Prediction and extension of curves of distillation of vacuum residue using probability functions

3rd International Meeting for Researchers in Materials and Plasma Technology; 1st Symposium on Nanoscience and Nanotechnology

物理学;材料科学

León, A.Y.^1 ; Riaño, P.A.^1 ; Laverde, D.^1

Universidad Industrial de Santander (UIS), Bucaramanga, Colombia^1

关键词: Akaike information criterion;    Bayesian information criterion;    Characterization studies;    Compositional analysis;    Correlation coefficient;    Distillation curves;    Hydrocarbon mixture;    Probability functions;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/687/1/012093/pdf
DOI  :  10.1088/1742-6596/687/1/012093

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

The use of the probability functions for the prediction of crude distillation curves has been implemented in different characterization studies for refining processes. The study of four functions of probability (Weibull extreme, Weibull, Kumaraswamy and Riazi), was analyzed in this work for the fitting of curves of distillation of vacuum residue. After analysing the experimental data was selected the Weibull extreme function as the best prediction function, the fitting capability of the best function was validated considering as criterions of estimation the AIC (Akaike Information Criterion), BIC (Bayesian information Criterion), and correlation coefficient R2. To cover a wide range of composition were selected fifty-five (55) vacuum residue derived from different hydrocarbon mixture. The parameters of the probability function Weibull Extreme were adjusted from simple measure properties such as Conradson Carbon Residue (CCR), and compositional analysis SARA (saturates, aromatics, resins and asphaltenes). The proposed method is an appropriate tool to describe the tendency of distillation curves and offers a practical approach in terms of classification of vacuum residues.