A deep kernel method for lithofacies identification using conventional well logs

How to fit a properly nonlinear classification model from conventional well logs to lithofacies is a key problem for machine learning methods.Kernel methods(e.g.,KFD,SVM,MSVM)are effective attempts to solve this issue due to abilities of handling nonlinear features by kernel functions.Deep mining of log features indicating lithofacies still needs to be improved for kernel methods.Hence,this work employs deep neural networks to enhance the kernel principal component analysis(KPCA)method and proposes a deep kernel method(DKM)for lithofacies identification using well logs.DKM includes a feature extractor and a classifier.The feature extractor consists of a series of KPCA models arranged according to residual network structure.A gradient-free optimization method is introduced to automatically optimize parameters and structure in DKM,which can avoid complex tuning of parameters in models.To test the validation of the proposed DKM for lithofacies identification,an open-sourced dataset with seven con-ventional logs(GR,CAL,AC,DEN,CNL,LLD,and LLS)and lithofacies labels from the Daniudi Gas Field in China is used.There are eight lithofacies,namely clastic rocks(pebbly,coarse,medium,and fine sand-stone,siltstone,mudstone),coal,and carbonate rocks.The comparisons between DKM and three commonly used kernel methods(KFD,SVM,MSVM)show that(1)DKM(85.7%)outperforms SVM(77%),KFD(79.5%),and MSVM(82.8%)in accuracy of lithofacies identification;(2)DKM is about twice faster than the multi-kernel method(MSVM)with good accuracy.The blind well test in Well D13 indicates that compared with the other three methods DKM improves about 24%in accuracy,35%in precision,41%in recall,and 40%in F1 score,respectively.In general,DKM is an effective method for complex lithofacies identification.This work also discussed the optimal structure and classifier for DKM.Experimental re-sults show that(m_(1),m_(2),O)is the optimal model structure and linear svM is the optimal classifier.(m_(1),m_(2),O)means there are m KPCAs,and then m2 residual units.A workflow to determine an optimal classifier in DKM for lithofacies identification is proposed,too.

Acoustic reflection well logging modeling using the frequency-domain finite-element method with a hybrid PML

In this paper, we propose a hybrid PML （H-PML） combining the normal absorption factor of convolutional PML （C-PML） with tangential absorption factor of Mutiaxial PML （M-PML）. The H-PML boundary conditions can better suppress the numerical instability in some extreme models, and the computational speed of finite-element method and the dynamic range are greatly increased using this HPML. We use the finite-element method with a hybrid PML to model the acoustic reflection of the interface when wireline and well logging while drilling （LWD）, in a formation with a reflector outside the borehole. The simulation results suggests that the PS- and SP- reflected waves arrive at the same time when the inclination between the well and the outer interface is zero, and the difference in arrival times increases with increasing dip angle. When there are fractures outside the well, the reflection signal is clearer in the subsequent reflection waves and may be used to identify the fractured zone. The difference between the dominant wavelength and the model scale shows that LWD reflection logging data are of higher resolution and quality than wireline acoustic reflection logging.

Research and Application of Log Defect Detection and Visualization System Based on Dry Coupling Ultrasonic Method

In order to optimize the wood internal quality detection and evaluation system and improve the comprehensive utilization rate of wood,this paper invented a set of log internal defect detection and visualization system by using the ultrasonic dry coupling agent method.The detection and visualization analysis of internal log defects were realized through log specimen test.The main conclusions show that the accuracy,reliability and practicability of the system for detecting the internal defects of log specimens have been effectively verified.The system can make the edge of the detected image smooth by interpolation algorithm,and the edge detection algorithm can be used to detect and reflect the location of internal defects of logs accurately.The content mentioned above has good application value for meeting the requirement of increasing demand for wood resources and improving the automation level of wood nondestructive testing instruments.

