(596b) Optimal Control of Chemical Looping Reactors for Process Intensification of Natural Gas Upgrading

Upgrading natural gas to valuable aromatics locally at the extraction sites has proven to be economical primarily because this saves the otherwise high transportation costs. Nonoxidative dehydroaromatization (DHA), which only involves methane as a reactant, is a promising approach for converting natural gas into valuable aromatics [1]. However, the thermodynamics of DHA severely limits aromatics conversion efficiency. Such thermodynamic constraints can be overcome with a technique known as chemical looping. In this presentation, we will focus on the reactor structure, modeling, and control of chemical looping processes.

Chemical looping has been proposed as an effective technique for process intensification in the DHA process through the reactive separation of hydrogen, which shifts the thermodynamic equilibrium of the process and increases the yield of aromatics conversion [1]. The DHA reaction produces a mixture of methane, aromatics, and H₂, following which H₂ is selectively oxidized using a solid oxygen carrier (reducible metal oxides) to produce a reduced oxide (reduction phase) and water. The reduced oxide is re-oxidized (oxidation phase) using steam to produce H₂, thus, effectively separating H₂ from the DHA effluent. A temperature-swing adsorber is used to remove the water during the adsorption phase, and the adsorbent is regenerated during the desorption phase. This process, which is repeated to convert methane into aromatics, is carried out in multiple interconnected packed bed reactors. An optimization-based process control strategy is proposed to maximize the aromatics conversion efficiency while minimizing the operating cost in the face of the uncertainty.

A hierarchical control system implementation scheme typically used for industrial processes is adopted for the looping process. This scheme is chosen because it makes possible the design of a controller that will keep the process at desired operating conditions that simultaneously meet the economic objectives [2]. For this purpose, a first principles-based partial differential equation model for packed bed reactors is developed to model the chemical looping process. However, because the looping process exhibits cyclic behavior and never quite reaches a steady-state, the dynamics of looping reactors are different from the regular dynamics of processes that achieve steady-state conditions. Hence, "off-the-shelf" techniques for controller design may not be appropriate for such processes. In this presentation, we will also discuss the challenges faced during the design and testing of the control system and the approaches employed to deal with them effectively.

References

1. Brady, C., Murphy, B. and Xu, B., 2017. Enhanced Methane Dehydroaromatization via Coupling with Chemical Looping. ACS Catalysis, 7(6), pp.3924-3928.
2. Tatjewski, P., 2010. Supervisory predictive control and on-line setpoint optimization. International Journal of Applied Mathematics and Computer Science, 20(3), pp.483-495.