Inversion of time-dependent nuclear well-logging data using neural networks

The purpose of this work was to investigate a new and fast inversion methodology for the prediction of subsurface formation properties such as porosity, salinity and oil saturation, using time‐dependent nuclear well logging data. Although the ultimate aim is to apply the technique to real‐field data, an initial investigation as described in this paper, was first required; this has been carried out using simulation results from the time‐dependent radiation transport problem within a borehole. Simulated neutron and γ‐ray fluxes at two sodium iodide (NaI) detectors, one near and one far from a pulsed neutron source emitting at ～14 MeV, were used for the investigation. A total of 67 energy groups from the BUGLE96 cross section library together with 567 property combinations were employed for the original flux response generation, achieved by solving numerically the time‐dependent Boltzmann radiation transport equation in its even parity form. Material property combinations (scenarios) and their correspondent teaching outputs (flux response at detectors) are used to train the Artificial Neural Networks (ANNs) and test data is used to assess the accuracy of the ANNs. The trained networks are then used to produce a surrogate model of the expensive, in terms of computational time and resources, forward model with which a simple inversion method is applied to calculate material properties from the time evolution of flux responses at the two detectors. The inversion technique uses a fast surrogate model comprising 8026 artificial neural networks, which consist of an input layer with three input units (neurons) for porosity, salinity and oil saturation; and two hidden layers and one output neuron representing the scalar photon or neutron flux prediction at the detector. This is the first time this technique has been applied to invert pulsed neutron logging tool information and the results produced are very promising. The next step in the procedure is to apply the methodology to real data.     

Energy Group optimization for forward and inverse problems in nuclear engineering: application to downwell-logging problems

Simulating radiation transport of neutral particles (neutrons and γ‐ray photons) within subsurface formations has been an area of research in the nuclear well‐logging community since the 1960s, with many researchers exploiting existing computational tools already available within the nuclear reactor community. Deterministic codes became a popular tool, with the radiation transport equation being solved using a discretization of phase‐space of the problem (energy, angle, space and time). The energy discretization in such codes is based on the multigroup approximation, or equivalently the discrete finite‐difference energy approximation. One of the uncertainties, therefore, of simulating radiation transport problems, has become the multigroup energy structure. The nuclear reactor community has tackled the problem by optimizing existing nuclear cross‐sectional libraries using a variety of group‐collapsing codes, whilst the nuclear well‐logging community has relied, until now, on libraries used in the nuclear reactor community. However, although the utilization of such libraries has been extremely useful in the past, it has also become clear that a larger number of energy groups were available than was necessary for the well‐logging problems. It was obvious, therefore, that a multigroup energy structure specific to the needs of the nuclear well‐logging community needed to be established. This would have the benefit of reducing computational time (the ultimate aim of this work) for both the stochastic and deterministic calculations since computational time increases with the number of energy groups. We, therefore, present in this study two methodologies that enable the optimization of any multigroup neutron–γ energy structure. Although we test our theoretical approaches on nuclear well‐logging synthetic data, the methodologies can be applied to other radiation transport problems that use the multigroup energy approximation. The first approach considers the effect of collapsing the neutron groups by solving the forward transport problem directly using the deterministic code EVENT, and obtaining neutron and γ‐ray fluxes deterministically for the different group‐collapsing options. The best collapsing option is chosen as the one which minimizes the effect on the γ‐ray spectrum. During this methodology, parallel processing is implemented to reduce computational times. The second approach uses the uncollapsed output from neural network simulations in order to estimate the new, collapsed fluxes for the different collapsing cases. Subsequently, an inversion technique is used which calculates the properties of the subsurface, based on the collapsed fluxes. The best collapsing option is chosen as the one that predicts the subsurface properties with a minimal error. The fundamental difference between the two methodologies relates to their effect on the generated γ‐rays. The first methodology takes the generation of γ‐rays fully into account by solving the transport equation directly. The second methodology assumes that the reduction of the neutron groups has no effect on the γ‐ray fluxes. It does, however, utilize an inversion scheme to predict the subsurface properties reliably, and it looks at the effect of collapsing the neutron groups on these predictions. Although the second procedure is favoured because of (a) the speed with which a solution can be obtained and (b) the application of an inversion scheme, its results need to be validated against a physically more stringent methodology. A comparison of the two methodologies is therefore given.     