Comparison of different inversion methods of D-T_(2)two-dimensional nuclear magnetic resonance logging and applicability analysis

D-T_(2)two-dimensional nuclear magnetic resonance(2D NMR)logging technology can distinguish pore fluid types intuitively,and it is widely used in oil and gas exploration.Many 2D NMR inversion methods(e.g.,truncated singular value decomposition(TSVD),Butler-Reds-Dawson(BRD),LM-norm smoothing,and TIST-L1 regularization methods)have been proposed successively,but most are limited to numerical simulations.This study focused on the applicability of different inversion methods for NMR logging data of various acquisition sequences,from which the optimal inversion method was selected based on the comparative analysis.First,the two-dimensional NMR logging principle was studied.Then,these inversion methods were studied in detail,and the precision and computational efficiency of CPMG and diffusion editing(DE)sequences obtained from oil-water and gas-water models were compared,respectively.The inversion results and calculation time of truncated singular value decomposition(TSVD),Butler-Reds-Dawson(BRD),LM-norm smoothing,and TIST-L1 regularization were compared and analyzed through numerical simulations.The inversion method was optimized to process SP mode logging data from the MR Scanner instrument.The results showed that the TIST-regularization and LM-norm smoothing methods were more accurate for the CPMG and DE sequence echo trains of the oil-water and gas-water models.However,the LM-norm smoothing method was less time-consuming,making it more suitable for logging data processing.A case study in well A25 showed that the processing results by the LM-norm smoothing method were consistent with GEOLOG software.This demonstrates that the LM-norm smoothing method is applicable in practical NMR logging processing.

Numerical simulation and response analysis of microspherical focused logging in inclined micro-fractured formation

Tight and unconventional reservoirs have become the focus with the progress of petroleum exploration and development.Micro-fractures in these reservoirs can effectively improve reservoir permeability,and well-developed micro-fractures can serve to directly improve productivity.Compared with the centered electrical well logging method,the Micro Spherical Focused Logging(MSFL)is more suitable for microfracture identification due to its high resolution and near borehole wall measuring method.In this study,an anisotropic model is used to depict micro-fractured formation.First,a forward model with microfractured formation,borehole,logging instrument and surrounding rock is established.Subsequently,MSFL responses under different micro-fracture porosity,resistivity,dip angle and borehole radius,are calculated based on the finite element method(FEM).Finally,the MSFL responses under different microfracture parameters are analyzed with the response laws clarified.

A SPECTRAL STREAMLINE DIFFUSION FINITE ELEMENT COUPLING METHOD OF UNSTEADY TRANSPORT EQUATION IN THE FIELD OF NEUTRON LOGGING

In this paper, a new numerical method, the coupling method of spherical harmonic function spectral and streamline diffusion finite element for unsteady Boltzmann equation in the neutron logging field, is discussed. The convergence and error estimations of this scheme are proved. Its applications in the field of neutron logging show its effectiveness.

Self-adaptive inversion method of electromagnetic-propagation resistivity logging while drilling data

Because of the constraint mode of the inversion objective function in the traditional resistivity-inversion method of electromagnetic-propagation resistivity logging while drilling(EPR-LWD),obvious differences appear in the radial and vertical investigation characteristics between the amplitude-ratio and phase difference,which affect the practical application of EPR-LWD data.In this paper,according to the EPR-LWD data,a self-adaptive constraint resistivity-inversion method,which adopts a self-adaptive constraint weighted expression in the objective function to balance the contributions of the phase difference and amplitude attenuation,is proposed.A particle swarm optimization algorithm is also introduced to eliminate the dependence of the accuracy and convergence on the initial value of the inversion.According to the inversion results of multiple classical formation models for EPR-LWD,the differences between the adaptive constraint inversion-resistivity logs with the traditional amplitude-ratio and the phase difference of the resistivity logs are discussed in detail.The results demonstrate that the adaptive resistivity logs take into account the advantages of the amplitude-ratio logs in the radial investigation and phase difference logs in the vertical resolution.Further,it is superior in thin-layer identification and invasion-effect appraisal compared with the single-amplitude-ratio and phase difference logs.The inversion results can provide a theoretical reference for research on the resistivity-inversion method of electromagnetic wave LWD.

FMLogs:基于模板匹配的日志解析方法

日志主要记录软硬件的运行信息,通过查看系统日志,可以找到系统出现的问题及原因,确保系统的稳定性和正常运行。日志解析的目的是将半结构化的原始日志解析为可阅读的日志模板,现有解析方法往往只注重于对原始日志的解析,而忽略了后期模板处理,导致结果的精度不能进一步提高。自此,提出了一种日志解析方法FMLogs(logs parsing based on frequency and MinHash algorithm)。该方法通过设计正则表达式和调节阈值参数以获得最佳性能,同时采用了字符级频率统计和MinHash方法对长度相同和不同的日志模板进行合并。FMLogs在七个真实数据集上进行了广泛的实验,取得了0.924的平均解析准确率和0.983的F 1-Score。实验结果表明,FMLogs是一种有效的日志解析方法,在解析日志的同时具有较高的准确性和效率,并能保证性能的稳定。

CCSD Well Logging Engineering Program

This paper introduces briefly the tasks and characteristics of China Continent Science Drilling (CCSD) Well Logging Engineering, the logging methods measured with CCSD, the quality control of original logging information, the logging plan of CCSD, the logging engineering management of CCSD, the logging interpretation and the results and reports made with CCSD.

