Journal of Applied Sciences ›› 2004, Vol. 22 ›› Issue (4): 509-512.

• Articles • Previous Articles     Next Articles

A Multiplex Forecasting Model Based onan Artificial Neural Network and Its Application

XUE Li-hua1, HUANG Hong-zhong1, SHI Zhang-ming2   

  1. Key Laboratory for Precision & Non-traditional Machining of Ministry of Education, Dalian University of Technology, Dalian 116023, China;
    2. School of Energy and Power Engineering, Centralsouth University, Changsha 410083, China
  • Received:2003-08-26 Revised:2003-12-15 Online:2004-12-31 Published:2004-12-31

Abstract: On the basis of our study of the industrial converting process as well as through comprehensive application of the BP neural networks and the experiential evaluation method, a multiplex forecast model is set up at the endpoint of copper blow period in a converting process. The fuzzy clustering anaysis method is also used to determine the number of hidden nodes in the neural networks. Experience shows that the forecast is highly dependable and capable of providing effective guidance to the production process.

Key words: artificial neural network, endpoint forecast, fuzzy clustering analysis, matte converting

CLC Number: