二. 油气储层预测和描述
（负责人：黄绪德 牟永光 陈小宏）
Prediction and Description of and Gas
Reservoir
(Conveners: HUANG Xude MU Yongguang CHEN Xiaohong)
in
Recognizing the Special Lithologic Reservoirs
Wang Yonggang Zhang Junhua
(University of Petroleum Dongying 257061)
联系地址: 山东东营市 石油大学(华东)物探教研室
邮编: 257061
E-mail Address: zhangjh@sunctr.hdpu.edu.cn
Peng Zhenming
(Cheng Du University of Technology, Cheng Du 610059)
Zhu Zhixin ,Liang Guosheng
(Exploration and Development Science Institution, ZPEB, Pu Yang 457001)
Subject heading: back-propagation network , genetic algorithm,
evolutionary learning , seismic parameters, sonic
logging extrapolation
3. The
application study of pattern recognition method in seismic data interpretation
Zeng Rui 1, Wang Runqiu2, and Hu
Tianyue1
1 Department
of Geophysics, Peking University, Beijing 100871, China
2 Department of Geosciences, University of Petroleum, Beijing 102200, China
Abstract
This paper discusses the three major parts used in pattern recognition: feature extraction, cluster analysis and classify discrimination. In the stage of feature extraction, Karhunen-Loeve transform is used to process the original data to reduce the number of the features based on the maintenance of the major information of the original features. An improved K-nearest neighbour (K-NN) algorithm is used in cluster analysis for the pre-processed data. Finally, FISHER linear arbiter is employed to classify the unknown data to achieve the purpose of the predicted classification. This paper processes and discriminates various groups of typical data to confirm the theory and the algorithm. The selection of the parameters used in K-NN algorithm and the threshold value in the discriminate functions are discussed. The discussed methods and conclusions would be useful in seismic data interpretation.
4．The Application of Kriging Techniques in the Transform from Seismic
Attribute to
Reservoir Characteristic Value
Zhang
Junhua Wang Yonggang et.al.
(University
of Petroleum Dongying 257061)
The Kriging technique has been famous for mining. Now it is used in oil industry also. People utilize its Theory of Regionalized Variables to solve complex geostatistic problem. In the paper, we simply review the development of this technique and provide the calculating formula of variogram. In the same time, the relation about the half variogram, variogram sill and covariance is given. Three effect Kriging methods(Kriging, Co-Kriging and Bayesian Kriging) are selected and the programs are developed by ourself. They are successfully used in the TJH area, the processing results are satisfied.
Subject words Kriging, variogram, seismic attribute, porosity
请送 专题(2) 油气储层预测和描述
联系地址: 山东东营市 石油大学(华东)物探教研室
邮编: 257061
E-mail Address: zhangjh@sunctr.hdpu.edu.cn
5．非线性地震信息油气检测
安鸿伟 李正文 许强
（成都理工学院）
Oil & Gas
Detection Using Nonlinear Seismic Information
Hongwei An Zhengwen Li
Qiang Xu
(Chengdu
University of Technology)
The geologic evolvement is a nonlinear process.
The information in seismic records used to detect stratum feature bear
nonlinear event. Oil & gas detection using nonlinear seismic information
firstly need to pick-up information obtained by nonlinear theory in a time
window. It includes four kinds: (1) Catastrophe information
based on tine Catastrophe model: Catastrophe number, sudden jumping potential, sudden jumping time and sudden
jumping interval. (2) Fractal: information dimension and relating dimension.
(3) Chaos: Lyapunov index. (4) Wavelet parameter: signal fantastic nature,
wavelet coefficient reflecting signal catastrophe, wave impedance with wavelet
arithmetic. Secondly, oil & Gas detection adopt method of artificial nerval
network. If there are well data in survey area, it uses BP arithmetic.
Otherwise it uses SOM arithmetic. This method
has achieved better result in the actual application.
6．Study on the Relation
between Sedimentary Micro-Facies and
Pore Structure of Reservoirs
Sun Hong Zhi
(Exploration Dep. Of Daqing Oil Field Company , 163453)
The pore structure of reservoirs shows synthetically the storing and permeating capacity. Hydraulically dynamical condition while depositing is the basis that proves that the sedimentary micro-facies is related to the pore structure of reservoirs. This paper combines the sedimentary micro-facies with the pore structure of reservoirs, and gives theirs distribution law over an area. The result of this paper is beneficial to direct the well locations deploy and production capability prediction. The method of this paper is effect and economically.