Corrections for downhole NMR logging

Nuclear magnetic resonance logging （NMR） is an open well logging method. Drilling mud resistivity, formation resistivity and sodium ions influence its radio frequency （RF） field strength and NMR logging signals. Research on these effects can provide an important basis for NMR logging data acquisition and interpretation. Three models, water-based drilling mud--water bearing formation, water- based drilling mud--oil bearing formation, oil-based drilling mud--water bearing formation, were studied by finite element method numerical simulation. The influences of drilling mud resistivity and formation resistivity on the NMR logging tool RF field and the influences of sodium ions on the NMR logging signals were simulated numerically. On the basis of analysis, RF field correction and sodium ion correction formulae were proposed and their application range was also discussed. The results indicate that when drilling mud resistivity and formation resistivity are 0.02 Ω·m and 0.2 Ω·m respectively, the attenuation index of centric NMR logging tool is 8.9% and 9.47% respectively. The RF field of an eccentric NMR logging tool is affected mainly by formation resistivity. When formation resistivity is 0.1 Ω·m, the attenuation index is 17.5%. For centric NMR logging tools, the signals coming from sodium ions can be up to 31.8% of total signal. Suggestions are proposed for further research into NMR logging tool correction method and response characteristics.

基于改进ΔlogR方法估算低阻海相页岩储层总有机碳含量——以桂中坳陷A地区B组为例

总有机碳含量(total organic carbon,TOC)是确定烃源岩质量的最重要参数。为解决传统ΔlogR方法应用于沉积环境较复杂的海相页岩气储层精度低的问题,以桂中坳陷A区B组海相页岩气储层为研究对象,在分析研究区低阻原因的基础上,对测井曲线按照扩径范围进行井径校正,并利用自然伽马能谱测井中Th/K数据,与ΔlogR建立多元拟合关系,提出基于井径校正和自然伽马能谱测井的改进ΔlogR方法。结果表明:声波时差曲线经过校正后,降低了井径扩径带来的测井数据失真的影响,同时结合自然伽马能谱测井,能够适应研究区复杂的地质背景和较强的非均质性,大大提高总有机碳质量分数的预测精度,具有较好的推广性和适用性。该方法可以为桂中坳陷A区B组海相页岩气储层评价提供参考。

Study on Calculation Method of Compressional Velocities Based on Field Well Logs

In the past, most of the studies for compressional velocities are based on experimental measurements, which lack the support of field data. The purpose of this study is to estimate the compressional velocities based on well log data of delta front subfacies of Lower Tertiary ages of Ji-Dong oil field, China. At initial stage, we have chosen the well log parameters (effect factors) which strongly influence on compressional velocities and established a new modified equation for compressional velocities, which is based on these effect factors. Then Gardner, De-hua Han and this newly established equation were utilized to calculate the compressional velocities in each well. Finally, Least-square regression was carried out to check the fitting of each equation. Regression results clearly indicate that our purposed equation shows better fitting as compared to Gardner and De-hua Han equations.

A nonparametric spectrum estimation method for dispersion and attenuation analysis of borehole acoustic measurements

Dispersion and attenuation analysis can be used to determine formation anisotropy induced by fractures,or stresses.In this paper,we propose a nonparametric spectrum estimation method to get phase dispersion characteristics and attenuation coefficient.By designing an appropriate vector filter,phase velocity,attenuation coefficient and amplitude can be inverted from the waveform recorded by the receiver array.Performance analysis of this algorithm is compared with Extended Prony Method(EPM)and Forward and Backward Matrix Pencil(FBMP)method.Based on the analysis results,the proposed method is capable of achieving high resolution and precision as the parametric spectrum estimation methods.At the meantime,it also keeps high stability as the other nonparametric spectrum estimation methods.At last,applications to synthetic waveforms modeled using finite difference method and real data show its efficiency.The real data processing results show that the P-wave attenuation log is more sensitive to oil formation compared to S-wave;and the S-wave attenuation log is more sensitive to shale formation compared to P-wave.

