摘要
本文试图将模糊人工神经元网络及模糊信息分析理论引入到余震区长度与震级关系的识别之中。采用信息扩散与ＢＰ型人工神经元网络相结合的方法建立模型，该模型有较强的自适应能力及处理矛盾样本的功能。最后将该模型的识别结果与统计结果进行了比较，结果表明该模型具有一定的优越性。
By using information diffusion method combined with BP neural network,the authorssuggest a kind of fuzzy neuron in present paper.The model of relationship between length ofaftershock area and magnitude is discussed by the fuzzy neuron. This model has strong adapt-ability and can treat contradictory samples. Comparison of the results of this model with thoseof statistical method shows the advantage of the model.
出处
《西北地震学报》
CSCD
1995年第1期62-68,共7页
Northwestern Seismological Journal
关键词
模糊识别
信息处理
余震
震级
神经网络模型
Fuzzy identification,Information processing,Aftershock, Magnitude,Model
