基于数值模拟的油藏生产动态优化算法

对于大规模油藏开发优化,常规优化算法存在收敛速度慢、寻优效率低、难以与现场结合等问题。针对这些问题,建立了油水井动态调控优化模型,并结合模拟退火遗传算法和拉丁超立方采样算法寻找模型全局最优解,同时,利用同步扰动随机逼近算法提高了模型局部求解收敛速度,研发编制了油水井动态调控优化软件,并应用于大庆油田H区块。对比油田常规油水井生产制度,油水井动态调控优化模型的最优方案使H区块累计产油量在5年内增加了5.68×10⁴ m³,较好地解决了油水井生产动态控制优化问题,也为大规模油田高效开发提供了一种新的方法。     

人工神经网络在油藏数据训练中的应用

由于人工神经网络（ANN）较传统的模式识别方法更具优越性,可以应用到油藏数据的训练中。采用3种神经网络训练函数,即,网络的权值和阈值训练函数、利用Levenberg-Marquardt规则训练函数和阶梯变化梯度算法,分别对油藏数据进行训练,得到成功的训练结果,从而把预测的油藏参数和三维（3D）地震属性联系起来。实验表明,把ANN用于训练油藏数据是可行的。     

裂解炉数学模拟新方法

1. Introduction Thermal cracking of hydrocarbons for olefinproduction is normally carried out in long reactor tubessuspended in a large gas fired furnace. Many complexfluid dynamic phenomena occur in the thermal crackingfurnace, including mass transfer,…     

Intelligent Prediction of Porosity and Permeability from Well Logs for An Iranian Fractured Carbonate Reservoir

Abstract

One of the most important processes in reservoir engineering is reservoir characterization, in which the reservoir parameters such as porosity and permeability are calculated. These parameters have crucial importance in reservoir engineering computations like reserve estimates and reservoir management. Estimation of porosity and permeability from conventional well logs for uncored well intervals is a good suggestion, but the complexity of the fractured carbonate reservoir makes the application of traditional statistical models totally unreliable. In this article the power of the pattern recognition of artificial neural networks (ANNs) has been applied to develop a transformation map from available most related well logs to rock petrophysical properties of Darquvain reservoir in the southwest of Iran. Comparison of the obtained results illustrates that ANN models can yield more reliable results with respect to traditional models of estimating petrophysical properties. An ANN can be utilized as a flexible and powerful tool for reservoir characterization from available well logs in development projects in the oil and gas industry.     

基于神经网络的海水淡化系统产水模拟及预测

海水淡化是一个涉及传热传质等诸多因素的复杂的非线性过程。为了克服传统数学建模方法在模拟仿真精度和实时性方面的不足,对大样本空间的神经网络构造及网络学习加速方法进行了大量探索及尝试,成功地将神经网络技术引入海水淡化产水过程中构造新的产水模型,建立了以空气入口干球温度、预冷器进口冷却水温度、海水喷淋温度和海水喷淋量作为输入参数的海水淡化系统神经网络模型。分析表明,该模型不仅具有较高的仿真精度,而且保持了系统原有的光滑性,能满足系统实时仿真模拟预测的要求。     

模糊神经网络推理系统定量计算储层参数

储层参数是地层评价的基础。自适应模糊神经网络推理系统是综合了模糊逻辑与人工神经网络两者优势的一种人工智能方法．该方法既可用于模式分类．又可进行连续计算。这是一种基于数据的建模方法,模糊隶属度函数及模糊规则是通过大量已知数据的学习得到的．而不是按经验或直觉信息给定的。这对于那些特性还不为人们所了解或者特性非常复杂的系统尤其重要．因此特别适于复杂储层评价中的定性解释与定量计算。研究结果表明,这种方法具有较高的计算精度。     

应用神经网络计算储层参数

以岩心分析资料及多种测井信息为依据，首先利用样本信息的神经元模型（ＣＵＳＩ）解决了储层参数的计算问题，并利用改进后的自适应神经元模型（ＡＣＵＳＩ）提高了分析精度。最后利用前馈神经网络的误差反向传播模型（ＢＰ）网络的外延和信息表达能力解决了非储层的定量识别。应用上述方法对辽河油田四口井进行了逐步参数分析，分析结果与实际情况吻合很好。     

缝洞型碳酸盐岩油藏耦合数值模拟新方法

首次提出洞穴内两相流体动力学计算方法,并建立了洞穴与溶孔间的界面处理方法,创建了缝洞型油藏洞穴-溶孔耦合型两相流数学模型,研究了有限体积数值解法,编写了计算机应用程序,成功地应用于塔河缝洞型油藏注水驱油机理研究,首次提出了多井缝洞单元的注水开发模式,现场实施后增油效果显著。     

The Design of New Soft Sensors Based on PCA and a Neural Network for Parameters Estimation of a Petroleum Reservoir