基于阵列声波测井的井下多尺度压裂效果评价方法

为有效监测水力压裂效果、提高储层压裂效果评价精度,基于阵列声波测井资料的井下压裂效果评价方法,采用反射斯通利波提取和成像技术定量表征近井筒压裂缝,建立适用于测井观测系统的叠前深度偏移成像算法,以实现远井压裂缝的高精度成像,形成了基于阵列声波测井的井下多尺度压裂效果评价方法,并在中国西部某油田X1水平井成功应用。研究表明,基于阵列声波测井资料评价的压裂缝发育程度与压裂液注入量呈正相关关系,并且压裂缝最为发育层段与地震蚂蚁体属性图中天然裂缝发育位置一致。基于阵列声波测井的井下压裂效果评价方法的提出,对表征井周数十米范围内不同尺度压裂缝的发育情况具有重要意义,对更好地发挥水力压裂技术对能源勘探开发行业的支撑作用具有一定指导意义。

融合裂缝形态信息的电阻率测井成像图空白条带填充方法

电阻率测井成像图中存在宽度和位置不定的空白条带,而目前常用的空白条带反距离加权插值填充法仅依靠像素点距离参数计算插值,往往导致填充后区域与实测区域纹理不连贯,纹理易呈现直线状,与实际状况不符。为此,将裂缝区域形态与分布信息作为参量,构建插值计算模型,形成融合裂缝形态分布信息的空白条带填充法。该填充算法能根据插值点与裂缝形态拟合线的相对位置关系,动态调整参考点对插值点插值的权重系数,填充后的裂缝区域纹理连贯平滑,更符合裂缝实际形态和分布的规律。实际测井数据应用表明,相较于传统填充方法,采用该方法填充后的裂缝分布拟合线与霍夫变换后曲线相似度提升25%以上,不连贯填充点数量下降88%以上,性能提升明显。该方法的应用成果可为后续测井图像中各种裂缝的智能识别和精确分割提供更好的数据支撑。

双向驱动井径测井仪推靠系统工作行为分析

针对单向驱动井径测井仪存在的启动响应慢、速度集中和初始运动冲击等问题,提出一种具有双向驱动特征的新型结构井径测井仪,并对其推靠系统(实现测量臂伸缩)双向驱动过程进行矢量化建模求解及工作行为分析(运动性能仿真、测量性能仿真)。首先进行双向驱动测井仪推靠系统物理建模和理论研究,以双驱动丝杠系统参数(速度、位移)为输入,采用复数矢量法对测井仪推靠系统运动性能参数(位移、速度、加速度)进行参数化表征;然后以实际双驱动推靠系统为对象进行实例分析,对其运动特性进行理论求解,得到推靠杆端口和推靠销钉的运动曲线,并对双驱动速度匹配特性进行研究;最后基于ADAMS进行测井仪工作行为分析,将仿真与理论结果进行对比,结果表明推靠系统仿真运动曲线与理论解析曲线具有较高的吻合度,证明了测井仪推靠系统矢量化建模求解的正确性。

基于主成分自组织神经网络法的测井曲线分层技术

在砂岩型铀矿找矿工作中,提高测井岩性分层效率和精度至关重要。为提高砂岩型铀矿岩性分层效果,本文采用主成分分析法对多个测井曲线进行降维处理,将主成分分析法的第一主成分、第二主成分、第三主成分作为自组织神经网络的样本数据,进行自组织神经网络训练,将训练好的网络模型用于砂岩型铀矿岩性的自动化分层。实验结果显示:主成分自组织神经网络法岩性分层精度可达到85%以上,高于传统自组织神经网络算法78%的分层精度,具有更好的测井岩性分层效果。因此,主成分自组织神经网算法的岩性分层方法有效减少了输入样本的种类,简化了自组织神经网络结构,其自动化分层效果要优于传统的自组织神经网络算法。本文的研究结果表明,主成分自组织神经网算法在砂岩型铀矿领域岩性识别工作中具有较好的应用效果。

