基于LSTM的地震前兆数据分析算法设计与实现  

Design and implementation of seismic precursor data analysis algorithm based on LSTM

作　　者：王圆圆 孙可可 WANG Yuanyuan;SUN Keke(School of Emergency Management,Institute of Disaster Prevention,Sanhe Hebei 065201,China)

机构地区：[1]防灾科技学院应急管理学院,河北三河065201

出　　处：《智能计算机与应用》2020年第2期228-231,共4页Intelligent Computer and Applications

摘　　要：地震前兆数据具有短期、中期、长期的变化,研究地震前兆数据的变化规律对地震预测具有重要意义。论文采用目前最流行的深度学习技术之LSTM模型,对海量前兆数据进行学习,学习出其短期、中期、长期的变化规律,建立前兆数据深度模型,并根据模型预测出前兆波形。论文将预测的前兆波形与实际的观测的波形数据进行对比,实验表明,LSTM算法能很好地拟合观测数据。Earthquake precursor data have short-term,medium-term and long-term changes.Studying the change rule of earthquake precursor data is of great significance for earthquake prediction.This paper uses the most popular LSTM model of deep learning technology to learn the massive precursory data,learn its short-term,medium-term and long-term change rules,establish the depth model of precursory data,and predict the precursory waveforms according to the model.In this paper,the predicted precursory waveform is compared with the actual observed waveform data.Experiments show that the LSTM algorithm can well fit the observation data.

关 键 词：地震前兆数据 LSTM 数据拟合 时间序列分析 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]