An evolutionary approach is applied to solve the nonlinear well logging inverse problem. In the framework of the proposed interval inversion method, nuclear, sonic, and laterolog resistivity data measured at an arbitrary depth interval are jointly inverted, where the depth variation of porosity, water saturation, and shale volume is expanded into series using Legendre polynomials as basis functions. In the interval inversion procedure, the series expansion coefficients are estimated by using an adaptive float-encoded genetic algorithm. Since the solution of the inverse problem using traditional linear optimization tools highly depends on the selection of the initial model, a heuristic search is necessary to reduce the initial model dependence of the interval inversion procedure. The genetic inversion strategy used in interval inversion seeks the global extreme of the objective function and provides an estimate of the vertical distribution of petrophysical parameters, even starting the inversion procedure from extremely high distances from the optimum. For a faster computational process, after a couple of thousand generations, the genetic algorithm is replaced by some linear optimization steps. The added advantage of using the Marquardt algorithm is the possibility to characterize the accuracy of the series expansion coefficients and derived petrophysical properties. A Hungarian oil field example demonstrates the feasibility and stability of the improved interval inversion method. As a significance, the genetic inversion method does not require prior knowledge or strong restrictions on the values of petrophysical properties and gives highly reliable estimation results practically independent of the initial model and core information.
相似文献 |
|||||||
|
|||||||
|
共查询到20条相似文献，搜索用时 125 毫秒
1.
2.
3.
基于粒子群算法与混合罚函数法的有限元优化反演模型及应用 总被引：3，自引：0，他引：3
岩土工程优化反分析是一个典型的复杂非线性函数优化问题,采用全局优化算法是解决这个问题的理想途径。针对常规反演方法应用于岩土工程参数反演时搜索效率低的缺点,结合粒子群算法和遗传算法的特点,充分考虑二者的互补性,提出一种效率较高的全局优化算法,以测点的实测值与计算值建立一种新的评价函数,将多目标优化问题转化为单目标优化问题,用混合罚函数法将约束问题变为无约束问题,构建了一种新的目标函数,将有限元程序ABAQUS作为一个模块嵌入到优化算法程序中,编制了有限元优化反演分析程序。并给出了应用实例验证了该法的有效性和实用性,是一种可行的参数反演方法,可应用于实际工程中复杂岩土介质初始应力场反演、渗流场以及位移反分析 相似文献
4.
针对地下水模拟-优化模型约束优化的特点,本文结合最小代数代沟模型和Pareto强度指标概念,引入一种求解地下水模拟-优化模型的新型实数编码遗传算法,该算法将罚函数法求解约束优化问题的目标函数和违反约束条件的程度函数的权组合方式改为向量组合形式,从而将约束优化问题转化为两目标优化问题进行求解。通过经典地下水算例与其他优化方法的比较分析表明了新算法的可靠性、通用性和稳健性。 相似文献
5.
6.
双重贝寒尔函数积分的数值计算 总被引：2，自引：1，他引：2
双重贝塞尔函数积分则由于贝塞尔函数互乘项的强振荡与慢衰减特性而以以应用通常的数值积分算法。本文将被积区间[0,∞]划分为[0，λ0]、[λ0，∞]两部分，应用贝塞尔函数的克尔函数表述式及后者的大宗量渐近特性，区间[λ0，∞]的双重贝寒尔函数积分可被转化为Fourier正（余）弦变换，并可利用各种快速算法对其进行数值计算；区间[0，λ0]上的双重贝塞尔函数积分的计算可直接应用一般的数值积分算法并能获得较高的计算精度；当需大量计算有共同参量的双重贝塞 尔函数积分时，其计算效率仍显不足。此时，可应用贝塞尔函数的导数关系式对[0,λ0]内的双重贝塞尔函数积分进行恒等变换，再用差商近似导数，将其转化为对贝塞尔函数本身的积分，而该积分又仅需计算一次。故本算法对双重贝塞尔函数积分的计算效率有明显提高。 相似文献
7.
工程中的不确定性问题可以用区间理论、随机理论或模糊理论进行求解。采用区间分析方法来处理地下隧洞结构开挖支护过程中的不确定性问题时。将结构系统中的不确定性参数用区间数来表示,用有限元方法建立系统的控制方程。该控制方程是一个包含不确定性几何参数的复杂的非线性区间方程组。讨论了两种优化计算模型,分别将方程组中的所有区间数作为设计变量,区间量的变化区间作为相应的设计变量的边界约束,运用遗传模拟退火算法求出位移和应力等响应量的最大值和最小值。通过工程算例验证了文中方法的合理性及可行性,并与端点组合方法的结果进行了比较分析。 相似文献
8.
《物探化探计算技术》2020,(3)
探地雷达反演问题是高度非线性的,采用线性反演方法往往难以获得较好的反演效果,因此提出了将生物地理学优化算法同粒子群优化算法相结合的混合非线性反演方法。将该方法用于探地雷达时间域波形反演,采用时间域有限差分方法进行正演,以信号的均方误差函数作为目标函数,并针对波形反演的特点,在目标函数中加入波形的导数拟合差作为约束项,实现了结构层厚度和介电常数的波形反演。对比经典粒子群算法和生物地理学优化算法在多层介质仿真数据的一维波形反演中的效果,验证了该改进算法的有效性和抗噪性。 相似文献
9.
10.
确定河流纵向离散系数的快速SA法 总被引：4，自引：0，他引：4
将模拟退火算法应用求解分析河流水团示踪试验数据,确定河流纵向离散系数的函数优化问题.针对一般模拟退火算法收敛性能差的缺陷,采取了改变模型扰动、接受概率、降温方式以及增加附加约束条件和增加记忆功能等综合措施对算法进行了改进.实例分析结果表明,改进后的模拟退火算法确定河流纵向离散系数的优化问题不仅可行,而且求解精度较高,更重要的是收敛速度比其它算法快的多,这在一些高维复杂的优化问题中很值得推广. 相似文献
11.
