高阶神经网络在储层分布参数定量预测中的应用

文中介绍了高阶神经网络的原理和方法,提出把高阶神经网络应用于储层分布参数的定量预测方法,并给出双河Ⅳ4（5-9）油组顶层孔隙度与深度的预测算例,并讨论了不同的储层分布参数的预测方法的优缺点     

储层预测中BP神经网络的应用

论述了BP神经网络的基本原理及在油气预测中的应用，特别讨论了网络的学习和网络技术的改进，通过SN油田SGH组的油气预测，表明利用BP神经网络进行油气藏的储层预测是有效的。     

应用三重孔隙模型评价火成岩储层

火成岩岩性特殊,储层非均质性非常严重,裂缝、孔洞发育,属于典型的多重孔隙介质储层,定量计算火成岩储层基质和裂缝的孔隙度和饱和度是评价火成岩油气层的关键。由基质、天然裂缝和非连通孔洞组成的储层可以模拟成基质和裂缝的并联导电网络,非连通孔洞和基质／裂缝的组合的串联导电网络。采用密度测井资料求取总孔隙度,用深浅双侧向电阻率计算裂缝孔隙度,采用密闭取心分析的孔隙度作为基质孔隙度。通过建立以基质、天然裂缝和非连通孔洞组成的储层的三重孔隙解释模型,找到适合于火成岩缝洞型复杂储层的饱和度方程。将三重孔隙解释模型应用于松辽盆地北部火成岩储层,计算的含水饱和度平均相对误差由16．85％降低至12．42％,精度明显提高,为火成岩储层评价、储量计算与地质建模提供了定量参数。     

测井资料在储层预测研究中的应用

测井解释技术面临三方面制约因素:(1)长期以来仪器的发展速度快于解释评价的发展速度,使大量有用信息被长期掩盖;(2)现今测井评价方法强于微观解释而弱于宏观分析,明显限制了测井技术的发展和受重视程度;(3)油气勘探开发目标的日益复杂化,对测井技术提出了更高要求.开展测井储层预测技术的探索,在某种程度上可弥补上述不足.本文提出测井储层预测技术研究的二个依据,(1)将测井资料作为宏观地质作用与微观岩石信息统一关系论证的一个关键点,寻找测井信息能准确描述二者具有统一成因关系的证据;(2)以构造-沉积演化作为研究切入点,预测特殊储层在时、空的分布关系.基于上述认识,以印尼B油区裂缝型砂岩的预测以及大港油田Q50断块低阻油层的预测研究为例,分别介绍了测井储层预测技术的2个应用实例.     

基于多任务学习的测井储层参数预测方法

基于多任务神经网络模型, 提出一种多任务测井储层参数预测方法, 利用测井数据对储层孔隙度、渗透率及含水饱和度同时进行预测.分别采用同架构和异架构多任务模型对测井储层参数进行预测, 通过数值实验对比, 多任务预测模型有效提升了单任务储层参数预测模型的效果, 且提升幅度与模型结构有关, 异架构多任务模型的总体预测效果好于同架构多任务模型.以平均相对误差(MAPE)作为模型评价标准, 针对本研究所采用的数据集, 同架构多任务模型的孔隙度、渗透率和含水饱和度在测试集上的MAPE约为6%、17%和9%, 相较于单任务模型, 预测效果分别提升约30%、20%和10%.异架构多任务模型的孔隙度、渗透率和含水饱和度, 在测试集上的MAPE约为6%、13%和6%, 相较于单任务模型分别提升超过2%、60%和10%.