Abstract

A new set of soft sensors is presented based on principal component analysis (PCA) and artificial neural network (ANN) methodologies for parameters estimation of a petroleum reservoir. The crude diagrams of reservoir parameters provide valuable evaluation for petrophysical parameters. These parameters, however, are usually difficult to measure due to limitations on cost reliability considerations, inappropriate instrument maintenance, and sensor failures. PCA is utilized to develop new soft sensors to incorporate reliability and prediction capabilities of ANN. For this purpose, a PCA model is derived to reconstruct a parameter from other reservoir parameters using their redundancy relations. The developed soft sensors are applied to reconstruct parameters of Marun reservoir located in Ahwaz, Iran, by utilizing the available geophysical well log data. The experimental results demonstrate that the proposed hybrid PCA-NN algorithm is able to reveal a better performance than the PCA and the conventional back propagation–based NNs.     

三维地质建模与数值模拟技术在裂缝型有水气藏开发中的应用

以典型碳酸盐岩裂缝型有水气藏为例,阐述了裂缝型气藏三维地质建模与数值模拟一体化气藏评价技术。首先应用三维地质建模技术,利用地质、测井等测试资料,进行地层格架、断层、裂缝及相模型的建立,为气藏数值模拟提供了初始静态地质模型;其次,通过数值模拟技术,利用生产动态测试资料,通过历史拟合调整完善数值模型;最后,对不同采气速度、增压开采工艺、排水采气工艺等开发措施进行指标预测评价,最终确定合理的开发调整方案。生产实践表明,应用气藏一体化评价技术能实现不同开发阶段的协同管理,为复杂气藏调整开发方案提供了依据。     

一种确定低渗透油藏启动压力梯度的新方法

原油中的表面活性组分在多孔介质的孔道壁面上吸附形成了边界层,从而使得低渗透油藏的启动压力梯度远大于中高渗油藏的启动压力梯度。首先分析了边界层的形成及影响因素,然后建立一个孔隙网络模型来进行渗流模拟。通过在孔道中引入边界层,可以利用孔隙级网络模型作出压力梯度与渗流速度之间的关系曲线,即渗流曲线。由渗流曲线便可进一步确定启动压力梯度。该方法首次在确定启动压力梯度时直接考虑了边界层的影响,而且将孔隙级网络模型的应用进一步扩展到了低渗透油藏。     

Estimation of Petroleum Reservoir Parameters Using an Integrated Approach Neural Network,Principal Component Analysis and Fisher Discriminant Analysis

A new predictive methodology is introduced, based on a combined principal component analysis (PCA), Fisher discriminant analysis (FDA), and artificial neural network (ANN) methodologies for parameters estimation of a petroleum reservoir. Prediction of continuous petrophysical parameters is often time consuming and complicated because of geological variability such as facies changes due to sedimentary and structural changes. The petrophysical parameters, however, are usually difficult to measure due to reliability considerations, limitations insights on cost, inappropriate instrument maintenance, and sensor failures, evaluated by crude diagrams of reservoir parameters valuably. PCA and FDA provides an optimal lower dimensional representation in terms of discriminating among classes of data and are developed utilizing the reservoir historical data to incorporate reliability and prediction capabilities of ANN. The developed soft sensors are applied to predict the parameters of Marun reservoir located in Ahwaz, Iran, by utilizing the available geophysical well log data. The resulting outcomes demonstrate the promising capabilities of the proposed hybrid PCA-FDA-NN methodology than the conventional back-propagation NN, FDA, and PCA algorithms.     

复杂裂缝性油藏历史拟合中的特殊做法

克拉玛依油田八区 上二叠统下乌尔禾组油藏是一个具有边底水的特低渗透裂缝性巨厚砾岩油藏,在油藏数值模拟研究过程中,针对油藏的具体特点,对“非油层”出油问题、注入水漏失问题、水淹水窜井问题、地层水侵入问题以及不同水驱油模式问题采取了一些特殊做法,对400多口井、23年的生产历史进行了精细拟合,单井的压力、含水率拟合符合率达85%以上。获得了很好的拟合结果,为类似油藏的数值模拟提供了可供借鉴的经验,这些做法可推广应用于其它低渗透裂缝性油藏。     

利用神经网络建立储层宏观参数动态模型--以胜坨油田二区为例

建立描述和表征储层宏观参数随油田开发过程发生动态变化的物理和数学模型,可有效提高油田管理水平及最终采收率。以长期注水开发的胜坨油田二区沙河街组二段第8砂层组第3小层三角洲相储层为例,在研究了储层参数变化规律的基础上,利用神经网络建立了表征储层宏观参数变化的动态模型及数学表达式。该模型可有效地预测不同开发阶段储层宏观参数的变化过程和变化规律,为油田开发提供了科学依据。     

Dealing With the Challenges of the Rate-Dependent Skin Phenomenon in a Gas Condensate Reservoir: A Simulation Approach

