姚休义^1,,
滕云田²,
杨冬梅²,
姚远³
1. 云南省地震局, 昆明 650224
2. 中国地震局地球物理研究所, 北京 100081
3. 川滇国家地震预报实验场云南分中心, 昆明 650224

基金项目: 中国地震局科技星火计划（XH18041Y），国家自然科学基金（41504129），国家重大科学仪器设备开发专项项目（2014YQ100817）共同资助


详细信息

作者简介: 姚休义, 女, 1988年生, 云南省地震局助理研究员.研究方向为地磁学.E-mail:xiuyiyao@126.com

中图分类号: P318

收稿日期:2017-02-21
修回日期:2018-03-21
上线日期:2018-06-05

Reconstruction of geomagnetic data based on artificial neural network

YAO XiuYi^1,,
TENG YunTian²,
YANG DongMei²,
YAO Yuan³
1. Earthquake Administration of Yunnan Province, Kunming 650224, China
2. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China
3. Yunnan Sub-center of China Earthquake Science Experiment, Kunming 650224, China


MSC: P318

--> Received Date: 21 February 2017
Revised Date: 21 March 2018
Available Online: 05 June 2018


摘要
摘要:在距离数据缺失台站一定范围内选取参考台作为输入，构建非线性BP神经网络并进行地磁观测数据重构研究.数据仿真结果显示，重构数据和原始记录数据吻合程度较高，重构残差较小，磁静日重构平均残差仅为0.11 nT，磁扰日平均重构残差为0.23 nT.重点对磁场活动最剧烈时段内的数据进行了短时重构，平均残差由0.4 nT降低到0.2 nT，重构效果得到较大改进.计算了原始数据与重构数据的功率谱密度，除部分高频信号外，二者变化特征基本相同，相关性高达1.0.从时域和频域验证了BP神经网络在地磁相对记录数据重构上的有效性，并将其运用于实际缺失数据重构，取得较好效果.
关键词: BP神经网络/
数据重构/
地磁观测/
检验

Abstract:In this work, we used the back-propagation neural network to reconstruct missing geomagnetic data based on data of adjacent observatories. The simulation results show that reconstructed data are close to original data. The correlation of their power spectral density is about 1.0. The average residual between reconstructed data and original data is about 0.11 nT in quiet days, and reaches 0.23 nT in disturbed days. When we use this method to reconstruct the data of a short time period with intense magnetic disturbance, the residual reduces to 0.2 nT from 0.4 nT. Based on the comparison of time and frequency domains, we suggest that the back-propagation network is an effective tool for geomagnetic data reconstruction.
Key words:Back-Propagation Network/
Data reconstruction/
Geomagnetic data/
Test



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=dqwlxb_14550