首页 | 本学科首页   官方微博 | 高级检索  
     

应用人工神经网络确定声波孔隙度
引用本文:夏克文,宋建平,李昌彪.应用人工神经网络确定声波孔隙度[J].小型微型计算机系统,2004,25(4):716-718.
作者姓名:夏克文  宋建平  李昌彪
作者单位:西安交通大学,电子与信息工程学院,陕西,西安,710049
基金项目:中国石油天然气集团公司科学技术基金项目 ( 2 0 0 0 2 0 6-2 )资助
摘    要:利用声波测井获得的时差求取地层孔隙度是石油测井解释中一项重要任务,传统的方法主要是利用Wyllie实验得到的时间平均公式以及其改进形式或经验公式,均为统计学方法,在具体应用上是很不方便的,优越于统计学理论的人工神经网络方法具有高度的自学习、自适应和抗干扰性等优点,采用带有非线性连接权的二层前馈神经网络能够取代三层BP网络的功能,实际应用表明,应用神经网络能够很好地确定声波孔隙度.

关 键 词:人工神经网络  非线性连接权  声波孔隙度
文章编号:1000-1220(2004)04-0716-03

Determining Acoustic Porosity with Artificial Neural Network
XIA Ke wen,SONG Jian ping,LI Chang biao.Determining Acoustic Porosity with Artificial Neural Network[J].Mini-micro Systems,2004,25(4):716-718.
Authors:XIA Ke wen  SONG Jian ping  LI Chang biao
Abstract:It is an important task to determine the formation porosity with slowness acquired from acoustic logging in oil logging interpretation. The traditional methods are the time average equation derived from experiment by Wyllie, and its improved formulae or some experience formulae based on statistics theory, but these equations are inconvenient in concrete application. The method of artificial neural network (ANN) is superior to that of statistics theory, and has the characteristics of high self study, self adaptation and interference resistant. A two layer feed forward ANN with nonlinear connected weights can replace three layer BP algorithm, Actual application shows the method of ANN to determine acoustic porosity is very good.
Keywords:artificial neural network (ANN)  nonlinear connected weights  acoustic porosity
本文献已被 CNKI 维普 万方数据 等数据库收录!
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