A modified scaled variable reduced coordinate (SVRC)-quantitative structure property relationship (QSPR) model for predicting liquid viscosity of pure organic compounds

Seongmin Lee, Kiho Park, Yunkyung Kwon, Tae Yun Park, Dae Ryook Yang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Liquid viscosity is an important physical property utilized in engineering designs for transportation and processing of fluids. However, the measurement of liquid viscosity is not always easy when the materials have toxicity and instability. In this study, a modified scaled variable reduced coordinate (SVRC)-quantitative structure property relationship (QSPR) model is suggested and analyzed in terms of its performance of prediction for liquid viscosity compared to the conventional SVRC-QSPR model and the other methods. The modification was conducted by changing the initial point from triple point to ambient temperature (293 K), and assuming that the liquid viscosity at critical temperature is 0 cP. The results reveal that the prediction performance of the modified SVRC-QSPR model is comparable to the other methods as showing 7.90% of mean absolute percentage error (MAPE) and 0.9838 of R2. In terms of both the number of components and the performance of prediction, the modified SVRC-QSPR model is superior to the conventional SVRC-QSPR model. Also, the applicability of the model is improved since the condition of the end points of the modified model is not so restrictive as the conventional SVRC-QSPR model.

Original languageEnglish
Pages (from-to)1-10
Number of pages10
JournalKorean Journal of Chemical Engineering
DOIs
Publication statusAccepted/In press - 2017 Jul 25

Keywords

  • Liquid Viscosity
  • Modeling
  • Molecular Descriptor
  • Property Estimation
  • QSPR
  • SVRC

ASJC Scopus subject areas

  • Chemistry(all)
  • Chemical Engineering(all)

Fingerprint Dive into the research topics of 'A modified scaled variable reduced coordinate (SVRC)-quantitative structure property relationship (QSPR) model for predicting liquid viscosity of pure organic compounds'. Together they form a unique fingerprint.

  • Cite this