The application of artificial neural network expert system to the transient electromagnetic method inversion
The application of artificial neural network expert system to the transient electromagnetic method inversion
复制标题
DOI：
--
复制
发表时间：
2001
期刊：
影响因子：
--
通讯作者：
Li Shi
中科院分区：
文献类型：
--
作者：
Li Shi
目前瞬变电磁法反演多采用拟合方法，但其理论体系复杂，不易掌握，缺乏继承性，难以推广，因此要求科研人员具有丰富的工作经验，对反演结果的精度要求较高。提出了一种基于人工神经网络的专家系统，避免了繁琐的电磁计算，只要经过适当的训练和学习，就能有效地解决实际问题，而且，它的存储能力增强了瞬变电磁法反演领域的连续性和继承性，随着专家系统的完善，将有沿着光明的前景。
Fitting is used mostly in the transient electromagnetic method inversion at present.However there is a few of deficiencies as following.First,its complex theoretical system is no easy to be mastered;secondly,it is difficult to be popularized in the shortage of inheritance Therefore,scientific researcher is requested to have rich work experience;thirdly,computation complicated for its huge data package.In this paper,an expert system based on artificial neura network is put forward,which avoids detailed and knotted electromagnetic computation.Once afte appropriate training and learning,it can solve practical problem effectively.Furthermore,it memory capacity enhances the continuity and inheritance in the field of TEM inversion.A brigh outlook will be brought along with the consummation of this expert system.