The Modeling Mechanism of Cascade-forward Back Propagation Neural Network Based on Self-correlation and Its Application

摘要：

In this paper, based on self-correlation, a new method of modeling mechanism of cascade-forward back propagation neural network is put forward. First, the original sequence is divided into sub-sequences according to the prominence of self- correlative coefficients. Then,the framework of neural network is structured and input and output are defined reasonably; and then ameliorating BP algorithm with momentum factor is studied. Finally the method presented in this paper is applied in building the model of total residence number and the results show the model has very high precision.

关键词： Self-correlation Theory Feed-forward Neural Network Modeling and Forecasting

作者: Wenzhan Dai Junfeng Li

作者单位: College of Information and Electronic Engineering, Zhejiang Gongshang University Hangzhou, China Department of Automatic control, Zhejiang Sci-Tech University Hangzhou, China;College of Mechanical

会议类型: 国际会议

会议名称: 2006 International Symposium on Distributed Computing and Applications to Business,Engineering and Science(2006年国际电子、工程及科学领域的分布式计算应用学术研讨会)

会议地点: 杭州

会议语种:英文

页码: 727-730

在线出版日期: 2006-10-12（万方平台首次上网日期，不代表论文的发表时间）