PNN测井方法的蒙特卡罗模拟结果研究

PNN（脉冲中子-中子）测井是利用He-3管记录热中子时间谱,通过获取地层宏观吸收截面来确定含水饱和度的方法.本文利用蒙特卡罗方法（MCNP-4C）模拟了不同地层水矿化度、孔隙度、饱和度和井眼等条件下的热中子时间谱,研究了地层宏观吸收截面与地层水矿化度的关系,从理论上确定PNN测井适合的地层水矿化度范围约为10～100 g/L,在地层水矿化度为50 g/L时适于测井的孔隙度下限约为10%.井眼流体不同,地层的热中子计数率不同,但对地层宏观吸收截面影响较小.利用远近探测器热中子计数比值可以确定孔隙度,并提出了根据不同岩性和饱和度的地层宏观吸收截面与孔隙度的交会图来评价骨架、含水饱和度以及确定油层、水层和气层的方法.PNN测井方法在低地层水矿化度、低孔隙度地层比其他方法具有优势.     

脉冲中子双伽马谱饱和度测井方法及数值模拟研究

脉冲中子双伽马谱饱和度测井方法是利用特定的脉冲和测量时序设计,采用远、近伽马探测器记录非弹性散射和俘获伽马能谱以及伽马射线时间谱,通过获取C/O和地层宏观俘获截面等信息来确定地层含油饱和度.采用大直径仪器和BGO晶体探测器时,其采集的非弹性散射和俘获伽马能谱计数统计性优于国外仪器,重复脉冲发射中子后伽马射线时间谱仍满足指数衰减规律,且能使伽马射线时间谱的道计数累积,计数统计性好于常规中子寿命测井,得到的地层宏观俘获截面与单脉冲发射近似相同.在地层孔隙度未知的情况下,利用C/O和地层宏观俘获截面交会技术可以定量确定含油饱和度.     

A modelling of effects of variable saturation in invaded zone near borehole on neutron tool response in the gas-bearing Miocene formations of the Carpathian foredeep,Poland

Invasion of mud filtrate from the borehole into the surrounding formation can significantly alter the response of logging tools.In this paper the neutron tool response was calculated for a gas-bearing rock crossed by the borehole.The invaded zone is characterized by variable fluid and gas saturation and is defined by a specific hyperbolic function. The saturation function is based on the characteristics of Polish Carpathian Foredeep Miocene formations reached in gas-bearing rocks. The whole system including neutron tool equipped with Am-Be neutron source and a set of neutron detectors, borehole, invaded zone, and virgin zone, was numerically modeled and used as the input to the Monte-Carlo neutron transport code. The response of each detector was calculated for different combinations of porosity and saturation of invaded zone and virgin zone.     

核磁共振双TW测井数据联合反演与流体识别

针对核磁共振测井双TW观测数据分析和流体识别的需要,研究了基于全局搜索的遗传算法和局部搜索的最小二乘法的联合反演算法,实现了核磁共振双TW观测数据处理.首先,研究了饱和油气水岩石物理模型的核磁共振双TW观测模式的测井响应机理;然后,利用全局搜索性能优良的遗传算法, 对核磁共振回波差数据进行了反演,计算出了流体的核磁弛豫性质及其体积;最后,以遗传算法的反演结果为初值, 利用阻尼最小二乘方法对双TW回波串进行更精细的反演,计算出了双TW的T₂分布、孔隙度和流体饱和度.理想模型的合成数据和实际测井资料应用表明,遗传算法与最小二乘方法相结合是一种行之有效的联合反演方法,能较好地实现核磁共振测井双TW观测数据的处理和流体评价.     

