首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于专家知识的神经元网络方法及其在潜在震源区定量划分中的应用
引用本文:胡银磊,张裕明.基于专家知识的神经元网络方法及其在潜在震源区定量划分中的应用[J].中国地震,1996,12(3):261-268.
作者姓名:胡银磊  张裕明
作者单位:[1]香港大学土木与结构工程系 [2]中国北京100029国家地震局地质研究所
摘    要:将专家知识和神经元网络方法结合起来,利用专家知识和规则构造规则样本的方法来提高神经元网络学习样本的质量,通过神经元网络对规则样本的学习,形成基于专家知识的神经元网络模型。在潜在震源区划分专家系统中有关规则的基础上,将基于专家知识的神经元网络方法引入到潜在震源区定量划分中,并以首都圈地区为例,对其潜在震源区作了定量划分,结果表明,规则样本能较好的反映专家的知识和规则,利用构造规则样本的方法,可以提高

关 键 词:潜在震源区  神经网络  专家知识  地震  定量划分

Expert Knowledge Based on Artificial Neural Network and Its Application to the Identification of Potential Seismic Source
Hu Yinlei, Zhang Yuming.Expert Knowledge Based on Artificial Neural Network and Its Application to the Identification of Potential Seismic Source[J].Earthquake Research in China,1996,12(3):261-268.
Authors:Hu Yinlei  Zhang Yuming
Abstract:In this paper, an approach is developed to elevate the quality of the learning samples in Artificial Neural Network (ANN) by integrating the expert knowledge and rules to make expert rule samples,through the training of using these samples, an expert knowledge based on ANN is further developed.The method is introduced into the field of identification of potential seismic source on the basis of the rules of an expert system. And the method is applied to the quantitative identification of potential seismic sources in Beijing and its adjacent area. The result indicates, the expert rule samples can well incorporate the expert knowledge, the quality of the samples, the efficiency of training and the accuracy of the result are elevated.
Keywords:Sample  Potential seismic source  Expert rule  ANN  
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号