储层密度预测技术研究
|
摘要
储层密度预测技术是一种基于改进的混合智能学习算法的地震非线性预测方法。它是利用地震波阻抗剖面在测井密度数据约束下,对目标问题编码产生染色体,将禁忌搜索算法加在遗传操作的交叉点上,使染色体不断进化,并按一定的概率自始至终执行遗传算法和模糊神经网络算法,且概率自适应变化,以达到混合算法均衡,实现储层密度预测,获得高分辨率和高精度的储层密度剖面。在获得的储层密度剖面基础上,利用流体密度计算技术,即可得到流体密度剖面。应用实例与统计表明,流体密度从富集油气层的密度变至水层的密度,其异常相对幅度可达70%以上,因此流体密度是预测油气层的一个“绝好参数”,它是反映流体性质的最直接证据。可为地震勘探直接寻找油气提供一个关键性参数。
The reservoir density prediction technique is a non-linear seismic prediction approach based on improved hybrid intelligent learning algorithm.Using seismic acoustic impedance section and the constraint of logging density data,the approach produces chromosome by encoding the target problem and adds the tabu searching algorithm at the cross points of genetic operation,making the chromosome evaluate constantly and carrying out genetic algorithm and ANFIS (Adaptive-Network-based Fuzzy Inference System)algorithm according to certain probability from the beginning to the end;meanwhile, the probability changes adaptively,making the hybrid algorithm reach equilibrium,realizing reservoir density prediction and yielding the reservoir density sections with high-resolution and high-precision.On the basis of obtaining reservoir density sections and using fluid density computation technique,the fluid density sections can be yielded.The application cases and statistics showed the relative amplitude anomaly could reach to 70% and above if fluid density changed from the density of oil/gas-rich layer to that of aquifer bed,so the fluid density is an excellent parameter of oil/gas layer prediction,which is direct proof reflecting the fluid nature and provides a key parameter to HI (hydrocarbon indicator) approach of seismic exploration.
The reservoir density prediction technique is a non-linear seismic prediction approach based on improved hybrid intelligent learning algorithm.Using seismic acoustic impedance section and the constraint of logging density data,the approach produces chromosome by encoding the target problem and adds the tabu searching algorithm at the cross points of genetic operation,making the chromosome evaluate constantly and carrying out genetic algorithm and ANFIS (Adaptive-Network-based Fuzzy Inference System)algorithm according to certain probability from the beginning to the end;meanwhile, the probability changes adaptively,making the hybrid algorithm reach equilibrium,realizing reservoir density prediction and yielding the reservoir density sections with high-resolution and high-precision.On the basis of obtaining reservoir density sections and using fluid density computation technique,the fluid density sections can be yielded.The application cases and statistics showed the relative amplitude anomaly could reach to 70% and above if fluid density changed from the density of oil/gas-rich layer to that of aquifer bed,so the fluid density is an excellent parameter of oil/gas layer prediction,which is direct proof reflecting the fluid nature and provides a key parameter to HI (hydrocarbon indicator) approach of seismic exploration.
引文
[1]黄绪德.油气预测与油气藏描述——地震勘探直接找油气.南京:江苏科学技术出版社,2003
[2]王小平,曹立明.遗传算法——理论、应用与软件实现.西安:西安交通大学出版社,2002
[3]刘勇等.非数值并行计算(第2册)——遗传算法.北京:科学出版社,1995
[4]Goldberg D E.Genetic algorithms in search,opti miza-tion,and machine learning.Reading,MA,Addison-Wesley,1989
[5]Whitleyet D.Genetic algorithms and neural networks:opti mizing connections and connectivity.Parallel Compution,1990,14:347~361
[6]焦李成.神经网络的应用和实现.西安:西安电子科技大学出版社,1993
[7]戴葵等译.神经网络设计.北京:机械工业出版社,2003
[8]Jang J S R.ANFIS:Adaptive-network-based fuzzy in-ference system.IEEE Transaction on System,Man and Cybernetics,1993,23(3):665~685
[9]Glover F.Future paths for Integer programming and links to artifical intelligence.Computers and Opera-tions Research,1986,5:533~549
[10]李勇等.地震优化非线性反演方法及应用研究.矿物岩石,2004,24(1):101~104
[11]Biot M A.Theory of propagation of elastic waves in a fluid-saturated porous solid.The Journal of The A-coustical Society of America,1956,28(3)
[2]王小平,曹立明.遗传算法——理论、应用与软件实现.西安:西安交通大学出版社,2002
[3]刘勇等.非数值并行计算(第2册)——遗传算法.北京:科学出版社,1995
[4]Goldberg D E.Genetic algorithms in search,opti miza-tion,and machine learning.Reading,MA,Addison-Wesley,1989
[5]Whitleyet D.Genetic algorithms and neural networks:opti mizing connections and connectivity.Parallel Compution,1990,14:347~361
[6]焦李成.神经网络的应用和实现.西安:西安电子科技大学出版社,1993
[7]戴葵等译.神经网络设计.北京:机械工业出版社,2003
[8]Jang J S R.ANFIS:Adaptive-network-based fuzzy in-ference system.IEEE Transaction on System,Man and Cybernetics,1993,23(3):665~685
[9]Glover F.Future paths for Integer programming and links to artifical intelligence.Computers and Opera-tions Research,1986,5:533~549
[10]李勇等.地震优化非线性反演方法及应用研究.矿物岩石,2004,24(1):101~104
[11]Biot M A.Theory of propagation of elastic waves in a fluid-saturated porous solid.The Journal of The A-coustical Society of America,1956,28(3)
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心