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。     

Surrogate-based Optimization of Horizontal Well Placement in a Mature Oil Reservoir

Abstract

Exploitation of mature oil reservoirs has forced researchers to find ways for enhancing production from such reservoirs. An effective way to do so is drilling horizontal wells. There are many advantages for these kinds of wells but they require extra investment for drilling and completion. The objective of this research is to choose location, direction, and length of a new horizontal well to maximize accumulative oil production over a planning horizon. The authors employ various metamodeling techniques in order to build a good approximate model for estimation of accumulated outflow from each well scenario. This leads to a significant reduction in the computation time. The authors built and compared 18 metamodels based on quadratic, multiplicative, and radial basis functions for the reservoir under study. Then, the best model was introduced to a global optimization search using a Genetic algorithm. Numerical results showed that 3 million barrels of more oil could be produced from the designated well as compared with those of the initial scenarios, which is a significant improvement.     

Predicting Dew Point Pressure: Using a Hybrid Intelligent Network

This article presents a new model in predicting dew point pressure, which is reduced the complexities of the prior fuzzy model by implementing genetic algorithm as a feature recognition tool. Feature recognition is a tool that makes the problem less complex and create the opportunity to have a better model. In this study, 15 parameters decreased to six by using the feature recognition tool. This tool discovered if fuzzy model inputs include reservoir temperature and mole fractions of H2S, N2, C3, iC5, and C6, the proposed model captures the physical trend better than previous models. The recommended model has an average relative deviation of 1.16% and average absolute deviation of 3.08% for testing data points.     

非机理模型优化问题的NN-GA方法研究

针对化工过程系统中难于建立机理模型的过程或设备,提出了集成运用神经网络和遗传算法的优化方法（ＮＮ－ＧＡ方法）,并介绍了对神经网络训练算法的有效改进以及ＮＮ－ＧＡ方法的实施过程,给出了应用实例,分析了ＮＮ－ＧＡ方法与统计回归分析相比所具有的独特优势。     

在测井中用一种组合进化神经网络识别油水层

分析了基于进化算法的神经网络,指出基于遗传算法的神经网络具有强的全局搜索能力,基于进化规划的神经网络具有强的局部寻优能力。在此基础上,提出了综合上述两种神经网络优点的组合进化神经网络,并应用于油水层测井解释中,降低了误判率。     

斜井多井系统中一口井压力降落曲线试井分析

本文提出一套斜井多井系统中单井测试曲线的试井分析方法,其中包括任意倾斜角度、任意变化产量、任意排列形式的任意多口斜井的多井系统压力响应的算法和适合于斜井多井系统多解特征的求解试井参数的非线性规划算法。两种算法都通过算例得到了验证该计算方法简单明了,计算率和计算精度可靠。     

基于虚拟样机技术的海洋修井动力猫道结构优化

为解决海洋修井动力猫道设备整体工作效率低、自重过大等问题,结合虚拟样机技术,进行海洋修井动力猫道机械结构运动学动力学仿真分析,在仿真基础上利用遗传优化算法,选取各不同组件可进行优化的结构尺寸作为优化变量,分别以最大处理载荷和最快处理速度为目标对海洋修井动力猫道实现结构优化。结果表明：在以最大处理载荷为目标的结构优化中,安全因数均大于1.5的情况下各组件减重12.5%~33.0%;在以最快处理速度为目标的结构优化中,安全因数均大于1.5的情况下各组件减重16.4%~38.0%。虚拟样机技术分析方法可为海洋平台其他设备结构轻量化设计提供参考。     

The Use of Optimization Procedures to Estimate Minimum Miscibility Pressure

An important factor in the design of gas injection projects is the minimum miscibility pressure (MMP). A new genetic algorithm (GA)–based correlation and two neural network models (one of them is trained by back propagation [BP] algorithm and another is trained by particle swarm optimization algorithm) have been developed to estimate the CO₂–oil MMP. The correlation and models use the following key input parameters: reservoir temperature, molecular weight of C⁺ ₅, and mole percentage of the volatiles and intermediate components (for the first time, the mole percentages are used as independent variables). Then results are validated against experimental data and finally compared with commonly used correlations reported in the literature. The results show that the neural network model trained by BP algorithm and the correlation that has been developed by GA can be applied effectively and afford high accuracy and dependability for MMP forecasting.     

