1，自引：0，他引：1
鉴于金伯利岩普遍经受强烈的热液蚀变作用,新鲜的金伯利岩比较少见,这就给研究金伯利岩的矿物成分和岩石化学成分带来一定困难。因此,研究金伯利岩中比较稳定的微量元素分布特征,查明含金刚石程度不同的金伯利岩中微量元素含量差别,探讨根据微量元素判别和预测金伯利岩含金刚石程度的可能性,具有比较普遍的实际意义。 相似文献
9.
10.
火成岩岩石成因研究包括确定岩石性质、源岩的矿物成分和化学成分、源岩熔融过程和岩石形成的物理化学条件,以及同化混染作用、分异作用及流体对岩石的怍用等内容。对这些内容的研究以前多由实验岩石学来承担。近几年来,痕量元素地球化学,包括浠土元素地球化学的进展,使火成岩成因研究工作大大推进了一步。稀土元素定量模式是稀土元素地球 相似文献
11.
通过岩浆岩的地球化学特征判别岩浆形成的大地构造环境和岩浆源区的化学性质是地球化学全岩分析最重要的应用之一。该方法利用全岩地球化学数据,包括主量元素、微量元素和同位素组成数据,对给定岩浆岩(玄武岩、花岗岩等)的大地构造环境进行判别。作为人工智能技术在地球化学研究领域中的新尝试,机器学习判别方法逐渐成为经典判别图解法的补充研究手段。然而,高维数据特征筛选和繁多未知参数确定是影响算法分类准确性的两个主要因素。为此,提出一种遗传算法优化神经网络耦合判别方法(GA-NNDM)。该方法利用特征选择、参数确定和分类性能之间的反馈联系,将分类准确率、所选特征数量和特征代价作为适应度函数,通过迭代演化寻求最佳特征子集和未知参数,从而达到减少特征、优化参数和提高性能的目的。此外,根据公开玄武岩样品地球化学数据,通过K折交叉验证等方法设置纵向、横向比较实验来验证GA-NNDM在玄武岩构造环境判别方面的准确性、稳定性和外延性。仿真实验结果表明,GA-NNDM具有优良的判别效果和泛化能力,其总体分类准确率能达到90%。因此,GA-NNDM值得在地球化学领域做进一步推广应用。 相似文献
12.
Steffen H. Büttner 《Mineralogy and Petrology》2012,104(1-2):129-135
Rock Maker is a simple software tool that computes bulk rock compositions resulting from mixing or unmixing of rocks or minerals. The calculations describe the chemical expression of processes such as magma mixing, fractional crystallization, assimilation, residual melt extraction, or formation of solid solutions. Rock Maker can also be used for the elimination of thermodynamically inactive or unwanted chemical components from the whole rock composition, such as cores of porphyroblasts that are considered not to be in equilibrium with the matrix. The calculation of the resulting rock composition is essentially based on modal proportions and compositions of different components in rocks, which may include specific portions of the rock or individual mineral phases. Compositional data, obtained using XRF, ICPMS, EDS, or EPMA, may include major and trace element concentrations. Depending upon the nature of the problem to be solved, the concentrations of oxidic and elemental components can be added to, or subtracted from, each other, producing the calculated normalized whole rock composition after completion of the investigated process (mixing, unmixing, depletion, enrichment, etc.). Furthermore, the software allows the calculation of whole rock compositions from ideal mineral compositions, for which modal proportions can be chosen from pre-defined mineral compositions. The data set includes the most common rock forming minerals and allows the addition of further phases. This section can be used to calculate the approximate whole rock compositions from petrographic modal analysis. This part of Rock Maker is specifically suitable as a teaching tool that illustrates the interrelationship between mineral compositions, modes, and the corresponding whole rock compositions. 相似文献
13.
