Prediction of the flash points of alkanes by group bond contribution method using artificial neural networks

Pan, Yong; Jiang, Juncheng; Wang, Zhirong

doi:10.1007/s11705-007-0071-z

Cited by 8 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequently, BP was combined with the group bond contribution and topology to predict the flash points of 44 alkanes and 40 fatty alcohols, respectively. 29,30 In general, the results showed that the predicted flash points are in good agreement with experimental data, with the absolute mean relative error being less than 2.25 %, which is superior to those of traditional group contribution methods. In 2007, Pan and co-workers 31 developed a model combining the group bond contribution method with the back-propagation (BP) neural network for 92 alkenes.…”

Section: Pure Component Flash Point Prediction Methodsmentioning

confidence: 64%

“…In 2007, Pan et al 28 applied the group contribution method in the back-propagation (BP) neural network model and used 32 kinds of molecular groups as input varieties for flash points of 258 organic compounds to determine the quantitative relation between molecular structure and flash point. Subsequently, BP was combined with the group bond contribution and topology to predict the flash points of 44 alkanes and 40 fatty alcohols, respectively.…”

Section: Pure Component Flash Point Prediction Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Research Progress on Flash Point Prediction

Liu

2010

J. Chem. Eng. Data

View full text Add to dashboard Cite

Flash point is one of the major physical and chemical properties used to determine the fire and explosion hazards of liquids; therefore, the prediction of flash points is an important safety consideration. In this paper, flash point prediction methods based on vapor pressure, molecular structure, composition range, and boiling point of flammable liquids are reviewed, respectively. Le Chatelier's rule and Antoine equations are used in the correlation between vapor pressure and flash point. Research on the correlations between flash point and composition range of the mixture has focused on flash point predictions for binary and ternary solutions, and further investigation for multicomponent solutions is required in the future. With regard to the correlation between molecular structure and flash point, some key breakthroughs have been made through quantitative structure-property relationship (QSPR) analysis. In particular, the artificial neural network (ANN) in QSPR is one possible correlation technique because of its nonlinear property and high accuracy and has the potential for wide application.

show abstract

Section: Pure Component Flash Point Prediction Methodsmentioning

confidence: 64%

Section: Pure Component Flash Point Prediction Methodsmentioning

confidence: 99%

Research Progress on Flash Point Prediction

Liu

2010

J. Chem. Eng. Data

View full text Add to dashboard Cite

show abstract

“…Albahri used structural group contribution method for predicting flammability characteristics such as auto‐ignition temperature, flash point, and the upper and lower flammability limits of pure hydrocarbon fluids . Yang et al presented a model for predicting the flash points of alkanes through a group bond contribution model using artificial neural networks . Rowley et al proposed a method for estimating the flash point of general organic compounds based entirely on structural contributions .…”

Section: Introductionmentioning

confidence: 99%

Two Reliable Simple Relationships between Flash Points of Hydrocarbon Kerosene Fuels and Their Molecular Structures

Zohari

Qhomi

2017

Zeitschrift anorg allge chemie

View full text Add to dashboard Cite

This study presents two new reliable simple correlations for predicting flash point of kerosene hydrocarbons using multiple linear regression method. The methodology assumes that the flash point of kerosene fuels can be expressed as a function of elemental composition and several structural parameters. The proposed correlations have determination coefficients of 0.910 and 0.977. Also, the first model has root mean square deviation (RMSD) and the average absolute deviations (AAD) of 10.6 and 8.2 K, respectively, for 111 kerosene fuels with different molecular structures as training set. The RMSD and AAD for the second improved model are 5.39 and 4.33 K, respectively. The predictive power of two correlations is checked using a cross validation method. (R2 = 0.977, Q2Ext = 0.975, and Q2LMO = 0.979). Also, these correlations give good predictions for further 25 kerosene fuels as test set. The proposed model can also be applied for designing novel kerosene fuels.

show abstract

“…The GCM models predict the FP as a function of the number and type of functional groups which constitute a compound . In most accurate GCM models, artificial neural networks are exploited to map the relationship between the functional groups and the FP . Artificial neural network is one of the most efficient machine learning based tools for mapping the linear and nonlinear dependencies between variables and has been extensively used to predict various properties .…”

Section: Introductionmentioning

confidence: 99%

A Neural Network QSPR Model for Accurate Prediction of Flash Point of Pure Hydrocarbons

Mirshahvalad

Ghasemiasl

Raoufi

et al. 2018

Molecular Informatics

View full text Add to dashboard Cite

The present study introduces a QSPR model to predict the flash point of pure organic compounds from diverse chemical families. We used the Maximum-Relevance Minimum-Redundancy (MRMR) as an efficient descriptor selection algorithm to select 20 the most effective out of 1926 calculated descriptors. The selected descriptors and their combination with the normal boiling point data were used as model inputs and their correlation with FP was mapped using feedforward artificial neural networks. Study-ing various models, the best result was obtained by a neural network with 2 neurons in the hidden layer for which a combination of the selected descriptors and normal boiling point data were used as model inputs. Evaluating the performance of this model for a dataset of 727 compounds resulted in average absolute relative errors of of 1.36 %, 1.34 %, 1.44 % and 1.42 % and average absolute deviations of 4.48, 4.41, 4.75 and 4.66 K for the overall, training, validation, and test datasets, respectively.

show abstract

Prediction of the flash points of alkanes by group bond contribution method using artificial neural networks

Cited by 8 publications

References 8 publications

Research Progress on Flash Point Prediction

Research Progress on Flash Point Prediction

Two Reliable Simple Relationships between Flash Points of Hydrocarbon Kerosene Fuels and Their Molecular Structures

A Neural Network QSPR Model for Accurate Prediction of Flash Point of Pure Hydrocarbons

Contact Info

Product

Resources

About