Artificial neural network forward modelling and inversion of electrokinetic logging data

An artificial neural network method is proposed as a computationally economic alternative to numerical simulation by the Biot theory for predicting borehole seismoelectric measurements given a set of formation properties. Borehole seismoelectric measurements are simulated using a finite element forward model, which solves the Biot equations together with an equation for the streaming potential. The results show that the neural network method successfully predicts the streaming potentials at each detector, even when the input pressures are contaminated with 10% Gaussian noise. A fast inversion methodology is subsequently developed in order to predict subsurface material properties such as porosity and permeability from streaming potential measurements. The predicted permeability and porosity results indicate that the method predictions are more accurate for the permeability predictions, with the inverted permeabilities being in excellent agreement with the actual permeabilities. This approach was finally verified by using data from a field experiment. The predicted permeability results seem to predict the basic trends in permeabilities from a packer test. As expected from synthetic results, the predicted porosity is less accurate. Investigations are also carried out to predict the zeta potential. The predicted zeta potentials are in agreement with values obtained through experimental self potential measurements.     

基于改进多分类孪生支持向量机的测井岩性识别方法研究与应用

岩性识别是认识地层及求解储层参数的基础,受地质环境复杂性和非均质性影响,测井曲线间存在着大量的信息冗余,数据集类间分布不平衡,常用的分类算法无法满足实际需求.针对常用分类算法容错性差,识别岩性单一和无法有效解决类间不平衡的问题,本文改进合成少数过采样技术(Synthetic Minority Over Sampling Technique,SMOTE)来处理数据集,可得到类间平衡的新数据集,并提出一种新的模糊隶属度函数改进模糊孪生支持向量机,在北美Hugoton油气田实际测井数据的基础上,用改进多分类孪生支持向量(Improve Multi Class Twin Support Vector Machine,IMCTSVM)综合自然伽马(GR)、电阻率(RL)、光电效应(PE)、中子密度孔隙度差异(DPHI)和平均中子密度孔隙度(PHIND)五种测井参数,以及相对位置(RELPOS)和非海洋/海洋指标(NM_M)两种地质约束变量,识别出9种岩性.将识别结果与传统支持向量机、深度神经网络等方法进行对比与分析,发现IMCTSVM算法优于上述两种分类算法,取得了较好的识别效果.     

松辽盆地深层火山岩含气储层产能预测

松辽盆地深层火山岩是当前大庆地球物理、地质、地球化学研究的主要领域之一,已取得丰硕的成果,火成岩含气储层产能作为一个表示动态特征的参数,是储层评价的重要指标之一,本文讨论了火成岩含气储层的产能与测井响应之间的关系,探讨了根据测井资料应用人工神经网络技术预测火成岩含气储层产能的方法,利用已知气井测试结果和测井资料作为网络的训练样本,根据网络学习训练结果,输入测井资料等静态参数,可预测储集层的产能,根据这种关系采用神经网络技术实现了测井对产能的预测评价,从而为大庆深部火成岩含气储层的开发提供了一定的依据。     

中低渗油藏剩余油饱和度监测技术系列优化与应用

套管井剩余油饱和度监测技术是研究剩余油分布最重要的测井方法.针对研究区块中-低孔隙度、中-低渗透率、高矿化度地层水的油藏特点与饱和度监测的现状,对比研究了现有监测技术与仪器装备的性能指标,同时考虑脉冲中子、非弹性散射与地层中元素的相互作用的特性,提出了适用于这种不同油藏地质特点的剩余油饱和度监测系列优化方案.针对某区块编制了整体监测方案,并且对监测资料进行精细处理与解释.多口井实施和应用表明,该方案与应用效果良好.     

小井径双源距碳氧比C/O测井的影响因素及处理

研究各种地层和井眼环境因素对碳氧比C/O测井长、短源距探测器的影响规律，可以为C/O测井仪器的刻度方法提供指导，为解释模型的建立和数据处理提供依据.本文用Monte Carlo方法，计算了C/O值随井眼直径、水泥环厚度、套管直径、孔隙度、含油饱和度、地层岩性和油密度的变化规律.从中看出，当井内流体为油时，井径或套管直径增大，C/O值增大，井眼影响增大；当井内流体为水时，井径或套管直径的增大，C/O值减小；水泥环厚度增加时C/O值减小；当井眼条件不变时，地层孔隙度越大，含油饱和度越大，C/O值越大，对测井越有利；反之，地层孔隙度越小，含油饱和度越小，C/O值越小，对测井不利；地层岩性对C/O值的影响显著，相同条件下，碳酸盐岩比砂岩的C/O值高；油密度越大，C/O值越大.文中还提出了一种消除这些因素影响的数据处理方法     

