基于Markov链模型的储层岩相随机模拟

在油气储层随机建模研究中，基于Markov链模型的方法是一类较受欢迎的技术，同时也是一类不成熟的技术，问题的症结之一在于侧向的转移概率矩阵很难求取，针对这种情况，作者在深入理解Walther相律的基础上，借鉴模拟退火算法的相应思路，提出了一种岩相模拟的新方法，该方法依据不同岩相的百分比进行随机模拟得到一幅初始图象，而后以按岩相组织剖面得到的垂向和侧向的岩相转移概率矩阵的相似性作为判别标准对图象进行扰动，直至得到满意的图象，二维模型试算结果表明了这种岩相随机模拟方法的可行性。     

沉积模型和储层随机建模

沉积模型是地层分析的重要工具，可以分为比例尺模型、概念模型和数学模型三大类型，其中数学模型又可分为确定性模型和随机模型。在实际地层分析及模拟工作中，特别是在小尺度问题的研究中，采用随机模型(或称统计学模型)往往更为有利。储层随机建模技术，作为这方面研究的典范，近年来成为储层预测和风险评价的一项较为有效的手段。然而，由于研究目标的复杂性，不同沉积模型之间的嵌套制约关系亦应引起重视。     

基于约束Markov决策过程的初至自动识别技术

随着地震数据采集技术的进步,地震数据量日益增加,全自动、高精度的地震初至走时拾取技术受到了更加广泛的关注.本文将初至拾取看作特征空间内带约束的Markov决策过程,在奖励函数空间,按一定准则全局寻优获得积累奖励值最大的路径,从而达到在高维空间自动拾取初至信息的目的.同时,状态值函数中包含与距离相关的折扣因子γ,使Markov决策过程拾取初至能够考虑地震数据的横向连续性,并且回避地震数据中的坏道信息.在此基础上,本文方法进一步引入受空间几何信息约束的动作（Actions）和转移概率（Transitions Probability）,从而降低了对起始状态和折扣因子选取的难度,让地震数据初至走时拾取更加准确和自动化.实际数据测试结果表明,在初至能量较弱（信噪比较低）情况或浅层存在相邻较近复杂波形时,本文提出的约束Markov算法仍能准确地进行初至走时的自动拾取,并且具有一定的质量监控能力,让拾取结果更有物理意义.

     

琼东南盆地崖南区少井岩性建模技术研究

崖南区是琼东南盆地已证实的富生烃区,几口已钻井都已证实主要目的层段为高压地层,利用常规的压力预测方法预测新钻井的压力会出现较大的误差.若是从区域应力角度入手预测新钻井的压力误差会减小,其预测基础为岩性模型.对于已开发的油气田,利通常规的岩性建模方法可以建立较好的岩性模型;但是对于崖南区而言,由于地震资料品质不是很好,同时本区钻井较少,很难通过常规的建模方法建立岩性模型,所以本区研究重点是如何利用少井建立岩性模型.通过研究认为若完成崖南区的岩性建模必须改进建模流程,改进的岩性建模流程克服了常规岩性建模在崖南区存在的问题,主要有三方面的优点:1)不采用相模型约束岩性建模,解决了由于研究区相模型划相较粗很难约束岩性模型建立的问题;2)属性模型控制岩性模型的横向变化趋势,解决了几种常规属性与岩性间没有较好关系的问题;3)利用泥质含量结合岩性资料建立岩性体,得到的岩性模型比较接近实际情况.C井钻前完成岩性模型建立,利用C井井点位置提取岩性数据与本井钻后录井岩性数据对比,发现预测岩性与录井岩性的吻合程度很高,证明改进的岩性建模思路在崖南少井区可用.     

