作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (17): 225-227. doi: 10.3969/j.issn.1000-3428.2007.17.077

• 人工智能与识别技术 • 上一篇    下一篇

基于CRBF神经网络的分类算法及应用

吕林涛1,李 翠1,白晓东2   

  1. (1. 西安理工大学计算机学院,西安 710048;2. 横河西仪有限公司系统部,西安 710075)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-09-05 发布日期:2007-09-05

Classification Algorithm and Application of Neural Network Based on Cosine Radial Basis Function

LV Lin-tao1, LI Cui1, BAI Xiao-dong2   

  1. (1. Institute of Computer, Xi’an University of Technology, Xi’an 710048; 2. System Dept., Yokogawa Xiyi Co., Ltd., Xi’an 710075)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-05 Published:2007-09-05

摘要: 通过对CRBF的分析研究,提出一种基于CRBF神经网络的分类算法。采用数学模型和几何模型构造其应用模型,通过分类算法的训练过程修改应用模型中的相关参数,使得分类结果更趋合理。通过CENTUM3000和Visual Basic6.0平台开发了化工厂爆炸监控系统。实践表明,分类结果与监控设备运行结果吻合得很好,满足了工厂监控系统的实际需求,证明该分类算法和应用模型具有较高的理论和实用价值。

关键词: 神经网络, CENTUM CS3000, 径向基函数, OPC

Abstract: Based on the analysis and study of cosine radial basis function(CRBF), a classification algorithm of neural network based on cosine radial basis function is proposed. Application model is constructed by mathematical model and geometric model, and the parameters of application model are modified through the training process of classification algorithm to make the classification result more reasonable. Classification result and supervision equipment are consistent after application of blast supervision system based on CENTUM CS3000 and Visual Basic 6.0, the factual demand is satisfied. The classification and application model are proved to be highly effective and valuable in practice and academic study.

Key words: neural network, CENTUM CS3000, radial basis function, OPC

中图分类号: