补偿模糊神经网络在储层参数预测中的应用  

APPLICATION OF CFNN IN PREDICTION OF RESERVOIR PARAMETERS

作　　者：甄兆聪[1] 肖慈珣[1] 

机构地区：[1]成都理工大学信息工程学院,成都610059

出　　处：《物探化探计算技术》2002年第2期124-128,共5页Computing Techniques For Geophysical and Geochemical Exploration

基　　金：国家高技术研究发展计划 ( 86 3计划 )项目 ( 82 0 -0 7-0 2 )

摘　　要：为了克服常规 BP神经网络法在预测储层参数中出现学习速度慢、无法结合专家知识等不足 ,我们引入了补偿模糊神经网络。它是一个结合了补偿模糊逻辑和神经网络的混合系统 ,由面向控制和面向决策的神经元组成 ,其模糊运算采用动态的、全局优化运算 ,学习速度快、学习过程稳定 ,将其用于储层参数预测效果良好。General BP neural networks(BPNN) has several lacks, such as slow learning speed and being not able to combine with the expert knowledge etc. In order to overcome these lacks we introduce the compensative fuzzy neural network(CFNN), which combines compensative fuzzy logic and neural network, and is composed of control oriented cell and decision-making cell. It's fuzzy computation is dynamic and global optimized, therefore its speed is fast. It turns out that CFNN is effective in prediction of reservoir parameters.

关 键 词：补偿模糊神经网络 模糊逻辑 神经网络 储层参数预测 ANH 

分 类 号：P618.130.8[天文地球—矿床学] TP183[天文地球—地质学]