QSPR and Artificial Neural Network Predictions of Hypergolic Ignition Delays for Energetic Ionic Liquids

摘要: Abstract : Due to their negligible volatility, energetic ionic liquids are being considered as next generation hypergolic fuels for replacing toxic monomethylhydrazine. One design challenge is maintain ignition delays close that of The process with an oxidizer, such nitric acid, a complex and, date, there no theoretical method predicting the delay. present work examined two correlation methods, Quantitative Structure Property Relationship (QSPR) and Artificial Neural Networks (ANNs), ability predict this quantity. A set five descriptors were chosen from pool more than 160 establish these correlations. good QSPR was obtained using descriptors. We also trained artificial neural network predictive extensive 5-fold cross validation same number data normalization techniques training validation. results show ANNs exhibit excellent prediction capabilities application.