Abstract

Optimal heat flux distributions along the axial position of steam reformer tubes are analyzed, aiming to maximize the outlet methane conversion without violating the upper bounds specified for the tube skin temperature, Tw(z) and the local heat flux, Q(z). The possible occurrence of the carbon deposition phenomenon is considered as an additional constraint in the optimization problem by means of a kinetic criterion which accounts for the rate of carbon formation (rC,net).The problem of finding the axial heat flux profile that leads to maximum methane conversion (or minimum tube skin temperature for specified production) without violating the upper limits imposed for Tw(z), Q(z) and rC,net is in principle a complex optimization process. In this work, a relatively simple semi-analytical method, that requires only iterative reactor simulations, is proposed.At conditions of fresh catalyst and for operations with typical feed compositions and temperatures, the carbon formation constraint is not active. For low values of the maximum allowable local heat flux (Qall), the optimal heating policies are monotonically increasing Tw(z) trajectories. Conversely, for higher Qall values the optimal Tw(z) shows an initial increase in the first tube section followed by an isothermal section. Decreasing axial wall temperature profiles or Tw(z) curves with maximum are clearly not optimal.When the catalyst is strongly deactivated, the optimal manipulation of Q(z), or Tw(z), is an appropriate procedure to achieve carbon free conditions. As this heating policy consists basically in reducing the firing in the top section of the reformer tube, it leads to unavoidable production losses.The shape of the catalyst activity axial profile has a considerable influence on the risk of carbon formation. Increasing activity distributions (similar to those found when sulfur poisoning takes place) may result in severe carbon formation in the reformer top. An optimal control of the heat input can contribute to minimize this practical operation problem.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call