Prediction of Coke Strength Based on Multiple Linear Regression Analysis
Wu Xianxi
Abstract
Wu Xianxi
Abstract
Coke strength was predicted by selecting the indexes of blending coal quality.The indexes were used as independent variables for multiple linear regression analysis and got regression models.Then took regression coefficient to make t test,eliminated the no significant index and renewed to make the linear regression models.The research indicated that the model can be used to effectively predict coke strength and the relative error forecasting results was within 5%.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Coke strength was predicted by selecting the indexes of blending coal quality.The indexes were used as independent variables for multiple linear regression analysis and got regression models.Then took regression coefficient to make t test,eliminated the no significant index and renewed to make the linear regression models.The research indicated that the model can be used to effectively predict coke strength and the relative error forecasting results was within 5%.
Key concepts: Linear regression, Regression analysis, Coke, Proper linear model, Statistics, Regression, Regression dilution, Coefficient of determination