摘要
测井资料中包含丰富的地层岩性信息,是岩性分析的基础资料。数理统计等传统方法难以准确地反映测井资料与地层岩性的非线性映射关系,而具有分布处理、自学习、自组织和高度非线性的神经网络能够较好地解决这个问题。将径向基神经网络应用到测井资料岩性识别中,结合准噶尔盆地某井的实际测井资料和岩性剖面资料,建立基于径向基神经网络的岩性识别模型。实际应用表明,径向基神经网络可以用来进行岩性识别,收敛速度快,且识别正确率较高。
.Well logging data contain rich formation lithologic information and it is the basic data of lithological identification. The traditional methods are difficult to reflect the nonlinear mapping relationship between well logging data and litholog, but the neural network can solve this problem. It has the advantages of distributed processing, automatic study, automatic organization ,high nonlinear and others. A radial basis function neural network mode of lithologic identification based on the radial basis function is established to study a real well logging data and lithologic data. Practical application shows that the accuracy of identification is high and the convergence speed is fast.
出处
《重庆科技学院学报（自然科学版）》
CAS
2008年第3期8-9,12,共3页
Journal of Chongqing University of Science and Technology：Natural Sciences Edition
基金
中石化重大科技项目"基于钻井工程地质数据库的钻井模拟"资助(JP04014)
胜利石油管理局博士后基金资助
关键词
径向基神经网络
测井资料
岩性识别
RBF neural network
well logging data
lithologic identification