复杂孔隙储层三维岩石物理模版

复杂孔隙储层往往同时发育孔缝洞等多种孔隙类型,这种孔隙结构的复杂性使得岩石的速度与孔隙度之间的相关性很差.经典的二维岩石物理模版只研究弹性参数与孔隙度和饱和度之间的定量关系,而不考虑孔隙结构的影响,用这样的模版来预测复杂孔隙储层的物性参数时带来很大偏差.本文首先证明多重孔隙岩石的干骨架弹性参数可以用一个等效孔隙纵横比的单重孔隙岩石物理模型来模拟;进而基于等效介质岩石物理理论和Gassmann方程,建立一个全新的三维岩石物理模版,用它来建立复杂孔隙岩石的弹性性质与孔隙扁度及孔隙度和饱和度之间的定量关系;在此基础上,预测复杂储层的孔隙扁度、孔隙度以及孔隙中所包含的流体饱和度.实际测井和地震反演数据试验表明,三维岩石物理模版可有效提高复杂孔隙储层参数的预测精度.     

基于单能窄束伽马射线算法计算高分辨率屏蔽自然伽马探测器的三维测井响应

本文应用单能窄束伽马射线理论建立高分辨率屏蔽自然伽马探测器三维测井响应的数值模拟算法.首先利用单能窄束原理,用稳态扩散方程描述伽马光子密度的空间分布,根据扩散方程基本解、放射源空间分布、探测器位置和屏蔽情况,将屏蔽探测器上的总伽马通量表示成放射性地层中的有效探测区域上的体积分和晶体表面上有效接收面的面积分形式.并根据伽马射线传播路径和探测器屏蔽情况,给出有效探测区域的解析表达式,通过数值积分法计算任意复杂情况下探测器上的自然伽马通量,获得伽马测井响应的3D数值模拟算法.并通过数值模拟结果与模型井数据的对比验证了该算法的有效性.最后通过3D数值模拟算法系统研究考察晶体形状（圆柱形晶体、方形晶体）、晶体长度、仪器在井轴中的位置（居中屏蔽探测器、贴井壁屏蔽探测器）、以及测速等不同情况下自然伽马测井响应,设计出新型高分辨率自然伽马测井仪器.     

Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study

Estimations of porosity and permeability from well logs are important yet difficult tasks encountered in geophysical formation evaluation and reservoir engineering. Motivated by recent results of artificial neural network (ANN) modelling offshore eastern Canada, we have developed neural nets for converting well logs in the North Sea to porosity and permeability. We use two separate back-propagation ANNs (BP-ANNs) to model porosity and permeability. The porosity ANN is a simple three-layer network using sonic, density and resistivity logs for input. The permeability ANN is slightly more complex with four inputs (density, gamma ray, neutron porosity and sonic) and more neurons in the hidden layer to account for the increased complexity in the relationships. The networks, initially developed for basin-scale problems, perform sufficiently accurately to meet normal requirements in reservoir engineering when applied to Jurassic reservoirs in the Viking Graben area. The mean difference between the predicted porosity and helium porosity from core plugs is less than 0.01 fractional units. For the permeability network a mean difference of approximately 400 mD is mainly due to minor core-log depth mismatch in the heterogeneous parts of the reservoir and lack of adequate overburden corrections to the core permeability. A major advantage is that no a priori knowledge of the rock material and pore fluids is required. Real-time conversion based on measurements while drilling (MWD) is thus an obvious application.     

双频介电测井方法与实现

介电测井是一种测量地层中电磁波传播速度和相位差的测井方法,能计算出地层的介电常数和电导(阻)率,在油田广泛推广用于评价地层孔隙度和含水饱和度,以区分油、水、气层.本文结合在国产数控测井地面系统配接Atlas1531/1532介电测井仪的经验,详细介绍了介电的测井原理和数据处理,给出了测井方法的具体实现和应用.该方法已经在多种地面系统和测试台架得到验证,通过现场测井取得良好的应用效果.     

