This work presents an approach for
the model-based design of a
load-flexible fixed-bed reactor for performing the partial water–gas
shift (WGS) reaction in the gas cleanup section of a coal-to-methanol
plant. Three optimization problems with various combinations of objective
functions (H2/CO ratio, pressure drop, and operating costs)
are formulated and solved using the nondominated sorting genetic algorithm
(NSGA-II) to obtain the Pareto-optimal fronts, and the selected solutions
are subjected to dynamic stability analysis. The optimum values of
decision variables such as the weight of the catalyst, catalyst particle
size, ratio of the reactor length and diameter, feed gas temperature,
and steam to CO ratio are estimated and are found to be dependent
on the choice of the objectives and tolerable deviation from their
required values. The dependence of the H2/CO ratio on the
inlet gas temperature varies with the choice and combination of the
objectives. In contrast, higher values of the steam to CO ratio (S/C) are associated with a lower deviation
from the desired H2/CO ratio for all three cases. However,
optimum values of S/C are found
to be below 1:1 because of its effect on the pressure drop and operating
costs. The solutions obtained in the equilibrium-limited regime are
found to be more stable in the presence of fluctuations in the feed
flow rates but show intermediate stability with fluctuations in the
inlet gas temperature. Four designs are recommended as the feasible
designs, and the criteria of their application are discussed. Even
though the work is related to a WGS reactor, the methodology used
can be applied for the design of any other reactor system operating
under variable feed conditions.