南海神狐海域天然气水合物储层参数定量评价方法

海洋型天然气水合物以固态形式赋存于尚未固结的泥质粉砂和粉砂质泥地层中,不同于常规油气藏在固结砂岩地层中的赋存模式,因此造成水合物储层测井评价的难题。2007年以来我国南海神狐地区实施了4次水合物钻探,取得丰富的地球物理与岩心资料并且证实该区存在着巨大的天然气水合物资源潜力。本文通过对神狐地区常规测井、特殊测井资料以及取心资料的研究分析,总结水合物地层各项测井响应特征,利用岩心数据刻度测井,建立储层岩性、物性以及水合物饱和度计算模型,形成一套适合神狐地区天然气水合物储层参数的定量评价技术。利用本文方法对研究区部分井进行综合评价,水合物地层孔隙度平均为49.91%,水合物饱和度平均为29.75%,孔隙结构以小孔隙为主,渗透性较差,评价结果与岩心实验结果基本一致,符合率达85%。该研究成果为神狐海域水合物资源量评价及成藏研究提供了新的参考。

川南长宁页岩气储层有机质成熟度测井计算方法适用性分析

川南地区五峰组-龙马溪组页岩地层受多期构造运动引起沉降与抬升等作用的影响,使其在不同区域上的热演化程度存在较大的差异。有机质成熟度是评价页岩气储层热演化程度的重要参数。利用地球化学、激光拉曼、沥青反射率、物性、含气量测试及测井曲线等资料,研究川南长宁地区页岩井有机质不同成熟度的热演化特征,厘定出不同成熟度范围下的敏感测井响应,利用变形ΔlogR法、多元回归法、CatBoost算法计算不同成熟度类型的等效镜质体反射率(RO),并对比了计算效果与适用性。结果表明:①基于有机质出现石墨化前后热演化特征及RO对敏感电阻率曲线的影响,以有机质石墨化的RO下限值3.5%划分出过成熟阶段Ⅰ和过成熟阶段Ⅱ这2种类型。②过成熟阶段Ⅰ的特征为孔隙度高、含水饱和度低、含气量高、电阻率高,电阻率(Rt)曲线和声波时差(AC)曲线重叠差异效果弱,Rt-AC测井曲线颜色成像图显示为“紫+红”模式;过成熟阶段Ⅱ的特征为孔隙度低、含水饱和度高、含气量低、电阻率低,电阻率曲线和声波时差曲线重叠且具有明显幅度差,Rt-AC测井曲线颜色成像图显示为“蓝+暗红”模式。③CatBoost法优选自然伽马、声波时差、中子、电阻率这4条曲线作为输入变量,过成熟阶段Ⅰ、Ⅱ型的RO计算值与实测值相关系数都超过了0.95,取得了较好的有机质成熟度预测效果。该方法一定程度上弥补了实验数据不连续的缺点,为判识页岩气储层热演化阶段及降低勘探开发风险提供了依据。

富油煤焦油产率测井评价方法研究

【目的】富油煤作为一种集煤、油、气属性为一体的煤基油气资源,对于保障我国油气资源供应、实现煤炭清洁高效利用具有重要意义。目前富油煤的识别主要依靠格金干馏试验测量的焦油产率来判断,而利用地球物理手段准确识别与评价富油煤的研究则相对薄弱。【方法】以新疆三塘湖盆地侏罗系西山窑组和八道湾组煤层为研究对象,基于岩石物理实验,在明确富油煤典型测井响应特征基础上,创新建立了一种煤焦油产率测井定量评价方法。【结果和结论】结果表明:(1)相比含油煤,富油煤具有富氢结构物质含量高、孔隙结构差的特点,这造成富油煤具有“低密度、高中子和高电阻率”的常规测井响应特征;(2)通过多状态二维核磁共振实验测量分析,明确了富油煤具有“二维核磁T1谱双峰,T_(2)<1、T_(1)/T_(2)>10区域信号强”的核磁测井响应特征,而含油煤相应区域信号不明显,这与富油煤中富氢结构物质含量高有关;(3)基于富油煤典型测井响应特征,提出利用电阻率和中子2个参数构建富油煤指示因子Z,Z越大,表明煤焦油产率越高。在煤焦油产率刻度基础上,建立了煤焦油产率与指示因子间的线性关系,实现了煤焦油产率的准确方便计算,可操作性强。上述认识为基于地球物理测井手段识别和评价富油煤提供了理论指导。