Title :
Research for breakout prediction system based on support vector regression
Author :
Qing, Tian ; Jia-Wei, Wang ; Ji-Shuang, Xue
Author_Institution :
Hebei United Univ., Tangshan, China
Abstract :
SVM is widely used in the pattern recognition. It shows prediction ability well. For the system nonlinear and complexity of the CCM bonding breakout forecast system nonlinear, complexity, and breakout forecast system based on the least squares support vector machine (LSSVM) is put forward. In forecast system, establish 0-1 more value data window to eliminate the redundant data. The simulation results show that the LSSVM model cans quickly the training sample parameters in the small sample. It shows strong recognition ability, high precision.
Keywords :
accident prevention; casting; industrial accidents; least squares approximations; nonlinear systems; pattern recognition; production engineering computing; regression analysis; support vector machines; CCM bonding breakout forecasting system; LSSVM model; breakout prediction system; breakout production accident; continuous casting production; data window; least squares support vector machine; nonlinear system; pattern recognition; redundant data elimination; support vector regression; system complexity; Bonding; Casting; Predictive models; Steel; Support vector machines; Temperature distribution; Temperature measurement; Breakout; LSSVM; Multi-value Data Window;
Conference_Titel :
Robotics and Applications (ISRA), 2012 IEEE Symposium on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4673-2205-8
DOI :
10.1109/ISRA.2012.6219155