Quantification of electron microprobe compositional maps of rock thin sections: an optimized method and examples 总被引：1，自引：0，他引：1
V. DE ANDRADE O. VIDAL E. LEWIN P. O'BRIEN P. AGARD 《Journal of Metamorphic Geology》2006,24(7):655-668
Quantification of discrete pressure–temperature domains in deformed chlorite + white mica‐bearing metapelites was undertaken on mineral compositions derived by two‐dimensional microprobe compositional mapping of selected areas of rock thin sections. In order to achieve compositional information at sufficient analytical precision, spatial resolution and sample coverage within a typical analysis time of 1 day, an optimization of measurement methods was necessary. The method presented here allows collection of raw counts for eight different element concentrations at an analytical precision of ～1–2 wt%. X‐ray intensity multiplane maps (one map per measured chemical element) are translated into concentration multiplane maps, utilizing selected conventionally measured spot analyses combined with the Castaing approximation for each mineral. As this step requires identification of the different minerals present in the mapped area, a statistical clustering technique to identify different groups of composition was developed, guided by simple petrographic inspection of the thin section, to delineate the important minerals in the mapped area. Finally, the compositions of each pixel are translated into a mineral structural formula thus yielding a new kind of image with a high content of petrological information. The reliability of the mineral composition images was emphasized by carrying out precision tests on the analytical data. The possible use of chemical maps to infer the P–T–deformation history of metamorphic rocks is illustrated with two samples from the Spitzbergen and the Sambagawa blueschist facies belts. In both samples, a strong correlation between structures and chemistry is observed. Qualitative estimates of P–T conditions from the Si‐content of mica and chlorite are in good agreement with their location in microstructures that formed at different times. Therefore, the combination of chemical maps with microstructural observations is a very powerful approach to understand both the evolution of complex metamorphic rocks and the control by deformation of mineral reactivity. 相似文献
14.
On thermobarometry 总被引：15，自引：2，他引：13
Thermobarometry, the estimation of the pressure–temperature ( P – T ) conditions of metamorphism, can be undertaken by using pseudosection calculations as well as by conventional methods. Conventional thermobarometry uses only the equilibrium thermodynamics of balanced reactions between end-members of minerals, combined with the observed compositions of minerals. In contrast, pseudosections involve a forward calculation of mineral equilibria for a given rock composition. When related to observed rock data such as mineral assemblages, mineral proportions and mineral compositions, pseudosections have the power to provide valuable additional thermobarometric information. This is because the rock composition provides added constraints on P – T , unavailable in conventional thermobarometry, such as when minerals in the mineral assemblage are no longer stable, or when additional minerals join the mineral assemblage. Considering both conventional and pseudosection thermobarometry, a minimum requirement is that they use the same thermodynamic data and activity–composition models for the minerals involved. A new thermocalc facility is introduced that allows pseudosection datafile coding to be used for conventional thermobarometry. Guidelines are given and pitfalls discussed relating to pseudosection modelling and conventional thermobarometry. We argue that, commonly, pseudosection modelling provides the most powerful thermobarometric tools. 相似文献
15.
Compositions, proportions, and equilibrium temperature of coexisting two-feldspar in crystalline rocks are of great importance to classification in petrography and interpretation of petrogenesis. Crystalline rocks are usually composed of 4-6 minerals (phases), depending on their independent chemical components and the equilibrium temperature of crystallizations. In general, number of mineral phases can be determined by the “Phase Rule”. According to the mass balance principle, bulk composition of coexisting two-feldspar could be evaluated from the bulk chemistry of a rock, provided that the compositions of the coexisting mafic mineral phases containing calcium, sodium, and potassium oxides are determined, e.g., by microprobe analysis. The compositions, proportions, and temperature of two-feldspar in equilibrium can thus be simultaneously resolved numerically from bulk composition of the rock, by incorporating the activity/composition relations of the ternary feldspars with the mass balance constraints. Upon the numerical approximation method presented in this paper, better-quality, internally consistent data on feldspar group could usually be obtained, which would be expected more realistic and accurate in consideration of thermodynamic equilibria in the system of crystalline rocks, as well as bulk chemistry of a rock and the composing minerals. 相似文献
16.
数值计算中利用随机分布对各单元赋予不同的物理力学参数来考虑岩石的非均质性,赋值过程是纯随机的,并没有考虑岩石矿物组成的结构特征。故提出一种新的岩石非均质参数赋值方法--岩石矿物细胞元随机性参数赋值方法,基于岩石矿物种类及其含量定义细胞元类别判定区间,利用Monte Carlo方法对各个细胞元进行矿物类别判定,并进行相应的参数赋值,通过各矿物细胞元的随机混合体来描述岩石的非均质性。该方法既考虑了组成岩石的矿物种类及其含量（结构特征）,又考虑了组成矿物在岩石中的随机分布特征。针对矿物分布的随机特征,利用两矿物细胞元混合模型和三矿物细胞元混合模型进行数值试验,研究了矿物细胞元随机分布特征对岩石宏观力学参数的影响。研究结果表明,岩石矿物细胞元随机性参数赋值方法具有结构性和随机性的双重特性,其随机性不依赖于随机参数,岩石宏观力学参数受细胞元随机性特征的影响很小。 相似文献
17.