     

用核磁共振测井资料评价碳酸盐岩等复杂岩性储集层

对于碳酸盐岩等复杂岩性油气藏,由于其储集空间复杂、非均质性强等因素,用常规测井技术难以进行准确描述.核磁共振测井测量的对象是储层空间中的流体,因而可以直接用来划分储集层,而且能提供几乎不受岩性影响的孔隙度和渗透率等参数;同时,由于其T2分布表征了岩石的孔隙结构,所以可以根据T2分布形态判断有效裂缝和溶蚀孔洞.通过多口井的岩心对比和成像测井对比,研究出了一套用MRIL和CMR 的T2分布形态评价储集空间的方法;此外,在特定条件下,根据核磁计算的有效孔隙度和可动流体体积给出了一个计算含油饱和度的公式.利用这些核磁测井技术及其分析方法对车古20古潜山碳酸盐岩储集层裂缝及溶蚀孔洞发育特征进行准确描述,还计算了埕北潜山复杂复杂岩性油气藏的饱和度,对于储量计算具有重要意义.     

砂砾岩储层测井评价研究

砂砾岩储层岩性复杂、非均质性强,储层间非渗透性隔层类型多,储层基质孔隙度有时很低,从而使测井资料准确划分有效储层有很大的难度;砂砾岩体储层母岩类型变化大,岩石骨架参数很难确定,电阻率测量受岩石骨架、粘土含量和孔隙结构影响严重,反映储层孔隙流体性质的信息弱,使储层流体性质难以判断,油、气、干层界限的电性特征极不明显.通过核磁共振和井壁微电阻扫描成像测井,可以直观观察到岩石成分和粒径的变化,通过T2谱分布直观显示核磁测量井段的孔径分布,计算出各种类型孔隙度和渗透率参数,为砂砾岩有效划分储层和测井评价提供了可靠的依据.     

神经网络在石油测井解释中的应用综述

常规测井解释往往通过"四性关系"研究建立储层参数解释模型,将测井信息转化成地质参数开展储层评价,但对于复杂岩性储层、低渗低阻等储层而言,将测井信息转化为储层参数及利用测井曲线开展储层评价往往存在多解性.作为一种非线性数学方法,神经网络具有多信息融合、综合预测的功能,在解决一些复杂、非线性问题领域展示了强大的生命力.将神经网络引入石油测井解释中,综合利用多种测井信息开展岩性、物性及含油性分析,可有效提高石油测井解释的精度和效率;利用神经网络技术开展流体性质识别和储层裂缝研究可有效破解此类世界性难题.为此,本文对神经网络技术在石油测井解释中的应用进行了综述,并对应用中应注意的事项进行分析,最后进行了应用展望.     

基于LSTM循环神经网络的储层物性参数预测方法研究

储层物性参数是反映储层油气储集能力的重要参数,表征了不同地质时期的沉积特征.地球物理测井参数由深及浅反映了不同地质时期的声、放、电等沉积特征,因而测井参数和泥质含量(孔隙度)之间有很强非线性映射关系,并具有时间序列特征.充分利用多种测井参数预测储层泥质含量和孔隙度对于储层精细描述具有十分重要的意义.深度学习技术具有极强的数据结构挖掘能力,目前,全连接的深度神经网络已经在泥质含量预测进行了初步尝试并取得了较好的效果.而长短时记忆(LSTM)循环神经网络更适合解决序列化的数据问题,因此本文提出基于LSTM循环神经网络利用多种测井参数进行泥质含量和孔隙度预测的方法,预测结果的均方根误差比常规全连接深度神经网络分别下降了42.2%和48.6%,实际应用表明,对于具有序列化特性的泥质含量和孔隙度,LSTM循环神经网络预测的准确性和稳定性要明显优于常规全连接深度神经网络.     

自组织特征映射神经网络在测井岩性识别中的应用

为了解决测井岩性识别问题,引入具有较强的聚类和容错能力的自组织特征映射(SOFM)神经网络．在说明SOFM网络的模型和算法的基础上,结合某地的实际测井资料,建立SOFM网络岩性识别模型,进行岩性识别的应用研究．结果表明,识别的准确率较高,证明SOFM网络可以用于解决测井岩性识别问题,具有很好的应用前景．     

基于BP神经网络的波阻抗反演及应用

人工神经网络是近期发展最快的人工智能领域研究成果之一.本文在介绍BP神经网络的有关原理的基础上,提出一种基于BP神经网络模型的波阻抗反演方法,该方法克服了常规基于模型的波阻抗反演方法严重依赖于初始模型的选择和易陷入局部最优等局限性.利用该方法对实际地震剖面进行了波阻抗参数反演处理,结果表明人工神经网络方法在波阻抗反演中的应用是可行的并且是有效的.     