7. 薄互储层参数预测方法研究
高美娟 田景文 谢美萍
（大庆石油学院 安达 151400） （哈尔滨工程大学自动化学院 哈尔滨 150006）
Study
on the Prediction Methods of Thin Interbedded Reservoir Parameter
GAO Meijuan TIAN Jingwen XIE Meiping
The prediction of the thin interbedded reservoir parameter has an important meaning for oil field production and effect increase, reducing production costs ,raising exploit effect and total economic effect．As the intensify of oil field prospect and the rise of oil field prospect level ,the easy judge oil Reservoir of structure becomes fewer and fewer，searching conceal rocky Reservoir and carrying on the prediction of reservoir parameter become the main object. In point of the thin interbedded reservoir the features of reservoir thickness thin and the great change in horizontal and complexity of the reservoir, we offer two new prediction methods of reservoir parameter— wavelet neural network and projection pursuit wavelet neural network
8. 克希霍夫型叠前深度偏移孔径的计算方法
（王润秋 石油大学 102200）
The computation method for aperture of kirchhoff-type prestack depth migration
(wang runqiu petroleum university of china 102200)
ABSTRACT
The size of the aperture has an importation influence on the result of prestack depthKirchhoff-type migration. This paper researched the method of calculation minimum aperture, Which is based summation of diffraction stack and zero-order ray theory. For purpose of realty seismic data processing, we have improved the method of computing aperture as follow:
1.We will manually create some reflection interval on meshed model, that adjust to the break point of original reflector, It is very helpful that obtain to the smooth reflection-time surface .
2.The concept of the Fresnel Zone was primary introduced in the frequency domain, so we must compensate seismic wave attenuation frequency with depth, then separately calculate an aperture and migration for each monofrequency .
3. Global optimization algorithm is used to obtain the stationary point.
The method of
minimum aperture have applied to marmousi model, The result of test show the
quantity of prestack migration prior to that without calculating the minimum
aperture.
9．The Nonlinear Characteristic Analysis of Seismic Records
(Exploration and Development Science Institute, ZPEB)
Seismic data contains complex and comprehensive information of sedimentary sequence, present the strong nonlinear characteristic. Seismic records probably give rise to some illusion because of low R/N ratio & low resolution resulting from acquisition and data processing limitation. In this paper, the author have brought forward the precondition of extracting nonlinear characteristic for seismic data, improved the calculating model of fractal associated dimension, catastrophe parameter and Lyapunov index. In finally, the author thought it will bring bright future for reservoir description and recognition combining nonlinear characteristic with linear characteristic of seismic records, using fuzzy clustering analysis, pattern recognition and artificial neural network.
Key Words: Seismic data, Fractal associated dimension, catastrophe parameter, Lyapunov index, Analysis.
Yu Wenhui Zhang Liqing Shi Xueming Wang Jiaying
Geoscience University of China (Wuhan) 430074
We know that frequency of conversed wave produced by directive P-wave on the subsurface is not smaller than that of reflection P-wave by analyzing and processing VSP data of several main land exploring area all over the nation. At present, the wave attained by the method of observing conversed wave may be conversed wave that explosion pressure wave generates on the free surface. Numerical analysis, computer simulation, ultrasonic waves test and large offset VSP experiment testify the conclusion.
Have been eliminating ghost reflection conversed wave on VSP section, we got high quality VSP conversed wave section and under the VSP method guidance, we tentatively processed a part of multi-wave data. After that, frequency compositions of conversed wave are restored, section quality and resolution are improved.
11. 高频电磁波测井资料反演的优选调参子空间方法
张美玲 邢光龙
刘曼芬 杨善德
(吉林大学物理系,吉林省长春市,130023)
An Optimum Choice Method of Adjusted Inversion
Parameters'
Sub-space on High Frequency Electromagnetic Wave
Logging Inversion
Zhang M. L. Xing G. L. etal.
(Jinlin University ,Changchun 130023,China)
Although the linearity inversion methods have the advantages that their solution's computation time is less and the convergent speed is rapid in some degree, some problems on unsteadiness and multi-solution become more serious in their inversion courses. We think this is a good way that converts the high-dimension search into multi-time low-dimension searches, and the low-dimension search direction of each time is a good search direction. This paper proposes a new method that is called An Optimum Choice Method of Adjusted Inversion Parameters' Sub-space. The method is an attempt in the aspect. It is beneficial to overcome the multi-solution's problems and save a deal of computation time and store space. The good effect has been obtained when it is used into the inversion problem of High Frequency Electromagnetic Wave logging.
