论文详情
BP神经网络在沉积微相自动识别中的应用
石油实验地质
2014年 36卷 第s1期
阅读：132
查看详情
Title
Application of BP neural network to sedimentary micro-facies identification
Authors
Liu Junling
Wu Xuedong
Wang Dongzhai
He Zhenggang
单位
中国石化 西北油田分公司 塔河采油三厂, 新疆轮台 841600
Organization
No.3 Tahe Oil Production Plant, SINOPEC Northwest Company, Luntai, Xinjiang 841600, China
摘要
提出了基于测井数据深度挖掘和前馈式(BP)人工神经网络算法的沉积微相识别方法.在测井数据较少、井多的条件下深入挖掘有限的测井数据,获取蕴含沉积学意义的参数,提高了测井数据的利用率.通过一系列实验,研究了BP人工神经网络拓扑结构的优选准则,并提出了成长型网络训练方法.最后利用建立的样本集和自然样本进行网络训练和微相识别,准确率达83%以上.在测井数据不足、微相特征复杂的条件下实现了高效率、高准确度的沉积微相识别.
Abstract
A method of sedimentary micro-facies identification based on logging data and BP neural network was proposed in this paper. Through deeply exploring limited logging data, sedimentological sample indexes were gained and the utilization rate of logging data was improved. A series of experiments were conducted in order to find the optimization criterion of the BP artificial neural network and a growing network training method was put forward. Finally, a actual case of net training and micro-facies identification by using sample set and natural samples was analyzed, which showed an accuracy ratio for 83% and realized both high efficiency and precision of micro-facies identification.
关键词：
BP神经网络;
测井数据深入挖掘;
成长型神经网络;
沉积微相自动识别;
Keywords:
BP neural network;
further study of logging data;
growing artificial neural network;
identification of microfacies;
DOI
https://doi.org/10.11781/sysydz2014S1052
论文详情