A generalized image and boundaries of the Bazhenov Formation: constraints from well log data for Salym-type sections (West Siberia)

A technique akin to Galton’s composite portraiture is suggested for creating a generalized image of a stratigraphic unit based on well log data from a group of well sections. The procedure begins with superimposing stratigraphically equivalent section points according to pairwise correlation models followed by weighted summation of similar logs. The superimposition highlights persistent common features of the sections, while their minor individual details become suppressed. A generalized model of the Bazhenov Formation in the Salym area has been constructed from gamma-ray, apparent resistivity, and self-potential logs. Radioactivity and resistivity anomalies, which are often used as constraints on the Bazhenov Formation boundaries, show a significant spatial discrepancy. According to the correlation, the formation volume in each well depends on the depth interval corresponding to the hypostratotype. The results were used to map the Bazhenov Formation thickness and its standard deviation, as well as average values of different logs.     

动态模式匹配算法在地层对比中的应用

动态序列匹配算法已经被广泛地用于模式识别过程中。采用特殊的编码方法来模仿人眼的工作原理，对待识别信号与标准模式信号进行计算机自动匹配。从而达到识别未知信号的目的。针对利用测井曲线的地层对比，本文在动态匹配算法的基础上，为提高对比的可靠性，研究了以相似度S为主体，加上曲线的均值Y、曲线的方差σ、地层厚度△H以及作为改进参数—岩性特征值为辅助参数的匹配距离计算方法。按照最小匹配费用原则从匹配代价矩阵中确定了地层匹配路径。实际资料处理表明，该方法不仅能够较好找出匹配地层，还可有效判断地层尖灭现象。     

Correlation of geologic logs with spectral methods

Spectral studies of geologic logs demonstrate that automatic well—log correlation can be processed more efficiently in the frequency domain. Cross correlation of the power spectra of well logs identifies the direction and degree of thickening of stratigraphic sequences between two wells. Given the stretch, the displacement between logs is computed by correlation processes without relying on iterative procedures. Beginning with digitized log data of unequal lengths, power spectra are computed. The stretch factor between the two logs is observed as a difference in frequency scaling. A transform to logarithmic frequencies converts the spectra to a form that reduces the scaling effect of the frequencies to a simple displacement between the plots. A Lagrange interpolation procedure permits cross correlation of the two spectra with a variable window size. The peak value of the resultant correlation function identifies the displacement between the spectra and this, in turn, permits calculation of the stretch factor.     

一种基于核学习的储集层渗透率预测新方法

基于核学习的支持向量机,是一种采用结构风险最小化原则代替传统经验风险最小化原则的新型统计学习方法,具有完备的理论基础。这里提出了核学习技术在储集层非均质特性描述中渗透率参数预测的新用途。在复杂地层中,基于支持向量机的智能和自适应模式识别能力而建立了常规测井多参数信息输入的渗透率预测模型,然后对实际油田储集层渗透率进行了预测。与常规线性回归模型预测结果相对比,所提出的方法更易于使用,很少受不确定因素的影响,并具有较强的信息整合能力以及更高的预测准确性和可信度。     

Heterogeneities analysis using the generalized fractal dimensions and the continuous wavelet transform. Application to the KTB boreholes

The main goal of this paper is to identify heterogeneities from well logs data using the wavelet-based multifractal analysis. Firstly, the wavelet transform modulus maxima lines method is applied with a moving window of 128 samples to the raw well logs data. After that, the generalized fractal dimensions that correspond to the three first moments of the function of partition are estimated. Application to synthetic and real well logs data of the main and pilot Kontinentales Tiefbohrprogramm de Bundesreplik Deutschland wells shows that the information and the correlation dimensions can be used for heterogeneities analysis and lithofacies segmentation form well logs data.     

