摘要
为使瞬变电磁勘探的反演法简单和通用 ,提出了一种瞬变电磁勘探中的人工神经网络反演法 ,基于人工神经网络的BP算法 ,设计和开发了一个自适应的人工神经网络反演系统 .该系统避免了复杂的电磁场计算 ,只需经过学习训练就能够解决复杂的实际问题 ,而且具有记忆功能 ,从而使瞬变电磁法的反演工作具有延续性和可继承性 .通过对实际的地质数据进行反演 ,结果表明 ,该方法是切实可行的 .
A new inversion method of artificial neural network is presented for the purpose of simplification and more general application for the inversion method in transient electromagnetic exploration. On the basis of back propagation (BP) algorithm, an adaptive inversion system of artificial neural network is designed and developed. The system avoids complicated calculations of electromagnetic field and can be used to solve some practical problems through learning. It also possesses memorizing and make the inversion be of continuity which depends not only on current event but also the previous event. The inversion results are in good agreement with geological data.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2001年第6期604-607,615,共5页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(6 99780 19)
西安交通大学科研基金资助项目
关键词
瞬变电磁法
反演法
人工神经网络
电磁勘探
Backpropagation
Eddy currents
Feature extraction
Inverse problems
Neural networks
