Open AccessC-Mn yield forecasting model based on SVM
Author(s) -
Jinghang Li,
Jiao Bai,
Bozhong Yu
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/781/2/022024
Subject(s) - autoregressive integrated moving average , support vector machine , yield (engineering) , series (stratigraphy) , time series , smelting , computer science , data mining , machine learning , artificial intelligence , metallurgy , materials science , paleontology , biology
With the rapid development of economy, how to increase steel output and reduce plant cost is of great significance. Deoxy alloying is an important link in iron and steel smelting. This paper aims at the optimization of the burdening scheme of deoxy alloying of molten steel. The aim is to establish the SVM optimization model based on the data and predict the yield of C/Mn. First of all, we established a time series model based on the obtained results of the problem and the furnace times as the time, and used the ARIMA time series method in SPSS software to predict the yield of C/Mn. Secondly, we improve the model, use the historical data to train and forecast the support vector machine model (SVM), and compare the predicted values of the SVM model and the time series model, and get the conclusion that the SVM model is more accurate than the time series model.