Transforms to enhance correlation of mechanical well logs

Mathematical correlation of geologic sections from mechanical well logs depends on comparison of frequency components derived from the recorded amplitude variations of the well logs as they respond to changes in lithology. Techniques have been developed that compensate for changes in thickness of units within lithologic sequences and for moderate numbers of missing and truncated beds. However, accurate correlation depends on time stratigraphic units being composed of relatively consistent sequences of rock lithologic units. Correlation is uncertain in areas where time stratigraphic sequences exhibit considerable facies variation. Visual correlation experiences similar problems but the human eye can sort out facies changes more easily and may be better than mechanical correlation in problem areas. Both mechanical and visual correlation can be enhanced by the application of relatively simple transforms to the basic log data. They are applied by computer to digital logs and are designed to magnify specific diagnostics of the more continuous beds while subduing distracting events and noise. Both mechanical and visual correlation are improved by the use of transformed logs; moreover, the digital logs permit optically pleasing displays that maintain the overall advantage of trained geologists over mechanical intervention.     

Heterogeneities characterization from velocity logs using multifractional Brownian motion

Well logs provide additional information from the borehole that cannot be derived from other subsurface investigations. Their analysis may bring supplementary features about the earth’s heterogeneities. In previous researches, the logs were modeled using fractional Brownian motions indexed by Hurst (or Hölder) exponents quantifying their global regularity degrees. Indeed, these monofractal models, characterized by the same Hurst exponent, do not allow to study the depth-evolution of the local Hurst parameter. In order to overcome this problem, we propose to model the logs using multifractional Brownian motions and suggest an algorithm, developed previously for financial time series, to estimate the local Hurst exponent function. First, the potential of this algorithm is assessed through its application on synthetic sonic log data simulated by the successive random additions method. We note that the estimated Hurst functions (or regularity profiles) are very close to the theoretical Hurst functions. Second, this analysis is extended to sonic logs data from the KTB pilot borehole (Germany). Using the resulted regularity profiles, we carry out a lithological segmentation and fault identification on the geological layers crossed by the well. We derived also a correlation between the Hurst value variation and the lithological changes.     

Stratigraphic correlation of presumed Milankovitch cycles in the Blue Lias (Hettangian to earliest Sinemurian), England

Simple digital filtering methods allow the correlation of patterns of metre-scale cycles between outcrop sections and well logs through the Hettangian to earliest Sinemurian part of the Lower Lias in southern England. The matching of cyclic patterns is sufficiently confident to allow the identification of a significant break in the Blue Lias section at Lyme Regis (Dorset). The standard ammonite zones of the Hettangian Stage are shown to be of quantifiably unequal duration.     

Expert systems for automated correlation and interpretation of wireline logs

CORRELATOR is an interactive computer program for lithostratigraphic correlation of wireline logs able to store correlations in a data base with a consistency, accuracy, speed, and resolution that are difficult to obtain manually. The automatic determination of correlations is based on the maximization of a weighted correlation coefficient using two wireline logs per well. CORRELATOR has an expert system to scan and flag incongruous correlations in the data base. The user has the option to accept or disregard the advice offered by the system. The expert system represents knowledge through production rules. The inference system is goal-driven and uses backward chaining to scan through the rules. Work in progress is used to illustrate the potential that a second expert system with a similar architecture for interpreting dip diagrams could have to identify episodes—as those of interest in sequence stratigraphy and fault detection- and annotate them in the stratigraphic column. Several examples illustrate the presentation.     

测井数据滑动窗关联维计算和沉积学分析

应用滑动窗关联维计算方法对鲁北济阳坳陷某钻井剖面石炭系—二叠系陆表海沉积层段GR测井响应数据进行了分形维数计算,结果表明海侵体系域关联维数一般较小,且维数曲线振荡变化较弱,而高水位体系域则相反。这与海侵体系域形成时海平面上升的主控因素压制或掩盖了其他因素的显现有关,同时也证明测井序列分形分析能从另一个新的视角观察识别测井序列中所包含的地质信息。随着分形方法应用研究的不断开展,必将从测井数据中挖掘出更多的有用信息,指导油气等资源勘探与开发,丰富及完善测井地质学理论。相信将来分形会成为地质数据分析的标准工具。     

孤北-桩海地区东营组层序地层对比及层序地层格架

笔者通过大量的钻井岩芯观察、地震剖面解释、测井资料分析,详细研究了东营组层序划分和特征,共划分出了两个长期旋回,4个中期旋回。在层序划分的基础上,进行了全区的地层对比,各层序在不同地区发育程度不同是由于当时的古地理背景不同以及后期构造抬升剥蚀程度不同所造成的。分别建立了东西和南北向的层序地层格架模型,不同时期各中期旋回在不同地区有不同的发育特征。     