《International Geology Review》2012,54(10):1253-1277
ABSTRACTSeafloor subduction and subduction-zone metamorphism (SZM) are understood to be the very cause of both subduction-zone magmatism and mantle compositional heterogeneity. In this article, we compile geochemical data for blueschist and eclogite facies rocks from global palaeo-subduction-zones in the literature, including those from the Chinese Western Tianshan ultrahigh pressure (UHP) metamorphic belt. We synthesize our up-to-date understanding on how chemical elements behave and their controls during subduction-zone metamorphism. Although the compositional heterogeneity of metamorphic minerals from subducted rocks has been recently reported, we emphasize that the mineral compositional heterogeneity is controlled by elemental availability during mineral growth, which is affected by the protolith composition, the inherited composition of precursor minerals, and the competition with neighbouring growing minerals. In addition, given the likely effects of varying protolith compositions and metamorphic conditions on elemental behaviours, we classify meta-mafic rocks from global palaeo-subduction-zones with varying metamorphic conditions into groups in terms of their protolith compositions (i.e. ocean island basalt (OIB)-like, enriched mid-ocean ridge basalt (MORB)-like, normal [N]-MORB-like), and discuss geochemical behaviours of chemical elements within these co-genetic groups rather than simply accepting the conclusions in the literature. We also discuss the geochemical consequences of SZM with implications for chemical geodynamics, and propose with emphasis that: (1) the traditionally accepted ‘fluid flux induced-melting’ model for arc magmatism requires revision; and (2) the residual subducted ocean crust cannot be the major source material for OIB, although it can contribute to the deep mantle compositional heterogeneity. We also highlight some important questions and problems that need further investigations, e.g. complex subduction-zone geochemical processes, different contributions of seafloor subduction and resultant subduction of continental materials, and the representativeness of studied HP–UHP metamorphic rocks. 相似文献
18.
19.
一种计算岩石中矿物组成的新方法 总被引：1，自引：0，他引：1
介绍根据显微下观察,应用岩石化学全分析结果和分子量计算法计算岩石中矿物组成的新方法。所得各种矿物的组成是整块岩石在三维立体体积中各种矿物的质量分数,比目前沿用的目测一个岩石切面上各种矿物的面积百分比法更准确,比CIPW标准矿物分子法更符合岩石样品组合的实际情况,而且可用于含云母族、绿泥石族和铀矿物等蚀变矿化的花岗岩及沉积岩、变质岩的矿物组成计算。 相似文献
20.
The analysis of modal and normative composition of sedimentary rocks is widely used for studying their sources and tectonic settings. The normative calculation of the mineral composition of rocks in this study is formulated as a linear programming problem and is solved by means of the simplex method. This enables both simultaneous and successive subtraction of a set of basic minerals from a rock sample represented by its chemical composition {SiO2...LOI}. Such an approach provides a more exact calculation of the contribution of basic minerals in the rock. This mathematical approach is used to study two representative sets of sandstones and fine-grained rocks from a Meso- to Neoproterozoic marginal basin of southeastern Siberia (Uchur–Maya region, Yakutia) and a Pennsylvanian-Lower Permian uplifted continental block in Colorado, USA. The calculated normative mineral compositions of the Siberian sandstones are consistent with the observed modal compositions. These sandstones vary from K- Feldspar rich arkoses at the base of the sequence (the Uchur Group, lower Riphean) to quartz arenites or lithic sandstones and wacke in transgressive successions of the middle-upper Riphean. Arkoses and quartz arenites are dominant in Meso- to Neoproterozoic Siberia. These samples represent craton interior uplifted basement and quartzose, recycled orogen provenance of a stable craton in Rodinia. There are higher but consistent discrepancies between the calculated and observed compositions for the Pennsylvanian to Lower-Permian arkoses and quartz arenites (Sangre de Cristo, Belden, and Maroon Formations). The differences between the predicted and observed mineralogy may be due to uncertainties in the modes in the matrix and cement of the sandstones. This normative program should supplement modal calculations and provide better genetic constructions, especially in case of matrix-rich sandstones. 相似文献