Key words: Optimum Choice Method of Adjusted Inversion Parameters' Sub-space, Inversion of High Frequency Electromagnetic Wave Logging.
12. 黄骅坳陷第三系砂岩弹性波参数的超声波实验研究
毛宁波 彭大均
（成都理工学院，成都 610059）
The ultrasonic experiment of elastic wave parameters of HuangHua depression teriary sandstone
(Mao Ningbo Peng Dajun) (Chengdu university of Technology Chengdu 610059)
Abstract： The
elastic parameters are important in seismic prospecting. This paper presents
the ultrasonic data of velocities of P-wave and S-wave of sandstone rock cores
collected from HuangHua Depression Teriary ,China ,and their dynamic
elastic parameters (which include bulk compressibility modulus, Youngs
modulus, shearing modulus and Poissons ratio) measured by experiment under the
simulated formation temperature and pressure. From the experimental results,
the authors studied the variations of dynamic elastic parameters and the
factors influenced them, such as the buried depth of formation, temperature,
pressure, porosity and saturated water etc.
Wu Xijing (Petroleum University, Beijing, 102200)
Wang Jieyi (Xinjiang Oil Field, Keromayi, 834000)
In the oil well production, oil,water and gas flow up to surface along the
steel circular pipe, having the aid of sponting, lifting, or pumping. Regardless of whatever the fluid flow state is down hole, stable or unstable, the steady flow models are always employed during the log analyzing, and the interpreted
results are usually erroneous. In this paper, the unsteady flowing mechanism
of the fluid in rod pumping well are studied, and a new model is produced
for log interpretation based on the one dimension wave flow theory.
14.
GEOLOGICAL INTERPRETATION ON THE FEATURE OF THE PROBABILITY DISTRIBUTION TYPES OF
THE RESERVOIR PERMEABILITY IN DAQING OILFIELD
Lu
Yingzhong Chu
Zehan et al
(Petroleum
University,Beijing,102200)
Abstract
It
is necessary to do geological interpretation to the probability distribution
types of the reservoir permeability. The sedimentary facies ,the diagenesis and
the
probability distribution types of the
permeability from S,P or G oil reservoir respectively from Changyuan region in
Daqing oilfield was respectively contrasted with that from Longhupao oilfield
to the west of Changyuan region in it. It is showed that the remarkable
difference of diagenesis has caused the variation of the probability
distribution feature of the permeability of oil reservoir, but the difference
of sedimentary facies didn't change the probability distribution feature of it
and only led to the variation of numerical value of it.
15．The Multi-Parameters’ Distinguishing method’s Application on
High-Resistivity Water Beds’ Recognition
Sun H. Z. and Li X. H.
(Dep. Of Exploration, Daqing Oil Field Company ,163453)
High-resistivity water beds’ recognition is a difficult problem in the course of oil field’ exploration. This paper proposes to use the multi-parameters’ distinguishing methods to deal with some well logging measurements to overcome the problem. The well logging measurements include the deep-induction logging curve, the difference of bulk-porosity and neutron-porosity and the formation water’s resistivity. The sample lays include 13 oil lays and 17 water lays. The verified lays include 16 reservoir lays. And 14 of all is verified successfully. The fitness ratio is raised up to 87.5% , and when the method is not used the fitness ratio is only 63.7%.
16. The Evaluating
Model of Porosity and Permeability for Metamorphic Fractured Reservoir
HUAN Fei
(Cheng Du university of technology, Cheng du , 610059)
ZHANG Xiaoyan LU Zhanglin LU Dielin and ZHU Jianguang
(Exploration & Development Technology LTD. CO., ZPEB , PU Yang 457001)
It is generally difficult for metamorphic to form the effective reservoir because of its tight matrix, poor porosity and permeability. While affected by weathering, leaching and structure, it can lead to combine weathering crushed zone with structural fracture zone to form the effective reservoir. Then, we can obtain the petrophysical parameters by classification and description using logging and core data.
In this paper, the author have given rive to several indicating models including fracture development, classification, width, porosity and permeability estimation and so on . By above models, we reprocessed the logging data for 18 wells in studying area, interpreted 85 substratum and classified Ⅰ-class,Ⅱ-class and Ⅲ-class fracture.
Key words: Metamorphic Fractured reservoir, Porosity, permeability, Modeling,
Classification