Neural network/particle swarm method to predict flammability limits in air of organic compounds

A new neural network based on particle swarm optimization was developed to estimate lower and upper flammability limit percent in air of organic compounds using their molecular structures. The capability of this hybrid method was tested using 328 compounds in a correlated set and 90 compounds in a predicted set. The results show that the proposed method can estimate the flammability limits of organic compounds with low deviations and can be used with acceptable accuracy in thermal engineering.