三维地震神经网络反演在探明煤层富水灾害层的应用

在常规测井约束反演的的基础上,开展神经网络特征参数反演,将波阻抗等地震属性转化为与含水性更为密切的孔隙度、视电阻率数据体,使地震反演的地质属性与测井上的地质属性达到最优的相关性,从而实现应用三维地震对煤层顶板富水进行评价的目的。由于煤层顶板富水区的特殊性质,它同样也是地震后的易破坏层,因而对它的探明从抗震角度以及震害预测角度都是有价值的。以淮北某采区为例,通过孔隙度及电阻率的神经网络反演对研究区10#煤层顶板的富水性进行预测。反演结果表明采区北部发育一个强富水陷落柱,与钻孔揭示结果吻合。采区西部10#煤层顶板与第四系含水层呈不整合接触关系,神经网络反演结果预测为强富水区,同样与井下工程揭示富水特征吻合。利用多属性融合的神经网络反演可有效预测煤层顶板的富水特征,为煤矿安全生产以及抗震提供重要保障。     

密度测井和中子测井的相关性及其在识别天然气层中的应用

伽马射线与地层介质的康普顿（Ｃｏｍｐｔｏｎ）效应是密度测井的理论基础，介质对伽马射线的康普顿吸收系数取决于介质的原子核性质及其体积密度．（超热或热）中子测井主要反映地层介质的减速性质，而后者主要取决于介质的含氢量．介质含氢量是由介质的组分及其体积密度决定的．本文从上述两种测井方法的测量原理出发，从理论上证明了这两种测井方法并非两种互相独立的测量手段．对于含油气盆地常见沉积岩石及矿物，两种测井测量结果具有相关性．这一结论为密度、中子测井方法组合地质应用提供了理论依据，利用密度－中子测井的相关性可有效地识别轻质油气层特别是天然气层．实例表明，无论水层还是油气层，两种测井曲线均具有良好的相关性．利用这种相关性明显改善了识别天然气层和划分油－气界面的精度．     

Neutron and density logging responses to gas reservoir for well-balanced and under-balanced logging:Gas reservoirs of sandstone in a western China field

Studying the response differences between neutron and density logging of gas reservoir for well-balanced and under-balanced logging will be of significance in evaluation of gas reservoir under the under-balanced condition and application of logging data. With Monte Carlo simulation technique,the paper obtains the relationship between neutron and density logging measurement and borehole di-ameter,porosity or gas saturation for well-balanced and under-balanced logging. The conclusions show that the response trend of under-balanced logging to gas reservoirs agrees with that of well-balanced logging with small borehole,and under-balanced logging data can be used usually as well-balanced logging data. When borehole diameter is large,under-balanced logging data should be corrected for the influences of borehole.     

密度测井和中子测井的相关性及其在识别天然气层中的应用

伽马射线与地层介质的康普顿（Ｃｏｍｐｔｏｎ）效应是密度测井的理论基础，介质对伽马射线的康普顿吸收系数取决于介质的原子核性质及其体积密度．（超热或热）中子测井主要反映地层介质的减速性质，而后者主要取决于介质的含氢量．介质含氢量是由介质的组分及其体积密度决定的．本文从上述两种测井方法的测量原理出发，从理论上证明了这两种测井方法并非两种互相独立的测量手段．对于含油气盆地常见沉积岩石及矿物，两种测井测量结果具有相关性．这一结论为密度、中子测井方法组合地质应用提供了理论依据，利用密度－中子测井的相关性可有效地识别轻质油气层特别是天然气层．实例表明，无论水层还是油气层，两种测井曲线均具有良好的相关性．利用这种相关性明显改善了识别天然气层和划分油－气界面的精度．     

Experimental research on the rock electrical parameter dispersion properties in the displacement process

Based on experiments on rock electrical parameter dispersion properties in both water displacing oil and oil displacing water processes, it is found that the frequency dispersion properties of electrical parameters in water-bearing rocks are quite different from those in oil-bearing rocks. Oil saturation can severely impact the rock electrical parameter dispersion properties and as the oil saturation increases, the properties are more distinguishing. The variation of formation water salinity has a significant effect on the rock resistivity, hence the modulus of the rock complex resistivity varies with water saturation and a U-shaped curve occurs over a single frequency, just like the normal resistivity. However, the effect of formation water salinity on the rock frequency dispersion is less than that of oil saturation. As a result, the rock electrical parameter dispersion properties will have a significant potential of application in the evaluation of watered-out zones as well as identification of oil-bearing zones and water-bearing zones.