基于马尔科夫链模型的鄱阳湖流域水文气象干旱研究

基于气象和水文干旱的二维变量干旱状态基础上,通过一阶马尔科夫链模型对二维变量干旱状态进行频率、重现期和历时分析,建立水文气象干旱指数,从干旱灾害形成、演变和持续3方面对干旱灾害进行研究,同时预测未来6个月非水文干旱到水文干旱的概率.结果表明:(1)修河流域在干旱形成中危害大,抚河流域和修河流域在干旱演变中危害大,赣江流域和饶河流域在干旱持续中危害大;(2)鄱阳湖流域状态4(气象、水文干旱)发生的频率最高,为0.30,连续湿润或者干旱的概率最大,湿润状态(状态2)与水文干旱(状态4、状态5(气象湿润、水文干旱))的相互转移概率最低;(3)在长期干旱预测中,鄱阳湖流域从状态2转到状态4和状态5的平均概率为0.11,属最低,而状态1(气象、水文无旱)和状态3(气象干旱、水文湿润)到达状态4的概率为0.23,发生概率最大.修河流域在非水文干旱状态下未来发生气象、水文干旱状态的平均概率为0.28,是"五河"中最高的,而赣江流域在正常或者湿润状态下未来发生气象、水文干旱的概率最低,为0.18,该研究对于鄱阳湖流域水文气象干旱的抗旱减灾具有重要理论与现实意义.     

基于含砂率的相控地质统计学反演储层预测技术研究

陆相湖盆三角洲沉积储层横向变化快、纵向薄砂泥互层,储层预测难度大.针对该问题,本文以乌石油田群流沙港组二段中层序为背景,提出了一种基于含砂率的相控地质统计学反演储层预测技术.通过单井统计,建立含砂率与沉积微相的关系;再利用平面含砂率横向约束建立三维含砂率初始模型;进而运用贝叶斯理论将三维含砂率初始模型作为约束条件融入到...     

一种基于圆弧片状压电振子的相控圆弧阵随钻声波测井辐射器

随钻三维反射声波成像测井技术可以实时地对井周围的地层构造和地质体进行成像,为地质导向钻井提供必要的信息,是下一代声波测井技术的发展方向.针对该项技术,提出了一种基于圆弧片状压电振子的相控圆弧阵声波测井辐射器,推导了该辐射器的声学性能在波数-频率域的数学描述,并采用了实轴积分的方法对该辐射器在无限大液体、井旁地层中产生的波场进行了求解.研究结果显示,无论在无限大液体中还是在充液井孔内,该声源均可以向任意方位定向辐射能量,其水平指向性图主瓣明显,旁瓣级低,具有较高的方位分辨率;相控阵技术能够使得即使在较低的频率下,该声源辐射的声场仍具有较好的方位特征;与传统的单极子反射成像技术相比,有希望利用该声源发展一种具有较好的方位分辨能力和更深径向探测深度的随钻反射成像测井方法;与偶极子反射成像技术相比,采用该声源可以在周向上360°范围内确定反射体的方位,能够消除井旁地层界面方位测量的多解性.     

基于物理的随机地震动模型研究

基于物理联系研究地震动随机性,建立了随机地震动与基底输入傅氏谱、场地固有圆频率和场地等价阻尼比之间的物理关系,从随机傅氏谱函数角度描述了地震动随机过程的随机性本质。结合Ⅳ类工程场地的实测地震动记录资料,由数值方法识别了给出基本随机变量的概率分布参数。与实测记录对比表明,本文建立的随机地震动模型具有明确的物理概念,可充分反映地震动的变异性特征。     

基于随机地震动模型的结构随机地震反应谱及其应用

本文考虑给定地震烈度下地震地面运动的随机过程性，得强震记录统计确定的地震持时和我国地震规范采用的地震地面最大加速度平均值，确定了平稳过滤有色噪声地震动模型的参数；通过大量计算和回归分析，得到了单质点振子均方地震位移的实用计算公式，提出了随机地震反应谱，等效随机地震静荷载及结构地震随机反应和可靠性分析的实用方法，把结构在随机地震动作用下的动力可靠性分析转化成了结构在等效随机地震静荷载作用下的静力可靠     

基于FFT-MA的随机介质建模方法

随机介质是描述地球介质小尺度非均匀性的有效模型,在地震散射波场分析、储层描述等领域具有广泛应用,快速准确的随机介质建模方法是开展相关研究的基础与前提.本文首次将FFT-MA（Fast Fourier Transform Moving Average）算法引入到随机介质建模研究中,分析了该方法与传统随机介质建模方法相比具有的优势,并提出了基于FFT-MA的非平稳随机介质建模方法.建模实验表明,与传统的基于谱分解定理的随机介质建模方法相比,基于FFT-MA的方法在空间域产生随机数序列,使随机数序列与结构参数分离,因此,随机数序列与所建模型存在空间上的对应关系,可以分区域建模和局部修改模型.在非平稳随机介质模型建模时,滑动窗口的大小可以根据自相关长度变化而变化,避免了每个采样点都建立一次完整大小的模型,提高了建模效率.因此,FFT-MA随机介质建模方法能准确构建满足自相关函数要求的平稳及非平稳随机介质模型,具有建模效率高、灵活、实用的优点.

