End-point Temperature Prediction based on RBF Neural Network

Authors:	Deng Changhui, Key Laboratory of Process Industry Automation, Ministry of Education, Northeastern University, China Wang Shu, Key Laboratory of Process Industry Automation, Ministry of Education, Northeastern University, China Mao Zhizhong, Key Laboratory of Process Industry Automation, Ministry of Education, Northeastern University, China Wang Fuli, Key Laboratory of Process Industry Automation, Ministry of Education, Northeastern University, China
Topic:	6.2 Mining, Mineral & Metal Processing
Session:	Technology in Mining and Metal Processing Industry
Keywords:	neural network method,prediction,error correction, k-meansclustering,vacuum induction furnace

Abstract

An end-point temperature prediction model based on RBF neuralnetwork is developed to reduce the measuring cost and improve themeasuring accuracy in a vacuum induction furnace. It can givereliable predictions of tapping time and temperature of moltensteel in the first-round prediction. And the prediction accuracycan be improved by the error correction in the second-roundprediction. 120 set of data are used for model building andvalidation.The experimental results show that the proposed methodis effective and the real-world application ispotential.