Multiobjective Optimization of Reactor-Regenerator System Using Ant Algorithm

This article focuses on the development of a multiobjective optimization algorithm for a tubular reactor-regenerator system with a moving deactivating catalyst. The task is to find the optimal temperature profile along the tubular reactor, catalyst recycle ratio, and the regeneration capacity for maximizing the process profit flux, selectivity, and conversion. A new heuristic technique, viz, ant colony optimization method has been employed to obtain the Pareto optimal set of solutions.     

基于模糊神经网络的非线性系统建模方法

针对工业中难以建立精确数学模型的复杂过程，提出了一种基于模糊神经网络建模的方法。模糊神经网络隐层节点数随输入改变而改变，采用三角形隶属函数，同时给出了模糊规则中参数的给定方法，利用误差反传法修正参数。仿真实验验证了该建模方法的有效性。     

特高含水期水驱砂岩油田井网部署自动优化方法

注水油田开发设计的重要环节是井网设计,在布井时应充分考虑储层的非均质性与各向异性以及注入水在地下的驱替动态,布置与其相匹配的井网,即矢量井网。油田是一个复杂的系统,井网调整时井组间相互制约,为了确定最优的井网部署方案,基于数学建模和最优化理论,建立了井网部署优化模型,以各注采连线上见水时间相同作为目标函数,辅以适当的约束条件,优化各井的位置。井网部署优化模型属于大规模、多变量的复杂优化问题,采用遗传算法进行求解,速度快、效率高。文中推导并建立了考虑非均质和各向异性非活塞式驱替见水时间计算方法,相比传统的通过耦合数值模拟程序求解开发指标方法,效率更高;建立了包含25口井的概念模型,验证了该方法的有效性,优化后可实现井网与储层非均质与各向异性的最大匹配。     

Porosity Estimation of a Reservoir Using Geophysical Well Logs and an Interval Type-2 Fuzzy Logic System

The authors propose a novel interval type-2 fuzzy logic structure that uses the advantages of type-2 fuzzy sets to estimate the formation porosity of the reservoir using geophysical well logs. The proposed system is constructed on the base of a set of fuzzy rules that includes type-2 fuzzy sets in the antecedent part and type-1 interval representing the centroid of a type-2 fuzzy interval in the consequent part of the rules. For structure identification, a fuzzy clustering algorithm is implemented to generate the rules automatically and for parameter adjustment the back propagation algorithm is used. To demonstrate the superiority of the proposed model, the results of the proposed type-2 fuzzy logic structure are compared with its type-1 counterpart.     

基于蒙特卡罗遗传算法的换热网络优化问题

在换热网络超结构及其数学模型的基础上,提出了换热网络优化的蒙特卡罗遗传混合算法,利用蒙特卡罗方法在解空间进行全局搜索,得到最佳换热匹配,由此引入遗传算法对网络优化问题中的连续性变量进一步优化,降低换热网络年综合费用。实例表明,应用蒙特卡罗遗传混合策略能在保证算法的全局搜索能力的前提下,提高换热网络优化效率,并能使换热匹配更加合理,减少加热器和冷却器的投入,降低网络的综合费用。     

基于改进的遗传算法的天然气管网系统运行优化

在天然气管网系统安全稳定运行的基础上,为了实现节能减排,充分合理地利用管道的输配能力,将目标函数定义为天然气的最大流量,同时考虑管道内天然气稳定流动、各节点流量平衡、节点及管段压力等约束条件,建立了天然气管网系统优化数学模型。采用整数编码来进行管径编码,用模拟退火罚函数转化约束条件,并合理地将遗传算法的全局寻优能力和模拟退火的局部搜索能力互补融合起来,实现算法的改进和优化。将改进的遗传算法应用到某大型天然气管网优化设计的实例中,计算结果表明,改进的遗传算法在解的质量上和收敛的速度上都优于基本遗传算法,验证了所建立的优化模型是高效可行的。     

