A Deep Learning Framework in FCC Process Control

The controlling of exhaust gas from gasoline is crucial for atmospheric environment protection. Research Octane Number (RON) loss and restricted sulphur (S) content matter the quality of gasoline. To obtain gasoline with high quality, the paper proposes a novel data-driven optimization model integrating deep neural network (DNN) and genetic algorithm (GA) to model for Fluid Catalytic Cracking (FCC) process then optimize. To begin with, the DNN is used to fit the relations between 13 related input variables and output variables in FCC. Subsequently, the FCC process is modelled and GA is proposed to solve the optimization model. Ultimately, 305 samples from real datasets have been analysed to testify feasibility and effectiveness of the method. This paper provides a guideline for the production process of FCC gasoline.

Liquid deposition modification of nano-ZSM-5 zeolite and catalytic performance in aromatization of Hexene-1

The Olefin aromatization is an important method for the upgrade of catalytic cracking (FCC) gasoline and production of fuel oil with high octane number. The nano-ZSM-5 zeolite was synthesized via a seed-induced method, a series of modified nano-ZSM-5 zeolite samples with different Ga deposition amount were prepared by Ga liquid deposition method. The XRD, N2 physical adsorption, SEM, TEM, XPS, H2-TPR and Py-IR measurements were used to characterize the morphology, textural properties and acidity of the modified ZSM-5 zeolites. The catalytic performance of the Hexene-1 aromatization was evaluated on a fixed-bed microreactor. The effects of Ga modification on the physicochemical and catalytic performance of nano-ZSM-5 zeolites were investigated. The Ga species in the modified nano-ZSM-5 zeolites mainly exist as the form of Ga2O3 and GaO+, which provide strong Lewis acid sites. The aromatics selectivity over Ga modified nano-ZSM-5 zeolite in the Hexene-1 aromatization was significantly increased, which could be attributed to the improvement of the dehydrogenation activity. The selectivity for aromatics over the Ga4.2/NZ5 catalyst with suitable Ga deposition amount reached 55.4%.

Multi-Objective Nonlinear Programming Model for Reducing Octane Number Loss in Gasoline Refining Process Based on Data Mining Technology

To simultaneously reduce automobile exhaust pollution to the environment and satisfy the demand for high-quality gasoline, the treatment of fluid catalytic cracking (FCC) gasoline is urgently needed to minimize octane number (RON) loss. We presented a new systematic method for determining an optimal operation scheme for minimising RON loss and operational risks. Firstly, many data were collected and preprocessed. Then, grey correlative degree analysis and Pearson correlation analysis were used to reduce the dimensionality, and the major variables with representativeness and independence were selected from the 367 variables. Then, the RON and sulfur (S) content were predicted by multiple nonlinear regression. A multi-objective nonlinear optimization model was established with the maximum reduction in RON loss and minimum operational risk as the objective function. Finally, the optimal operation scheme of the operating variable corresponding to the sample with a RON loss reduction greater than 30% in 325 samples was solved in Python.

Achieving deep desulfurization with inverse-micellar polyoxometalates and oxygen

(CTA)PMo4V8 exhibited excellent activity in treating S-containing compounds (dibenzothiophene (DBT), 4,6-dimethyldibenzothiophene (DMDBT), benzothiophene (BT) and thiophene) in real oils, i.e. diesel and FCC gasoline.

Synthesis of Ni(II)-phosphotungstic acid/nanocrystalline HZSM-5 catalysts for ultra clean gasoline by single-stage reactor

Although considerable progress has been made in FCC gasoline hydrorefining, it is still a great challenge to develop new catalysts and single-stage reactor to realize deep refining and keep octane...

Niobium oxide (Nb2O5) as support for CoMo and NiW catalysts in the hydrodesulfurization reaction of 3-methylthiophene

The efficiency of niobium oxide as catalytic support of hydrodesulfurization (HDS) catalysts (CoMo and NiW) has been investigated in the HDS of a model molecule representative of sulfur compounds present in FCC gasoline (3-methylthiophene: 3MT). The NiW catalyst presented higher catalytic activity than CoMo calcined and non-calcined catalyst, however a better ratio pentane/pentene has been achieved by CoMo catalysts, which implies a lower formation of hydrogenated products. Indeed, the activity order for the catalysts evaluated is: NiW/Nb2O5 > CoMo/Nb2O5 calcined support > CoMo/Nb2O5 non-calcined support, despite the ratio pentane/pentene which has the inverse order. Furthermore, textural and chemical characterization techniques have been performed. From NH3-TPD analysis it was observed an acidity profile with a predominance of weak/strong and weak/moderate acid for CoMo and NiW catalysts, respectively. Meanwhile, the BET analysis has shown a low specific surface area for the catalysts supported by niobium oxide. Concerning the structure characteristic, the XRD analysis has suggested an amorphous phase in all catalysts analyzed.