Improved structure–property relationship models for prediction of critical properties

Godavarthy, Srinivasa S.; Robinson, Robert L.; Gasem, Khaled A. M.

doi:10.1016/j.fluid.2007.11.003

Cited by 51 publications

(61 citation statements)

References 93 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data used in the study of 7 were compiled from the multiple printed sources and their analysis is beyond the scope of this work. The data used by 11 Table 4 points to the source of the problem: both studies did not filter out nonexperimental data, and the data sets used contained about 60% to 70% of predicted data. Formally, this makes the problem even worse since, as seen in Table 4, addition of estimated data points greatly increases the overall data uncertainties, thus resulting in even greater degree of overfitting.…”

Section: ■ Resultsmentioning

confidence: 99%

Quantitative Structure–Property Relationship Predictions of Critical Properties and Acentric Factors for Pure Compounds

et al. 2015

View full text Add to dashboard Cite

Knowledge of critical constants and phase boundary pressure properties is essential to understanding thermodynamic behavior of substances and is often required in practical process design applications. Where critically evaluated data are unavailable, a quantitative structure−property relationship (QSPR) regression method can be used to relate molecular properties (descriptors) to properties of interest.The relationship is trained and tested using existing critically evaluated data and is dynamic; as new data become available, the relationship can be updated to reflect changes. In this work, we use support vector regression (SVR) to develop estimation methods for critical properties and acentric factors based on critically evaluated data for over 900 pure compounds. From threedimensional geometry and connectivity information, we calculate over 500 descriptors for each compound. A matrix of descriptor values defines the input vectors for SVR, whereas critically evaluated data for critical temperature, the ratio of critical temperature to critical pressure, and saturation reduced pressure form the targets. We determine optimal SVR parameters by minimizing the sum of absolute deviations between the SVR outputs and the target values. We use a genetic algorithm to find the Pareto front points that optimize the output fit while reducing the number of input vectors (descriptors). We use a single Pareto front point to make a final evaluation in SVR. To define uncertainties of predicted values, we use uncertainty propagation calculations based on a Monte Carlo method that employs Latin hypercube sampling.

show abstract

Section: ■ Resultsmentioning

confidence: 99%

Quantitative Structure–Property Relationship Predictions of Critical Properties and Acentric Factors for Pure Compounds

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Other models are also employed, for example non-linear [28] and complex models, such as neuronal networks [29,30]. In the latter, good adjustment is obtained with high correlation factors; nevertheless, for simplicity, linear correlations are only used here.…”

Section: Qspr Model Development and Selection Criterionmentioning

confidence: 99%

“…In the latter, good adjustment is obtained with high correlation factors; nevertheless, for simplicity, linear correlations are only used here. Note, however, the best descriptor set obtained with linear regression is normally used in non-linear cases [28].…”

Section: Qspr Model Development and Selection Criterionmentioning

confidence: 99%

A QSPR strategy to select models with extrapolative ability – Critical temperatures of linear and aromatic hydrocarbons

Peraza

Rojas

Bolívar

et al. 2014

JCM

View full text Add to dashboard Cite

This paper deals with the problem of evaluating the predictive and extrapolative ability of QSPR models by using the PaDEL-descriptor software. In this sense, the selections of training and external data sets were modified considering the molecular weight. Two criteria were used to evaluate the extapolative ability: high correlation factor (R 2 ) (criterion I) and positive ΔR 2 and high R 2 (criterion II). Based on internal and external validation, it is shown that criterion II has a better performance than criterion I. Other selection criteria were found by considering the maximum square correlation coefficient (Q 2 ext ) or the minimum standard deviation (σext) for the external set with high R 2 values. These facts are supported by a systematic variation of the correlation factor (VCF) and the variation of correlation coefficients (VCC), as analysis tools proposed in this article. The methodology was successfully applied to critical temperature (Tc) estimation of linear alkanes and aromatic compounds, considering extrapolation to the heaviest compounds. Descriptors obtained for studied cases, using the criterion II, are in some way in agreement with group contribution and QSPR methods from literature.

show abstract

“…The reason for selecting critical pressure as a target predicted property is that it is important to engineers since fluids at temperatures and pressures above their critical point become supercritical fluids whose properties are totally different compared to those of the fluids in their normal state. Thus, many researchers have attempted to estimate critical pressure by various methods (Daubert and Bartakovits, 1989;Godavarthy et al, 2008;Wang et al, 2008).…”

Section: Introductionmentioning

confidence: 99%

A Quantitative Structure-Property Relationship Model for Predicting the Critical Pressures of Organic Compounds Containing Oxygen, Sulfur, and Nitrogen

Park

Kim³

et al. 2017

J. Chem. Eng. Japan / JCEJ

View full text Add to dashboard Cite

In the present research, the critical pressures of organic compounds were selected as a model case and were predicted using a quantitative structure-property relationship model. The coverage of prediction contains hydrocarbons and nonhydrocarbon organic compounds containing O, S, and N atoms. In total, 802 hydrocarbons and 1144 non-hydrocarbon organic compounds were used to develop a model with the 3D structure of each compound being optimized by quantum mechanical calculations. Furthermore, appropriate descriptors to explain critical pressure e ectively were selected by forward selection regression and genetic algorithm. Multi-linear regression and neural networks were used to establish prediction models for the hydrocarbon and non-hydrocarbon organic compounds. The prediction models achieved su ciently high performances, with R 2 >0.96. This research also analyzes implications of selected descriptors, and the relationship between the descriptor and critical pressure.

show abstract

Improved structure–property relationship models for prediction of critical properties

Cited by 51 publications

References 93 publications

Quantitative Structure–Property Relationship Predictions of Critical Properties and Acentric Factors for Pure Compounds

Quantitative Structure–Property Relationship Predictions of Critical Properties and Acentric Factors for Pure Compounds

A QSPR strategy to select models with extrapolative ability – Critical temperatures of linear and aromatic hydrocarbons

A Quantitative Structure-Property Relationship Model for Predicting the Critical Pressures of Organic Compounds Containing Oxygen, Sulfur, and Nitrogen

Contact Info

Product

Resources

About