Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4980903 | Process Safety and Environmental Protection | 2016 | 24 Pages |
Abstract
Due to the combustible nature of hydrocarbons, accurate safety information for their safe utilization and transport is essential. The autoignition temperature (AIT) is an important safety parameter. In this study, a quantitative structure-property relationship (QSPR) approach was utilized to present a predictive model for the estimation of the AIT of 813 hydrocarbons from 69 different chemical families. A unified three-parameter model was constructed combining multivariate linear regression (MLR) and a genetic algorithm (GA). In order to investigate the complex performance of the selected molecular parameters an optimized artificial neural network (ANN) model was developed. The resulting models showed good prediction ability. Although the ANN prediction results are more accurate than the GA-MLR model, the application of the latter is more convenient.
Keywords
Related Topics
Physical Sciences and Engineering
Chemical Engineering
Chemical Health and Safety
Authors
Tohid Nejad Ghaffar Borhani, Afsaneh Afzali, Mehdi Bagheri,