Artificial intelligence modeling of DME conversion to gasoline and light olefins over modified nano ZSM-5 catalysts. (1st September 2016)

• Record Type: Journal Article
• Authors: Hajjar, Zeinab; Khodadadi, Abbasali; Mortazavi, Yadollah; Tayyebi, Shokoufe; Soltanali, Saeed
• Is Part Of: Fuel. Volume 179(2016)
• Page Start: 79
• Page End: 86
• Publication Date: 2016-09-01
• Subjects: Modified ZSM-5 -- DME conversion -- Neural network -- Genetic algorithm -- Modeling -- Optimization; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2016.03.046
• Languages: English
• ISSNs: 0016-2361

Abstract

Graphical abstract: Highlights: ZSM-5 are synthesized and modified with Fe and Cr to control its acidity. Impact of temperature, WHSV, DME percentage and catalyst acidity upon selectivity. The neural network is employed to model hydrocarbon selectivity. GA is applied for optimization of light olefins and gasoline-range selectivity. Abstract: The investigation of the effect of Fe and Cr promoters on HZSM-5 nanocatalyst in dimethyl ether (DME) conversion to gasoline has been carried out in this work. A model based on neural network is proposed for estimation of DME conversion to light olefins and gasoline-range hydrocarbons over ZSM-5 catalysts. The main operating variables affecting the hydrocarbon selectivity are temperature, WHSV (weight hourly space velocity), DME concentration and catalyst acidity, which are selected as input variables of the neural model. The predicated hydrocarbon selectivities are in very good agreement with the corresponding experimental values, with the correlation coefficients greater than 0.98. In addition, the genetic algorithm is employed to obtain the optimal selectivity values of light olefins and gasoline-range hydrocarbons and the appropriate operating conditions.