Title :
A neural network based on mechanism model and it´s application in prediction of rolling force
Author_Institution :
Dept. of Autom., Taiyuan Heavy Machinery Inst., China
Abstract :
A synthetic method of neural network and mechanistic modelling is presented. This method merges the mechanistic model knowledge with the neural network structure. It greatly reduces the number of weights of the network. The learning speed, nonlinear approximation precision and generalization performance of the network are improved. The effect of application in rolling force prediction is excellent
Keywords :
backpropagation; generalisation (artificial intelligence); neural nets; rolling mills; generalization performance; learning speed; mechanism model; mechanistic model knowledge; neural network structure; nonlinear approximation precision; rolling force prediction; Automation; Intelligent networks; Machinery; Neural networks; Prediction algorithms; Predictive models; Sun;
Conference_Titel :
Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress on
Conference_Location :
Hefei
Print_ISBN :
0-7803-5995-X
DOI :
10.1109/WCICA.2000.863403
