(212c) Impact of Energy Integration On Dynamics and Control of Process Networks

Integration plays an important role in increasing the energy efficiency of chemical processes, and the design and optimization of energy integrated process systems has been an area of active research for the past decades. The economic benefits that result from energy integration come, however, at the expense of an intricate dynamic behavior at the plant level which, in turn, lies at the origin of distinct control challenges. The `energy feedback' inherent to such designs results in complex, nonlinear plant dynamics. Furthermore, process integration typically results in a reduction of the number of degrees of freedom available for control. Analyzing the impact of energy integration on the dynamics and control of process systems is therefore an imperative requirement. In our prior work [1,2], we have considered two distinct categories of integrated process systems, namely, process networks with large energy recycle and process networks with high energy throughput, demonstrating that energy integration alters the dynamic behavior at the plant level, and proposed a control framework that accounts for the dynamics induced by integration.

In the present paper, we consider a generic prototype process, which captures the structural and dynamic properties of various energy integrated process networks, such as reactor-feed effluent heat exchangers, heat integrated distillation columns, multiple effect evaporators, etc (including the ones considered in [1,2]). Specifically, we focus on the case of tight energy integration, whereby large amounts of heat are typically exchanged at small temperature gradients and/or transported by material streams with large flowrates. The presence of energy flows of different orders of magnitude in such networks is shown to lead to a multi-scale dynamic behavior. We employ elements from singular perturbation theory to derive reduced-order models in the different time scales. Furthermore, we rely on the resulting non-stiff, nonlinear models to introduce a cadre for control system design which naturally accounts for the time scale multiplicity of the class of processes considered.

Two illustrating examples, a double effect distillation column configuration and a solid oxide fuel cell (SOFC)-reformer system, are considered. The double effect distillation system shows a two-time-scale dynamic behavior with energy balance variables evolving in the fast time scale and material balance variables evolving in the slow time scale (characteristic of the class of high energy throughout networks). Exploiting this time scale multiplicity, we propose a hierarchical control strategy specifically addressing the control of temperatures in the fast time scale and the control of compositions in the slow time scale. Along similar lines, the dynamic analysis and a subsequent hierarchical control system design are addressed for the SOFC-reformer system, which belongs to the class of high energy recycle. Simulation case studies are presented to demonstrate the effectiveness of the proposed controllers.

References:

[1] M. Baldea and P. Daoutidis, Modeling, dynamics and control of process networks with high energy throughput, Comput. Chem. Eng. 32 (2008) 1964-1983.

[2] S. S. Jogwar, M. Baldea, P. Daoutidis, Dynamics and control of process networks with large energy recycle, Ind. Eng. Chem. Res. Article ASAP (2009) DOI: 10.1021/ie801050b.