地震多参数BP神经网络预测煤层厚度Forecasting coal layer thickness by BP neural network from multiple seismic parameters
韩万林,张幼蒂,李梁
HAN Wan lin 1;ZHANG You di 2;LI Nai liang 2 (1 Tongji University;
摘要(Abstract)：
依据煤层反射波运动学和动力学特征 ,提取出了波峰波谷振幅 A1、平均频率 Fa、主频带能量 Qf1、低频带宽能量 Qf 和峰值频率 Fmain等 5个地震特征参数。选取 8组学习样本 ,利用 4层 BP(Back Propagation)人工神经网络模型 ,采用动量法和自适应调整的改进算法 ,训练 BP网络 ,用训练好的 BP网络预测煤层厚度。经实例验证 ,地震多参数 BP网络预测煤层厚度精度高 ,是一种有效的煤厚预测方法
Five seismic parameters such as amplitude of wave crest and hollow(A1),average frequency(Fa),energy in dominant frequency domain(Qf1),energy in low frequency domain(Qf),peak frequency(Fmain) are derived according to the seismic kinematics and dynamic characteristics of coal layer thickness.Eight groups of studying samples,made use of BP(Back Propagation)neural network of four layers improved by adopting momentum algorithm and self adaptive adjusting learning rate algorithm to train the BP neural network,and used the trained BP network to forecast coal layer thickness.It was proved that forecasting coal layer thickness by BP neural network from multiple seismic parameters had high accuracy by the practical data.and is an effective approach for forecasting coal layer thickness.
关键词(KeyWords)：
煤层厚度;地震特征参数;BP神经网络
coal layer thickness;seismic characteristics parameter;BP neural network.
基金项目(Foundation): 国家自然科学基金资助项目 (编号 :5 97740 0 5
作者(Authors):
韩万林,张幼蒂,李梁
HAN Wan lin 1;ZHANG You di 2;LI Nai liang 2 (1 Tongji University;
参考文献(References)：
- [1] 姚建阳.在频率域中求取薄层地层的厚度[J].石油地球物理勘探,1991,26(5):594—599.
- [2] 程增庆,吴弈峰,赵忠清等.用地震反射波定量解释煤层厚度的方法[J].地球物理学报,1991,34(5):657—662.
- [3] 窦易升.薄层厚度定量解释的振幅谱方法研究[J].石油物探信息,1994.
- [4] 胡强,刘福贵,冯德永等.利用振幅、频率信息综合反演储层厚度[J].石油地球物理勘探,1995,30(增刊1):96—101.
- [5] 中国煤田地质总局岩性勘探课题组.煤田地球物理岩性勘探技术文集[D].北京:煤炭工业出版社,1996.
- [6] 邵治龙,尹成,黄德济等.用多参量法计算薄层和薄互层厚度[J].石油物探,1997,36(4):92—98.
- [7] 楼顺天,施阳.基于MATLAB的系统分析与设计—神经网络[M].西安:西安电子科技大学出版社,1999.
- [8] 崔若飞.地震资料矿井构造解释方法及其应用[M].北京:煤炭工业出版社,199