共查询到19条相似文献，搜索用时 125 毫秒
1.
径向基神经网络（RBFNN）具有结构简单、学习速度快、不易陷入局部极小等优点,能够有效地提高电阻率层析成像反演的收敛速度和求解质量.本文针对电阻率层析成像反演的非线性特征,提出了一种基于汉南-奎因信息准则（HQC）的正交最小二乘法（OLS）学习算法（HQOLS）.该算法通过计算HQC的最优值来自动选择RBFNN的网络结构,避免了传统OLS学习算法中阈值参数的设定,保证了网络的泛化性能.通过比较聚类法、梯度法、OLS和HQOLS等学习算法的反演性能,构建了基于RBFNN的电阻率层析成像反演模型.数值仿真和模型反演的结果表明,该方法实现简单,在准确性上优于BP反演,成像质量优于传统最小二乘法反演. 相似文献
2.
3.
在研究大地电磁响应函数频散关系的基础上,构制了一套滤波系数算法,以用于由一套视电阻率资料估算相应的阻抗相位.理论模型和实际大地电磁观测资料的数字试验表明,该法是行之有效的.由频散关系估算的相位值与观测的相位资料的比较,可用于检验观测资料是否满足频散关系.利用经频散关系校正的阻抗相位值,进行大地电磁阻抗的联合反演则可望获得更为可靠的结果.研究了观测频带相互衔接的电偶源频率电磁测深和大地电磁测深视电阻率的一维联合反演问题.对两个实测点两种电磁法的观测资料进行了联合反演试验,与钻井资料对比表明,所获得的电性分层参数是较为可信的.在补充了由频散关系获取的电偶源频率电磁相位资料后,对于两种电磁法的视电阻率————阻抗相位、阻抗实部视电阻率-阻抗虚部视电阻率进行了拟大地电磁反演,获得了相近的反演结果. 相似文献
4.
时间域航空电磁中心回线(或重叠回线)装置晚期道数据受激电效应影响常出现符号反转现象.这类数据与多个激电参数相关，并且各参数之间灵敏度差异较大，导致反演存在严重的非唯一性.本文提出一种基于Pearson相关性约束和深度学习算法相结合的时间域航空电磁激发极化参数反演策略.该反演策略首先基于深度学习预测时间域航空电磁激电参数，进而给时间常数和频率相关系数一个较小的约束范围后再反演电阻率和极化率，由此大大减少反演的多解性.针对电阻率和极化率的反演，我们采用统计学中Pearson相关系数构建两种物性参数的相关性约束，进一步减少反演多解性.为验证反演策略的有效性，我们对双棱柱模型和拱形模型分别进行反演试算.理论测试结果表明，基于Pearson相关性约束的电阻率和极化率的反演结果比传统的高斯-牛顿反演结果更接近真实模型，而基于深度学习预测时间常数和频率相关系数后的电阻率和极化率反演结果与给定真实时间常数和频率相关系数后的反演结果效果相当.最后，我们对来自澳大利亚的带激电效应的航空电磁实测数据在考虑和不考虑激电效应条件下进行反演，结果表明考虑激电效应的反演无论数据拟合还是地电断面的连续性均得到明显改善... 相似文献
5.
可控源音频大地电磁法在资源勘探等领域中发挥着重要的作用.我们把有限差分数值模拟方法用于可控源音频大地电磁三维正演,结合正则化反演方案和共轭梯度反演的思路,将反演中的雅可比矩阵计算问题转为求解两次"拟正演"问题,得到模型参数的更新步长,形成反演迭代,实现了可控源音频大地电磁三维共轭梯度反演算法.该反演算法可用于对有限长度电偶源激发下采集到的可控源音频大地电磁全区(近区、过渡区和远区)视电阻率和相位资料进行三维反演定量解释,获得地下三维模型的电阻率结构.理论模型合成数据的反演算例验证了所实现的可控源音频大地电磁三维共轭梯度反演算法的有效性和稳定性. 相似文献
6.
根据电磁波的传播特性，把视电阻率随周期变化的曲线转化为电阻率随深度变化的曲线，并以此作为初始反演的地电模型．通过初始地电模型得到的视电阻率曲线与真实模型的视电阻率曲线的对比，对初始地电模型的电阻率值进行校正．校正后的反演模型的视电阻率曲线与真实模型的视电阻曲线的拟合程度有所提高．然后如此反复进行多次校正，获得与真实模型更为接近的反演模型，反演的拟合误差一般小于ｌ％．模型试验和实际例子表明，该方法的拟合程度优于Ｂｏｓｔｉｃｋ法． 相似文献
7.
