Least Squares Support Vector Machine Classifiers
Neural Processing Letters
Time Series Analysis, Forecasting and Control
Time Series Analysis, Forecasting and Control
Choosing Multiple Parameters for Support Vector Machines
Machine Learning
Hi-index | 0.00 |
In this paper, a support vector machine (SVM) classifier is designed for predicting the energy consumption level of Ironmaking process. To improve the accuracy, particle swarm optimization (PSO) is introduced to optimize the parameters of SVM. First, the consuming structure of Ironmaking process is analyzed so as to accurately modeling the prediction problem. Then the improved SVM algorithm is presented. Finally, the experimental test is implemented based on the practical data of a Chinese Iron and Steel enterprise. The results show that the proposed method can predict the consumption of the addressed Ironmaking process with satisfying accuracy. And that the results can provide the enterprise with effective quantitative analysis support.