Brittleness index predictions from Lower Barnett Shale well-log data applying an optimized data matching algorithm at various sampling densities

The capability of accurately predicting mineralogical brittleness index(BI)from basic suites of well logs is desir-able as it provides a useful indicator of the fracability of tight formations.Measuring mineralogical components in rocks is expensive and time consuming.However,the basic well log curves are not well correlated with BI so correlation-based,machine-learning methods are not able to derive highly accurate BI predictions using such data.A correlation-free,optimized data-matching algorithm is configured to predict BI on a supervised basis from well log and core data available from two published wells in the Lower Barnett Shale Formation(Texas).This transparent open box(TOB)algorithm matches data records by calculating the sum of squared errors be-tween their variables and selecting the best matches as those with the minimum squared errors.It then applies optimizers to adjust weights applied to individual variable errors to minimize the root mean square error(RMSE)between calculated and predicted(BI).The prediction accuracy achieved by TOB using just five well logs(Gr,pb,Ns,Rs,Dt)to predict BI is dependent on the density of data records sampled.At a sampling density of about one sample per 0.5 ft BI is predicted with RMSE～0.056 and R2～0.790.At a sampling density of about one sample per 0.1 ft BI is predicted with RMSE～0.008 and R2～0.995.Adding a stratigraphic height index as an additional(sixth)input variable method improves BI prediction accuracy to RMSE～0.003 and R2～0.999 for the two wells with only 1 record in 10,000 yielding a BI prediction error of＞±0.1.The model has the potential to be applied in an unsupervised basis to predict BI from basic well log data in surrounding wells lacking mineralogical measure-ments but with similar lithofacies and burial histories.The method could also be extended to predict elastic rock properties in and seismic attributes from wells and seismic data to improve the precision of brittleness index and fracability mapping spatially.     

陆东凹陷九佛堂组火山碎屑岩地层对比与划分

针对陆东凹陷稳定标志层不发育、地层对比难度大的问题,在岩电特征分析的基础上,将火山作用与沉积作用相联系,确定火山喷发期次。应用井震联合层位标定技术,精细制作合成记录,建立地震标志层与井上典型岩电特征的关系,完成地层对比与划分,有效指导了储层预测和井位部署。     

Cenozoic stratigraphy of the northern Sakhalin shelf

The analysis of diatom, palynofloral, and benthic foraminiferal assemblages made it possible to substantiate the age of Cenozoic sections recovered by wells on the northern and northeastern Sakhalin shelf. Biostratigraphic materials, lithological properties of stratigraphic units, and standard logs served as the basis for developing the first stratigraphic correlation scale of Cenozoic sequences on the Sakhalin shelf.     

Exploratory Factor Analysis of Wireline Logs Using a Float-Encoded Genetic Algorithm

In the paper, a novel inversion approach is used for the solution of the problem of factor analysis. The float-encoded genetic algorithm as a global optimization method is implemented to extract factor variables using open-hole logging data. The suggested statistical workflow is used to give a reliable estimate for not only the factors but also the related petrophysical properties in hydrocarbon formations. In the first step, the factor loadings and scores are estimated by Jöreskog’s fast approximate method, which are gradually improved by the genetic algorithm. The forward problem is solved to calculate wireline logs directly from the factor scores. In each generation, the observed and calculated well logs are compared to update the factor population. During the genetic algorithm run, the average fitness of factor populations is maximized to give the best fit between the observed and theoretical data. By using the empirical relation between the first factor and formation shaliness, the shale volume is estimated along the borehole. Permeability as a derived quantity also correlates with the first factor, which allows its determination from an independent source. The estimation results agree well with those of independent deterministic modeling and core measurements. Case studies from Hungary and the USA demonstrate the feasibility of the global optimization based factor analysis, which provides a useful tool for improved reservoir characterization.     