Abstract

Skin is one of the most challenging problems in oil and gas production, resulting in a reduction in reservoir deliverability. Mechanical skin is introduced to the pay zone during the drilling and completion phase, while rate-dependent skin comes into play as production starts, and is caused by infracting basic Darcy assumptions in a gas reservoir. Dissimilar to conventional gas reservoirs, the formation of condensate jots in a gas condensate reservoir when the bottomhole pressure drops below the dew point causes higher skin, which imposes a greater pressure drop. The authors investigated dependency of skin to flow rate in gas condensate reservoirs through a simulation approach. A radial synthetic reservoir compositional model was developed by using rock and fluid properties and well data based on information from a real gas condensate field in south Iran. A well test analysis was utilized to verify the validity and accuracy of the model. Due to heterogeneity existence in particular field, the authors tried to consider different reservoir and wellbore characteristics in reservoir simulation models. A wide range of rock permeabilities and mechanical skins were examined in different irreducible water saturations. Moreover, all the scenarios were run in two different initial reservoir pressures individually. The result of numerous simulations showed a similarity between the trends of rate-dependent skin versus flow rate for the entire tested model. It was also illustrated that in low-permeable gas condensate reservoirs, the existence of mechanical skin leads to inefficient production. Because mitigation of some kind of mechanical skin types are not efficient, the same simulation study should be performed before starting the drilling operation in a gas condensate reservoir to assess the effect of mechanical skin.     

一种开发无隔板底水油藏的新方法

在无隔板底水油藏开发过程中,许多油井面临着底水锥进速度快、临界产量低、无水采油期短等生产技术难题。借鉴前人研究思路,基于开发底水油藏的隔板理论和双层完井排液压锥技术,对无(天然)隔板底水油藏的开发模式进行了探讨。提出了一种人为注入堵剂(人工隔板)与排水压锥技术相结合的有效开发无隔板底水油藏的新方法,对于科学、高效、合理开发底水油藏具有一定的现实意义。     

A New Approach for Predrilling the Unconfined Rock Compressive Strength Prediction

Abstract

A reasonable knowledge of rock's physical and mechanical properties could save the cost of drilling and production of a reservoir to a large extent by selection of proper operating parameters. In addition, a master development plan (MDP) for each oilfield may contain many enhanced oil recovery procedures that take advantage of rock mechanical data and principles. Thus, an integrated rock mechanical study can be considered an investment in field development.

The unconfined compressive strength (UCS) of rocks is the important rock mechanical parameter and plays a crucial role when drilling an oil or gas well. A drilling operation is an interaction between the rock and the bit and the rock will fail when the resultant stress is greater than the rock strength. UCS is actually the stress level at which rock is broken down when it is under a uniaxial stress. It can be used for bit selection, real-time wellbore stability analysis, estimation an optimized time for pulling up the bit, design of enhanced oil recovery (EOR) procedures, and reservoir subsidence studies.

Rock strength can be estimated along a drilled wellbore using different approaches, including laboratory tests, core–log relationships, and penetration model approaches. Although this rock strength profile can be used for future investigation of formations around the wellbore, they are actually dead information. Dead rock strength data may not be useful for designing a well in a blind location (infill drilling). Rock strength should be predicted prior to drilling operations. These sort of data are helpful in proposing a drilling program for a new well.

In this research, new equations for estimation of rock strength in Ahwaz oilfield are formulated based on statistical analysis. Then, they are utilized for estimation of the rock strength profile of 36 wells in a Middle Eastern oilfield. An artificial neural network is then utilized for prediction of UCS in any predefined well trajectory. Cross-validation tests showed that the results of the network were compatible with reality. This approach has proven to be useful for estimation of any designed well trajectory prior to drilling.     

Black Oil and Compositional Reservoir Simulation for Increasing the Recovery Performance of an Iranian Fractured Carbonate Reservoir

Abstract

Naturally fractured reservoirs contain a significant amount of world oil reserves. Accurate and efficient reservoir simulation of naturally fractured reservoirs is one of the most important, challenging, and computationally intensive problems in reservoir engineering. Black oil and compositional reservoir simulators have been used to determine the reservoir management and production strategies to increase the oil recovery from a low-porosity, low-permeability fractured carbonate reservoir, with an average matrix permeability of 0.8 md, average fracture permeability of 500 md, and an average matrix porosity of 10%. This reservoir is a candidate for an enhanced oil recovery (EOR) process, because the reservoir production rate has been declined due to increasing the water cut as a result of rising the water oil contact. The injection techniques that have been considered in this study for black oil model include (a) gas injection, (b) water injection, and (c) simultaneous water alternating gas injection and for the compositional model include (a) dry gas injection, (b) CO₂ injection, and (c) N₂ injection. Simulation results show that CO₂ injection has the maximum oil recovery between the EOR scenarios.