A robust nonlinear model of steam reformers with applications to energy efficiency

Abstract

We have developed and implemented a computational methodology to solve for the heat flow and the gas concentrations in a class of industrial catalytic steam reformers. The problem under consideration consists of a nonlinearly coupled system composed of a boundary-value problem and reaction-diffusion partial differential equations. Such reformers have broad industrial applications and consume a large amount of energy and natural resources globally. The methodology is applied to tackle two outstanding questions regarding catalytic processes, for which our methodology delivers efficient answers to the following research questions: (i) How to solve the energy/catalyst investment dichotomy, i.e., locations with relatively inexpensive energy require less catalyst and in the opposite case of high energy costs, more catalyst is needed. (ii) How to express the energy input, which is difficult to ascertain for large burner designs, in terms of easily measurable parameters in real time, such as the input temperatures. These questions are answered by our approach leading to a powerful design tool for catalytic processes in general, with huge potential to mitigate environmental impacts. The complex nonlinear system of reaction-diffusion equations is solved by means of an implicit-explicit method and the nonlinear boundary value problem is solved by an iterative method. The coupling between the two systems is also performed through an iterative method. The convergence of the latter is confirmed by extensive numerical investigations, which are consistent with observed measurements in industrial installations. Furthermore, an analysis of the syngas production with different catalyst density, reactor inlet temperature, furnace energy flame and catalyst length are performed, which confirm that the variation of these parameters can result in a more optimized process. A typical consequence is the following: Fixing all parameters, except the catalyst density and the reactor inlet temperature, we show that an increase of 100 K in the latter allows us to use 3 times less catalyst density to achieve the same target CO concentration. In other words, we have a quantifiable and substantial trade off between energy and catalyst investment.