摘要
讨论了常用的欧拉矢量模型和函数拟合模型的优缺点,提出了基于欧拉矢量的BP神经网络模型。该模型运用欧拉矢量的地学性质,结合BP神经网络在处理需要同时考虑许多因素和条件的、不确定和模糊的信息时的优势,可以较好地区分块体整体的刚性旋转及内部的弹性形变。经实例验证,取得较好的精度。
In regional crustal movement research, mathematical model is always used to estimate with- out observation the points which need attention. Then, we can build up a relatively even and meaning- ful regional crustal movement velocity field. In this paper we analyze the strengths and weaknesses of the common Euler and function models, and propose a new model with BP neural network based on Euler vector. This proposed model uses the geological properties of a Euler vector and the superiority of BP neural network. It considers the various influences and uncertain information found in data pro- cessing. Therefore, the model can distinguish inner elastic strain from rigid-body of the plate. The proposed model obtains precision through specimen verification.
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2014年第3期362-366,共5页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金资助项目(40637034
41210006
41274005)~~
关键词
地壳运动
速率场
欧拉矢量
BP神经网络
crustal movement
velocity field
Euler vector
BP neural network