为推进大地电磁三维反演的实用化，本文实现了基于L-BFGS算法的带地形大地电磁三维反演.首先推导了大地电磁法三维反演的Tikhonov正则化目标函数以及Hessian矩阵逆矩阵近似表达式和计算方法，然后设计了一种既能保证空气电阻率固定不变又能保证模型平滑约束的协方差矩阵统一表达式，解决带地形反演问题.在反演算法中采用正则化因子冷却法以及基于Wolf条件的步长搜索策略，提升了反演的稳定性.利用开发的算法对多个带地形地电模型（山峰地形下的单个异常模型、峰-谷地形下的棋盘模型）的合成数据进行了三维反演，并与已有大地电磁三维反演程序（ModEM）进行对比，验证了本文开发的三维反演算法的正确性和可靠性.最后，利用该算法反演了华南某山区大地电磁实测数据，得到该区三维电性结构，揭示了研究区以高阻介质为基底，中间以低阻不整合面和相对低阻介质连续分布，浅部覆盖高阻介质的电性结构特征，进一步验证了本文算法的实用性. 相似文献
8.
应用地震资料约束下大地电磁资料反演古潜山或基底内部结构的新方法. 首先利用地震及井资料,反演得到古潜山或基底之上各层的厚度及电阻率,建立了古潜山或基底以上的地电模型;并利用大地电磁一维反演资料,建立了古潜山或基底及其以下的地电模型. 在初始模型建立的基础上,应用高效快速的二维大地电磁正演算法——模式匹配方法,进行正演计算. 利用共扼梯度迭代方法,进行正反演迭代计算. 通过对胜利油田南北618剖面反演, 基底以下在地震资料没有显示的构造信息,在反演结果中得到了较好地反映. 相似文献
9.
10.
为满足地球物理资料反演解释的高精度、快速、稳定的要求,本文结合免疫遗传算法寻优速度快和BP神经网络反演不依赖初始模型等优点,设计了一种将BP神经网络和免疫遗传算法进行有机结合的全局优化反演策略,并将该策略成功地应用于二维高密度电法数据反演.利用免疫遗传算法(Immune Genetic Algorithm,简称IGA)对神经网络的反演参数进行同步优化,提高了电阻率反演的精度.仿真和实验结果验证设计的全局优化反演策略取得了较好的效果,通过与线性反演方法和BP法以及遗传神经网络法等反演方法进行比较,得出该方法具有反演精度更高,反演时间更短等显著优势的结论. 相似文献
11.
大地电磁的反演问题是非线性,如果采用线性反演方法容易陷入局部极小,使得反演结果非唯一性严重.本文将人工鱼群算法引入到地球物理反演之中,提出了非线性的大地电磁人工鱼群最优化反演.该方法不需要进行偏导数的求取,可以对反演的范围进行约束,以减小反演结果的非唯一性.同时我们对搜索步长进行了改进,给出适用于大地电磁反演的人工鱼群参数.大量的理论数据试算表明,人工鱼群反演算法能够较好地寻找到全局最优解.实测数据的处理结果表明,该方法可以用来处理实际资料,并且能够取得很好的应用效果. 相似文献
12.
On the basis of the dispersion relation of magnetotelluric response functions (MTRF), a filter coefficient algorithm has been
made, with which the corresponding impedance phase data can be estimated using a set of apparent resistivity data. The tests
of theoretical models and observed magnetotelluric (MT) data show that this algorithm is effective. Comparing the impedance
phase estimated using dispersion relation with the observed phase, it can be checked whether the dispersion relation between
the observed apparent resistivities and phase data was satisfied. The use of phase data corrected using the dispersion relation
in the joint inversion for MT impedance is advantageous to obtain more reliable inversion results. The problems on the one-dimensional
joint inversion for the (MT) apparent resistivity and the apparent resistivity of the frequency electromagnetic sounding (FEMS)
with horizontal electric dipole, whose observed frequency bands are linked up each other, are studied. The observed data of
two kinds of electromagnetic (EM) methods at two sites are used to inverse, the comparison with the drilling data show the
results are more reliable. To supply the phase data of FEMS using the dispersion relation, for the apparent resistivity-phase
data and impedance real part-imaginary part apparent resistivities of two kinds of EM methods the imitated MT joint inversions
are made, and more similar results also are obtained.
The Chinese version of this paper appeared in the Chinese edition ofActa Seismologica Sinica,15, 91–96, 1993.
The projects sponsored by the Chinese Joint Seismological Science Foundation. 相似文献
13.
14.
THE DISPERSION RELATION, OF EARTH''S ELECTROMAGNETIC RESPONSE FUNCTIONS AND THE JOINT INVERSION OF DATA 
下载免费PDF全文
下载免费PDF全文 On the basis of the dispersion relations of MT field, the necessity and applied prospects of the joint inversions using a pair of MT response functions which are correlative with the dispersion relations, are infered. A filter coefficient algorithm is made, with which the corresponding impedance phase data can be estimated using a set of apparent resistivities. The tests for the observed MT data show that when comparing the impedance phase estimated using the dispersion relation with the ob served phase, it can be checked whether the dispersion relation between observed apparent resistivity and phase data is satisfied or not, and that the use of the phase data corrected using the dispersion relation in the joint inversion is advantageous to obtain more confident results. It is shown that joint inversions are more advantageous than single parameter inversions, and that in the most case the joint inversion using the apparent resistivities of impedance real and imaginary parts is more advantageous than the jointinversion using the normal apparent resistivity and impedance phase. The existence of the dipersion relations between the ratio apparent resistivity and corresponding impedance phase of the orthogonal electric and magnetic field horizontal Components in the frequency EM sounding with horizontal electric dipole(FEMS) are discussed, the better effect of the joint inversion using the pair of EM response functions is obtained. The problems on the one-dimensional joint inversion for the MT and FEMS apparent resistivities, for which the observed frequency bands partly overlape each other, are studied. It is shown that this joint inversion is applicable and effective:the joint inversions of the practical data for two kinds of EM methods at two sites give the results well corresponding to the drilling data. The simulated MT inversions for the data of two kinds of EM methods are made, and more confident results also are obtained. 相似文献
15.
