A Simple Method for Estimating Gasoline, Gas, and Coke Yields in FCC Processes

Abstract

In this work we propose a simple method to estimate gasoline, gas, and coke yields in the catalytic cracking process. The method requires only experimental information about the variation with time of product yields, which are correlated using a third-order polynomial. Combined cracking and decay constants in the 3-and 4-lump kinetic models reported in the literature are estimated by linear regression analysis using experimental data and the proposed methodology. The kinetic model differential equations were solved using a fourth order Runge-Kutta method in order to evaluate FCC product yield-conversion relationships. The proposed methodology gives accurate predictions of product yields in the FCC process with average deviations less than 3% with respect to experimental data.

Five-lump kinetic model with selective catalyst deactivation for the prediction of the product selectivity in the fluid catalytic cracking process

“…3, where the relationship between the experimental and predicted yields wt% to each lump versus conversion wt% is presented. For the case of non-selective catalyst deactivation, when all products are similarly affected by catalyst deactivation, the model responses should appear as polynomial functions of conversion, as proposed by Ancheyta-Juarez and Murillo-Hernandez [26], who developed simple relations of the product yields with the conversion or the time-on-stream. A slight deviation of the model from this rule is evident in Fig.…”

Effect of NaOH on asphaltene transformation in supercritical water

“…The variations of yields (asphaltene, gas + maltene and coke) with time can be calculated using the following polynomial functions [31]:…”

The inverse methodology of parameter estimation for model adjustment, design, simulation, control and optimization of fluid catalytic cracking (FCC) risers

“…In spite of all simplifying hypotheses, the two-dimensional fluid flow model consists of a more realistic and accurate model than the often used onedimensional plug-flow model. 4,7,14,16 The mass and momentum conservation equations are written as follows: In Equations (2) and (3), ρ, the mixture average fluid density is considered a function of the volumetric fraction of catalyst, gasoil and steam according to the input conditions of the riser (pressure and temperature). It is therefore assumed to be a fixed parameter in the model and calculated as reported by Souza et al 13 The mixture average viscosity is calculated by:…”

“…However, the problem with this methodology is that the kinetic constants are obtained empirically, as a function of the feedstock and catalyst properties. 4 Therefore, the kinetic model created using these constants will not be capable of predicting, with good quantitative accuracy, the mass fractions of products formed for plants operating under different conditions with a different feedstock/catalyst set from those in which the constants were originally calculated. Qualitative accuracy is still expected with respect to the variation of operating conditions and feedstock/catalyst set, since the reactions and physical phenomena remain similar.…”

“…14,16 The second application encompasses models used when the main goal is to study a specific physical phenomenon of gasoil catalytic cracking, such as analysis of a catalyst deactivation function, coke formation in the cracking process, adsorption phenomena in the catalytic cracking and the building of kinetic models. In these cases, it is preferable to have a simple and easy-to-solve mathematical formulation for the fluid flow (normally plug-flow), 4,7 allowing concentration of the analysis effort on the phenomenon of interest. The third type of model strives for plant optimization.…”