The End-Point Ingredient Prediction of Low Carbon Ferrochrome Smelt Based on the Working Condition
Construct the model of the end-point ingredient prediction of low carbon ferrochrome smelt based on the working condition of electrothermal silicon method using the method of multi-scale support vector machiness information fusion, where the best decomposition scale information is according to different smelt working conditions using Levenberg-Marquart algorithm to optimize the design, smelt working condition is judged by Bayesian classifier. Researches have proved that this method can improve the precision of prediction and make the prediction result more accurate, reasonable and practical.
N. N. Zhang et al., "The End-Point Ingredient Prediction of Low Carbon Ferrochrome Smelt Based on the Working Condition", Applied Mechanics and Materials, Vols. 128-129, pp. 1246-1249, 2012