Mansour A. Al-Garni 《Acta Geophysica》2010,58(1):143-162
A new approach is proposed in order to interpret spontaneous potential (self-potential) anomalies related to simple geometric-shaped
models such as sphere, horizontal cylinder, and vertical cylinder. This approach is mainly based on using neural network inversion
of SP anomalies, particularly modular algorithm, for estimating the parameters of different simple geometrical bodies. However,
Hilbert transforms are involved to determine the origin location in order to reduce the parameters which minimize the ambiguity
in the inverted models. The inversion has been tested first on synthetic data from different models, using only one well-trained
network. The results of inversion show that the parameter values derived by the inversion are identical to the true values
of parameters. Noise analysis has been also examined, where the results of the inversion produce acceptable results up to
10% of white Gaussian noise. The validity of the neural network inversion is demonstrated through published real field SP
taken from southern Bavarian Woods, Germany. A comparable and acceptable agreement is shown between the results of inversion
derived by the neural network and those from the real field data. 相似文献
16.
复电阻率法二维有限元数值模拟 总被引：9，自引：2，他引：9
伴随着复电阻率法的广泛应用，发展精确和快速的正演和反演算法成为复电阻率法研究的重点.本文采用基于三角单元剖分的有限单元法进行了复电阻率二维数值模拟研究.为了提高计算速度，对无穷远边界进行了近似处理.整个正演计算过程分为两步，首先采用有限单元法计算四个不同频率的视复电阻率数据，然后对前一步得到的视复电阻率数据采用递推算法计算视Cole-Cole参数.采用这种正演算法与一维正演的结果进行了对比，验证了本文方法的正确性.设计了两个二维极化模型，数值模拟结果表明视复电阻率和Cole-Cole视参数等值线断面图对于异常目标体都有比较明显的反映. 相似文献
17.
18.
3D inversion of DC data using artificial neural networks 总被引：2，自引：0，他引：2
Ahmad Neyamadpour W. A. T. Wan Abdullah Samsudin Taib Danesh Niamadpour 《Studia Geophysica et Geodaetica》2010,54(3):465-485
In this paper, we investigate the applicability of artificial neural networks in inverting three-dimensional DC resistivity
imaging data. The model used to produce synthetic data for training the artificial neural network (ANN) system was a homogeneous
medium of resistivity 100 Ωm with an embedded anomalous body of resistivity 1000 Ωm. The different sizes for anomalous body
were selected and their location was changed to different positions within the homogeneous model mesh elements. The 3D data
set was generated using a finite element forward modeling code through standard 3D modeling software. We investigated different
learning paradigms in the training process of the neural network. Resilient propagation was more efficient than any other
paradigm. We studied the effect of the data type used on neural network inversion and found that the use of location and the
apparent resistivity of data points as the input and corresponding true resistivity as the output of networks produces satisfactory
results. We also investigated the effect of the training data pool volume on the inversion properties. We created several
synthetic data sets to study the interpolation and extrapolation properties of the ANN. The range of 100–1000 Ωm was divided
into six resistivity values as the background resistivity and different resistivity values were also used for the anomalous
body. Results from numerous neural network tests indicate that the neural network possesses sufficient interpolation and extrapolation
abilities with the selected volume of training data. The trained network was also applied on a real field dataset, collected
by a pole-pole array using a square grid (8 ×8) with a 2-m electrode spacing. The inversion results demonstrate that the trained
network was able to invert three-dimensional electrical resistivity imaging data. The interpreted results of neural network
also agree with the known information about the investigation area. 相似文献
19.
地表起伏地形在野外矿产资源勘察中不可避免,其对直流电阻率法勘探影响巨大.近年来,电阻率三维正演取得诸多进展,特别是应用非结构网格我们能够进行任意复杂地形和几何模型的电阻率三维数值模拟,但面向实际应用的起伏地形下电阻率三维反演依然困难.本文基于非结构化四面体网格,并考虑到应用GPS/GNSS时,区域地球物理调查中可非规则布设测网的实际特点,实现了任意地形(平坦或起伏)条件下、任意布设的偶极-偶极视电阻率数据的不完全Gauss-Newton三维反演.合成数据的反演结果表明了方法的有效性,可应用于复杂野外环境下的三维电法勘探. 相似文献