自组织人工神经网络在地震综合预报中的应用
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
本文运用T.Kohonen自组织人工神经网络进行地震综合预报,以云南多震区──滇西南地震区为例进行实际判别计算。预测成功率较高。结果表明,神经网络方法性能较好,可望成为地震实际预报的一种有效的辅助手段。
The application of T. Kohonen's self-organization artificial neural network to synthetic prediction of earthquakes had not been reported before. And an area strucked by In this paper the frequent earthquake area in Yannan Province, i. e. , the southwestern Yunnan seismic zone was selected as an example to use this method for practical distinguishing calculation.The successful rate is high.The results indicated that the performance of the neural network approach is good, and therefore it might be referred as an effective assistant technique for synthetic prediction of earthquakes.
引文
[1]石绍先,沈斌,聚类分析在地震综合预报中的应用,地震研究,9.1.55-62.1986。[2]王学仁,地质数据的多变量统计分析,113-129,科学出版社,北京,1982。[3]石绍先,范扬,多级判别在地震综合预报中的应用,地震研究,13.4.330-339.1990。[4]尹红风,戴汝南,模式识别与人工智能,3(1):1-12,1990。[5]Lippmann,R.P.,Anintroductiontocomputingwithneuralnets,IEEEASSPMagazine,4-22,1987。[6]Kohonen,T.,Self-OrganizitionandAssociativeMemory(2nded.),Berlin:Springer-verlag,1988。

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心