一种基于神经网络的探地雷达信号解释研究

运用人工神经网络理论和方法,建立了用于隧道衬砌厚度探地雷达探测信号解释的BP神经网络模型,对某公路隧道衬砌检测厚度进行了分析应用,并与钻孔取芯结果进行比较,实践证明,该方法可提高探地雷达信号解释精度和工作效率．     

地震数据处理解释一体化在隐蔽油气藏勘探中的应用

Nowadays, it becomes very urgent to find remain oil under the oil shortage worldwide.However, most of simple reservoirs have been discovered and those undiscovered are mostly complex structural, stratigraphic and lithologic ones. Summarized in this paper is the integrated seismic processing/interpretation technique established on the basis of pre-stack AVO processing and interpretation.Information feedbacks occurred between the pre-stack and post-stack processes so as to improve the accuracy in utilization of data and avoid pitfalls in seismic attributes. Through the integration of seismic data with geologic data, parameters that were most essential to describing hydrocarbon characteristics were determined and comprehensively appraised, and regularities of reservoir generation and distribution were described so as to accurately appraise reservoirs, delineate favorite traps and pinpoint wells.     

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.     

准噶尔盆地砾岩油藏水淹层的测井评价方法研究

基于对准噶尔盆地某砾岩油藏的淡水水淹机理的实验认识,考察了砾岩油藏水淹层的测井响应特征,形成了一套定性识别水淹层和划分水淹级别的方法.在该基础上,按照"岩心刻度测井"的原则,统计建立了一套适用于该砾岩油藏淡水水淹层的测井评价方法.该方法在7个区块17口井的测井解释结果表明,水淹级别按6级划分的解释符合率达85%左右.     

微电阻率扫描成像测井方法应用及发展前景

阐述了微电阻率扫描成像（简称FMI）测井方法在我国的应用和发展前景。微电阻率扫描成像测井能够对井筒周围地层介质直观、清晰、高分辨率地成像,有助于解决当前测井技术面临的三大地质难题：砂泥岩薄互层储层的有效划分,裂缝性油藏的裂缝和储集性能分析,复杂岩性油藏的参数评价。但也存在测速慢、费用高、定量解释效果差等不足。国际三大测井公司在这方面的研究起步较早,并取得了重要的成果.且早已投入商用.我国在这方面的研究也有很大进展,在未来的发展中,应针对一些理论和技术上的难点分别突破,尽快推出更加适合我国地质特征的微电阻率扫描成像测井仪器和解释手段.     

Short‐term inflow forecasting using an artificial neural network model

The primary objective of this study is to investigate the possibility of including more temporal and spatial information on short‐term inflow forecasting, which is not easily attained in the traditional time‐series models or conceptual hydrological models. In order to achieve this objective, an artificial neural network (ANN) model for short‐term inflow forecasting is developed and several issues associated with the use of an ANN model are examined in this study. The formulated ANN model is used to forecast 1‐ to 7‐h ahead inflows into a hydropower reservoir. The root‐mean‐squared error (RMSE), the Nash–Sutcliffe coefficient (NSC), the A information criterion (AIC), B information criterion (BIC) of the 1‐ to 7‐h ahead forecasts, and the cross‐correlation coefficient between the forecast and observed inflows are estimated. Model performance is analysed and some quantitative analysis is presented. The results obtained are satisfactory. Perceived strengths of the ANN model are the capability for representing complex and non‐linear relationships as well as being able to include more information in the model easily. Although the results obtained may not be universal, they are expected to reveal some possible problems in ANN models and provide some helpful insights in the development and application of ANN models in the field of hydrology and water resources. Copyright © 2002 John Wiley & Sons, Ltd.     

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.