River channel sand body prediction by means of integrating well log and seismic attribute analysis in dense well patterns: a case study of the eastern Beierxi Block in the Sabei Development Area of Daqing Oilfield, China

The developmental areas in eastern China have been in the super-high water cut phase for a significant amount of time, and as a result the distribution of remaining oil has gradually become more and more complex. The effective prediction of the distribution of interwell river channel sand bodies in dense well patterns has become an urgent issue in enhanced oil recovery. In this paper, based on the comprehensive analysis of high-resolution well logs and seismic data, a case study of predicting interwell channel sand bodies is carried out, with the assistance of high-resolution well logs and 3D seismic data from the Beierxi Block of the Sabei Development Area. After interpretation of the horizons and faults, the seismic attributes are extracted on the premise of seismic horizon tracing in a dense well pattern, and seismic attributes on wells are determined by means of the inverse distance-weighted mean method. The linear statistical correlation, multivariate statistical correlation, clustering procedure correlation, and 2D crossplot analyses of the RAVE module between the seismic attributes and well logs are used to deduce the relationships between seismic data and physical information of the reservoir (including effective thickness and distribution of sedimentary facie). Finally, the possibility of using multiattribute results to foster an understanding of depositional-oriented trends in interwell river channel sand body distribution is demonstrated. The planar depositional microfacies distribution is mapped with the results of the well-seismic data, which will assist in achieving a new understanding of how remaining oil may be discovered and will also assist in making further decisions in terms of hydrocarbon exploration and development.     

综合应用测井及地震资料建立复杂断块地质模型

针对滩海油田具有完钻井少,以地震资料为主的特点,采用井震结合的方法进行地质建模。复杂断块油藏具有许多内部断层,充分利用地震解释层位和断层成果建立合理的构造模型。采用“多步建模”的思路,选用序贯高斯同位协同模拟算法建立储层参数模型。以测井资料为硬数据,地震反演数据为软数据,先后建立自然伽玛模型、泥质含量模型和砂泥岩相模型,然后以砂泥岩相模型为约束,建立储层参数模型。经研究表明,综合使用地震解释成果和三维地震数据体,有助于建立准确、合理的构造模型;而采用“多步建模”的思路,利用序贯高斯同位协同模拟算法,并综合测井资料、地震资料建立储层参数模型,有助于从地质的角度对储层参数模拟进行约束,提高储层随机建模精度,保证储层参数模型的准确和可靠性。     

Well-log correlation using a back-propagation neural network

We present a back-propagation neural network with an input layer in the form of a tapped delay line wich can be trained effectively on one or several well logs to recognize a particular geological marker. Subsequently, the neural network proposes locations of this marker on other wells in the field. Another neural network, similar in architecture to the first one, performs the same task for secondary markers using, in addition to the well logs, a depth reference function to the first marker. This method is shown to have better performance and better discrimination than standard cross-correlation techniques. It lends itself well for an interactive implementation on a workstation.     

Significance of the Toolebuc gamma ray anomaly in the search for and evaluation of oil shale in the Eromanga Basin

Chemical and radiometric analyses have been made on core samples from 15 drill holes intersecting the Toolebuc Formation at Julia Creek. The holes are located in the vicinity of the St Elmo Structure and include intersections of a potentially economic oil shale deposit west of this structure as well as non‐economic mixed coquinite/shale intersections to the east. The gamma ray activity of the formation is due largely to uranium in the shales. Wire‐line gamma ray logs, available for eight of these holes, reflect the distribution of organic carbon. Despite this, there is poor correlation between uranium and organic carbon in the oil shales as a result of remobilization of uranium from organic matter into phosphate during diagenesis. In consequence wire‐line gamma ray logs have no potential for evaluating oil yield. Density, on the other hand, is a good indicator of oil potential. On a hole by hole basis, the correlation between uranium and organic carbon in the formation ranges from 0.22 west of the St Elmo Structure (the potential oil shale resource) to 0.95 east of the structure. This reflects a systematic change in sedimentology from oil shale that is thick and well separated from coquinite (low correlation) to laterally equivalent but finely interbedded coquinite and shale (high correlations). Such trends could be used in exploratory drilling to indicate favourable areas elsewhere in the Toolebuc Formation.