人工神经网络在地震中期预报中的应用  

The application of artificial neural networks to mid-term earthquake prediction

作　　者：王炜[1] 宋先月[1] 王琤琤[2] 

机构地区：[1]上海市地震局,上海200062 [2]江苏省地震局,江苏南京210014

出　　处：《地震》2000年第1期10-16,共7页Earthquake

基　　金：中国地震局"九五"科研攻关项目!95-04

摘　　要：将ＢＰ神经网络用于地震中期预报。使用一些常用的地震学指标作为神经网络的输入，而将ＢＰ神经网络的输出作为表征地震活动增强的特征参数Ｗ１，并将其用于华北地区进行空间扫描。结果表明，中强地震前１～３年未来震中周围通常出现明显的Ｗ１值中期异常区，该方法具有较好的中期预报效果。BP neural network is used in mid-term earthquake prediction. Some usual pre- diction parameters of seismology are used as input units of neural networks. And the output units of neural networks are called as the characteristic parameter W, describing enhancement of seismicity.It is used for spatial scanning in North China. The result shows that the mid-term anomaly zone of W1-value usually appears obviously around the epicen- ter about 1-3 years before future earthquakes with good prediction effect.

关 键 词：BP神经网络 非线性关系 地震预报 中期预报 

分 类 号：P315.7[天文地球—地震学]