     

布谷鸟马尔科夫链蒙特卡洛混合高斯地质统计学随机反演

地质统计学随机反演可以获得比常规反演更高分辨率的结果,目前已成为储层高分辨率预测的主流方法.地下不同岩相储层参数存在明显差异,本文在地质统计学反演框架下构建了岩相和储层参数同步反演目标函数,实现不同岩相条件下储层参数分布精细描述.在求解该高维数据多参数同步反演问题时,本文将可以动态调节搜索步长的布谷鸟算法与马尔科夫链蒙...     

Streamflow simulation and skewness preservation based on the bootstrapped stochastic models

The stochastic model has been widely used for the simulation study. However, there was a difficulty in the reproduction of the skewness of observed series and so the stochastic model for the skewness preservation was appeared. While the skewness in the residuals of the stochastic model has been considered for the skewness preservation this study uses a random resampling technique of residuals from the stochastic models for the simulation study and for the investigation of the skewness coefficient. The main advantage of this resampling scheme, called the bootstrap method is that it does not rely on the assumption of population distribution and this study uses the combined model of the stochastic and bootstrapped models. The stochastic and bootstrapped stochastic (or combined) models are used for the investigations of skewness preservation and of the reproduction of probability density function between the simulated series. The models are applied to the annual and monthly streamflows of Yongdam site in Korea and Yakima river, Washington, USA for the streamflow simulation study then the statistics and probability density functions for the observed and simulated streamflows are compared. As the results the bootstrapped stochastic model reproduces the skewness and probability density function much better than the stochastic model. This evidences suggest that the bootstrapped stochastic model might be more appropriate than the stochastic model for the preservation of skewness and for simulation purposes of the series.     

Kriging metamodeling in seismic risk assessment based on stochastic ground motion models

An efficient computational framework is presented for seismic risk assessment within a modeling approach that utilizes stochastic ground motion models to describe the seismic hazard. The framework is based on the use of a kriging surrogate model (metamodel) to provide an approximate relationship between the structural response and the structural and ground motion parameters that are considered as uncertain. The stochastic character of the excitation is addressed by assuming that under the influence of the white noise (used within the ground motion model) the response follows a lognormal distribution. Once the surrogate model is established, a task that involves the formulation of an initial database to inform the metamodel development, it is then directly used for all response evaluations required to estimate seismic risk. The model prediction error stemming from the metamodel is directly incorporated within the seismic risk quantification and assessment, whereas an adaptive approach is developed to refine the database that informs the metamodel development. The ability to efficiently obtain derivative information through the kriging metamodel and its utility for various tasks within the probabilistic seismic risk assessment is also discussed. As an illustrative example, the assessment of seismic risk for a benchmark four‐story concrete office building is presented. The potential that ground motions include near‐fault characteristics is explicitly addressed within the context of this example. The implementation of the framework for the same structure equipped with fluid viscous dampers is also demonstrated. Copyright © 2015 John Wiley & Sons, Ltd.     

Modeling long term lake variations by physically based stochastic dynamic models

Several physically based stochastic dynamic models (SDM) are described including year-to-year variations of water volume in terminal and non-terminal lakes, streamflow of lake-fed rivers, and salinity of an inland sea (the Sea of Azov). All of these models are based upon the SDM of water volume of terminal lakes developed by Kritzky and Menkel in 1946 in co-operation with Kolomogorov. Explicit formulae are derived for second order statistical moments of the output processes, including variance, correlation function, spectra, etc., under the assumption that the forcing functions from stationary random sequences. The least-squares prediction problem is solved for both stationary and non-stationary cases. Some of the processes are shown to possess high statistical predictability. Actual predictions are compared with independent observations. Problems for further study are stated.     