遗传算法在磁异常反演中的应用 总被引：4，自引：0，他引：4
遗传算法是一种约束随机优化方法。它与传统的确定性优化方法相比较,具有全局收敛,不计算目标函数偏导数等优点；与传统的随机优化方法相比较,具有搜索效率高以及隐含并行计算等优点。本文首先讨论了遗传算法的基本原理和迭加机算步骤,其次讨论了该方法在磁化强度约束反演问题中的应用效果。 相似文献
12.
13.
14.
悬链线剖面是大位移井轨道的经典类型,在进行设计时需要求解一个以悬链线初始井斜角为未知数的非线性方程。由于未知数包含在多个三角函数和对数函数中,计算工作量较大,而且常用的迭代求解方法存在一些问题。通过数学变换将该方程转换成一个只包含对数函数和多项式函数的新方程,对新方程的函数性态做了几何分析,进而提出了寻找求解区间的步长搜索算法和自适应步长搜索算法。利用二分法在求解区间上能够快速求出新方程的数值解。利用大位移井设计实例验证了本文算法的有效性,并对圆弧井段井眼曲率与方程解的关系进行了讨论。本文提出的算法可用于大位移井轨道设计的计算机软件开发中。 相似文献
15.
Within the framework of limiting equilibrium approach, the problem of active earth pressure on rigid retaining wall is formulated in terms of the calculus of variations by means of Lagrange multipliers. It is transcribed as the functional of extreme‐value problem by two undetermined function arguments, and is further transformed into determining the minimax solution of restrained functions incorporating the geometrical relations of the problem. The function of (fmincon) in the optimization toolbox of MATLAB 6.1 can be used to find the minimax solution. Computation results show there exist two kinds of modes of failure sliding along plane surface and rotating around log‐spiral cylinder surface when the soil behind the walls reaches the critical active state. The magnitude of active earth pressure in the case of translational mode is less than that in the case of rotational mode. The location of action point of earth pressure in the case of translational mode is at or below $\frac{1}{3}$ height of the wall, and in the case of rotational mode, is above $\frac{1}{3}$ height of the wall. Preliminary study indicates a pair of numbers by two theoretical modes can be regarded as an interval estimation of active pressure. Copyright © 2009 John Wiley & Sons, Ltd. 相似文献
16.
Ramin Mohammad Hosseini Dokht Hamid Reza Ramazi Mohammad Talebi Kenari 《Arabian Journal of Geosciences》2013,6(6):1965-1975
The crucial advantages of a 3-D seismic survey are the proper migration of the reflection points and the more accurate study on the structural and stratigraphic targets, reservoir characterization, and joint study. In this article we will focus on Pardis Project in southwest of Iran where no 3-D seismic survey has ever been carried out. The local geological information, previous 2-D seismic, VSP data, and interval velocity information obtained from check shots were taken into consideration to determine the requirements of the survey. The objective is to adequately sample the primary and secondary targets at the depth of 1,200 and 5,000 m, respectively. On the base of the logic dominating a genetic algorithm, the best operational layout was offered to satisfy the geophysical requirements looking forward to satisfy financial constraints. Using Genetic Algorithm Toolbox in MATLAB, we could formulate a mathematical constrained optimization problem. Applying this technique we derived nominal designs which are needed to be evaluated to make sure how well they could image the targets. 相似文献
17.
Updating Markov chain models using the ensemble Kalman filter 总被引：1，自引：0，他引：1
18.
In this paper, a numerical model to predict flow‐induced shear failure along pre‐existing fractures is presented. The framework is based on a discrete fracture representation embedded in a continuum describing the damaged matrix. A finite volume method is used to compute both flow and mechanical equilibrium, whereas specifically tailored basis functions are used to account for the physics at discontinuities. The failure criterion is based on a maximum shear strength limit, which changes with varying compressive stress on the fracture manifold. The displacements along fracture manifolds are obtained such that force balance is achieved under conditions, where shear stress of the failing fracture segment is constrained to the maximum shear strength at the segment. Simultaneously, the fluid pressure is computed independently of the shear slip. A relaxation model approach is used to obtain the maximum shear limit on the fracture manifold, which leads to grid convergence. 相似文献
19.
根据岩体中不同区域裂隙发育规模的差异分为连续区域和离散区域,在不同区域中使用不同的地下水运动数学模型,应用区域分解算法来解决这类问题。其中连续区域采用了等效连续介质模型,离散区域采用了随机裂隙网络模型,通过区域公共边界上水位和流量连续的条件将两模型耦合求解。将基于区域分解算法的耦合模型应用于锦屏水电站坝址区三维渗流场的模拟中,通过钻孔观测水位和计算水位的对比发现,该方法是有效的,能够应用于实际工程。 相似文献
20.
In a statistical pattern recognition context, discriminant analysis is designed to classify, when possible, objects into predefined categories. Because this method requires precise input data, uncertainties cannot be propagated in the classifying process. In real case studies, this could lead to drastic misinterpretations of objects. A new nonparametric algorithm based on interval arithmetic has thus been developed to propagate interval-form data. They consist in calculating interval conditional probability density functions and interval posterior probabilities. Objects are eventually assigned to a subset of classes, consistent with the data and their uncertainties. The classifying model is thus less precise, but more realistic than the standard one, which we prove on a real case study. 相似文献
|
| 设为首页 | 免责声明 | 关于勤云 | 加入收藏 |
|
Copyright©北京勤云科技发展有限公司 京ICP备09084417号 |