低渗透油田有杆抽油系统模糊优化设计

低渗透油田地质情况复杂、开采难度大、成本较高,具有不同于常规油田的抽油系统特征,深入研究有杆抽油系统优化设计将产生明显的经济效益。有杆抽油系统模糊优化设计依据低渗透油田特点,以油层供液能力和采油设备协调为基础,采用优化的隶属函数,建立实用的权重矩阵,基于模糊评判方法进行方案优选,确定系统抽汲参数与各种机械设备类型和规格,预测相应抽汲参数下的工况指标。现场实例设计结果表明,模糊优化设计与油田实际设计原则一致,其结果准确可靠、简明直观。模糊优化设计可以提供多种方案,能够详细描述多种方案优劣程度,便于参考对比,灵活选择。最佳方案既可以按照综合评判结果直接确定,也可以综合对比多种方案,进一步进行人工优选。研究结果能够为新型采油设备的研发提供重要的参考数据。     

A Hybrid Symbiotic Evolution–Neuro Fuzzy Strategy for Permeability Estimation of the Reservoir

A key parameter for reservoir characterization is permeability distribution. Most well log data and core permeability values are corrupted by noise (such as uncertain depth-matching, core testing conditions, and thin beddings). In this work, the authors first used wavelet as a new powerful tool for de-noising data points and then they investigated how the integration of back propagation with group based symbiotic evolution improves the reliability and prediction capability of neuro-fuzzy systems for predicting permeability of real reservoir data.     

Optimization of Gas Allocation to a Group of Wells in Gas Lift in One of the Iranian Oil Fields Using an Efficient Hybrid Genetic Algorithm (HGA)

A hybrid genetic algorithm (HGA) was introduced to allocate optimum amount of gas. This method was applied to a group of wells in gas lift in the case of availability limited amount of gas. For testing the ability of the proposed HGA, the results of this work with those of previous works in a field with six wells were compared. Then for an Iranian southern west oil field with nine wells, gas allocation is performed using different amount of available gas. The results show that the introduced method (HGA) is very efficient tool in gas allocation issue.     

Permeability Estimation of the Reservoir Based on Particle Swarm Optimization Coupled with Artificial Neural Networks

Abstract

In this work we investigate how the integration of back-propagation (BP) with particle swarm optimization (PSO) improves the reliability and prediction capability of PSO. This strategy is applied to predict permeability in Mansuri Bangestan reservoir located in Ahwaz, Iran, utilizing available geophysical well log data. Our methodology utilizes a hybrid PSO–BP. The particle swarm optimization algorithm was shown to converge rapidly during the initial stages of a global search, but around global optimum, the search process will become very slow. On the contrary, the gradient descending method can achieve faster convergence speed around global optimum and with greater accuracy. The proposed algorithm combines the local search ability of the gradient-based BP strategy with the global search ability of particle swarm optimization. PSO is used to decide the initial weights of the gradient decent methods so that all of the initial weights can be searched intelligently. The experimental results show that the proposed hybrid PSO–BP algorithm is better than the PSO algorithm in convergence speed and accuracy.     

鄂尔多斯盆地碳酸盐岩储层测井产能预测研究

讨论了以模糊理论为基础的合试层产能分权法。由SOM自组织映射与TFI模糊聚类协同确定分类数及变量组合，以隶属度分权、动态地剖分合试层产能为预测单层样本数据体，以遗传神经网络建立预测模型来预测储层自然产能，形成独特、灵活的泛合试层测井产能预测方法。经对鄂尔多斯盆地中部气田16口井52个测试层的应用，取得较好的应用效果。     

油气集输系统新增产能建设拓扑优化方法研究

在油气集输新增产能建设中,为充分利用已建管网系统的处理能力,达到降低投资,提高经济效益的目的,进行了油气集输新增产能建设布局优化方法的研究。针对新建集输管网为树状结构的情况,建立了油气集输系统新增产能建设拓扑优化的数学模型,讨论了计算复杂性问题。根据模型的特点,将问题分解为拓扑级优化和几何级优化两个子问题,并通过它们之间的迭代进行求解。实例计算表明,所建立的数学模型正确,求解方法可行,能用于工程实际设计。