A comparative study of nonlinear Markov chain models for conditional simulation of multinomial classes from regular samples

Simulating fields of categorical geospatial variables from samples is crucial for many purposes, such as spatial uncertainty assessment of natural resources distributions. However, effectively simulating complex categorical variables (i.e., multinomial classes) is difficult because of their nonlinearity and complex interclass relationships. The existing pure Markov chain approach for simulating multinomial classes has an apparent deficiency—underestimation of small classes, which largely impacts the usefulness of the approach. The Markov chain random field (MCRF) theory recently proposed supports theoretically sound multi-dimensional Markov chain models. This paper conducts a comparative study between a MCRF model and the previous Markov chain model for simulating multinomial classes to demonstrate that the MCRF model effectively solves the small-class underestimation problem. Simulated results show that the MCRF model fairly produces all classes, generates simulated patterns imitative of the original, and effectively reproduces input transiograms in realizations. Occurrence probability maps are estimated to visualize the spatial uncertainty associated with each class and the optimal prediction map. It is concluded that the MCRF model provides a practically efficient estimator for simulating multinomial classes from grid samples.     

Drought forecasting using stochastic models

Drought is a global phenomenon that occurs virtually in all landscapes causing significant damage both in natural environment and in human lives. Due to the random nature of contributing factors, occurrence and severity of droughts can be treated as stochastic in nature. Early indication of possible drought can help to set out drought mitigation strategies and measures in advance. Therefore drought forecasting plays an important role in the planning and management of water resource systems. In this study, linear stochastic models known as ARIMA and multiplicative Seasonal Autoregressive Integrated Moving Average (SARIMA) models were used to forecast droughts based on the procedure of model development. The models were applied to forecast droughts using standardized precipitation index (SPI) series in the Kansabati river basin in India, which lies in the Purulia district of West Bengal state in eastern India. The predicted results using the best models were compared with the observed data. The predicted results show reasonably good agreement with the actual data, 1–2 months ahead. The predicted value decreases with increase in lead-time. So the models can be used to forecast droughts up to 2 months of lead-time with reasonably accuracy.     

基于马尔科夫随机场的岩性识别方法

通过地震反演数据识别岩性,是地震反演的一项基本任务.由于不同岩性的弹性参数范围常常存在一定程度的重叠,所以给岩性识别带来了很大的困难.本文以叠前反演的弹性参数为基础,通过马尔科夫随机场(Markov Random Field简写为MRF)建立先验模型,按照解释好的测井资料,对不同岩性的弹性参数进行统计,得到计算所需的参数,在贝叶斯(Bayesian)框架下建立岩性分类的目标函数,达到岩性识别的目的.通过马尔科夫随机场建立先验模型,能够建立相邻点间的相互作用关系,得到横向上延续的岩性剖面.本文使用一个楔形模型和Marmousi Ⅱ模型对该方法进行了测试,结果表明,该方法有效可行.同时,本文通过加入误差的方法,检验了反演存在误差对识别结果的影响.     

A stochastic framework to assess the performance of flood warning systems based on rainfall‐runoff modeling

Nowadays, Flood Forecasting and Warning Systems (FFWSs) are known as the most inexpensive and efficient non‐structural measures for flood damage mitigation in the world. Benefit to cost of the FFWSs has been reported to be several times of other flood mitigation measures. Beside these advantages, uncertainty in flood predictions is a subject that may affect FFWS's reliability and the benefits of these systems. Determining the reliability of advanced flood warning systems based on the rainfall–runoff models is a challenge in assessment of the FFWS performance which is the subject of this study. In this paper, a stochastic methodology is proposed to provide the uncertainty band of the rainfall–runoff model and to calculate the probability of acceptable forecasts. The proposed method is based on Monte Carlo simulation and multivariate analysis of the predicted time and discharge error data sets. For this purpose, after the calibration of the rainfall–runoff model, the probability distributions of input calibration parameters and uncertainty band of the model are estimated through the Bayesian inference. Then, data sets of the time and discharge errors are calculated using the Monte Carlo simulation, and the probability of acceptable model forecasts is calculated by multivariate analysis of data using copula functions. The proposed approach was applied for a small watershed in Iran as a case study. The results showed using rainfall–runoff modeling based on real‐time precipitation is not enough to attain high performance for FFWSs in small watersheds, and it seems using weather forecasts as the inputs of rainfall–runoff models is essential to increase lead times and the reliability of FFWSs in small watersheds. Copyright © 2013 John Wiley & Sons, Ltd.