(430g) New Directions on Process Operability: Bilevel and Parallel Programming Approaches for Process Intensification and Modularity | AIChE

(430g) New Directions on Process Operability: Bilevel and Parallel Programming Approaches for Process Intensification and Modularity


Lima, F. V. - Presenter, West Virginia University
Carrasco, J. C., West Virginia University
Vinson, D. R., Tufts University
Process operability emerged in the last decades as a powerful tool for the design and control of chemical processes, including a variety of academic and industrial-scale applications from companies such as Air Products and Chemicals (APCI) and DuPont (Lima et al., 2010; Georgakis et al., 2003). In particular, process operability was proposed as a method for verifying the ability of a process design, defined by the available input set (AIS), to reach an achievable output set (AOS) that considers production targets (Vinson and Georgakis, 2000). Recent efforts in process operability have been focused on the calculation of the desired input set (DIS) for process design and intensification of natural gas utilization applications described by nonlinear models. Among the applications addressed were a catalytic membrane reactor (MR) for the direct methane aromatization (DMA) conversion to hydrogen and benzene, as well as a natural gas combined cycle (NGCC) system for power generation (Carrasco and Lima, 2017a, b, c). These past results indicate that operability can be a tool for enabling process intensification of nonlinear chemical and energy processes, facilitating the realization of the concept of modular manufacturing. However, there is still a gap in terms of problem dimensionality that nonlinear operability methods can handle.

In this presentation, a historical perspective on process operability will be given including how the past concepts have evolved to address the current challenges and potential future directions. In particular, the incorporation of bilevel and parallel programming approaches into the classical process operability concepts will be discussed. Bilevel programming enables the formulation of a hierarchical structure for combining design and intensification objectives in one step. Additionally, parallel computing is proposed as the key to tackle the challenge of computational time limitation. Such approaches provide a systematic framework for process intensification of chemical and energy systems towards modularity. Modular systems have the potential to transform the US economy, by utilizing local stranded gas without the need of transporting it from the exploration areas to the utilization plants. However, the design and intensification of modular energy systems is a challenging task as these processes are represented by complex and nonlinear models, with large numbers of differential and algebraic equations that demand high computational effort for their solution. The proposed operability approach addresses this dimensionality gap by employing parallel programming tools.

Preliminary results on the implementation of the proposed method show a reduction in computational time up to 2 orders of magnitude, when compared to the original results without parallelization. For parallel computing purposes, the MATLAB Distributed Computing Server at WVU with 63 workers (cores) is employed. In terms of intensification, the results for a 4-D DMA-MR system show that different reactor designs with similar levels of performance may have significantly different process footprints (up to 90%) in terms of reactor volume. Also, a modular NGCC power plant design with a dramatic reduction in size, from 400 to 0.5 [MW], is produced by specifying conditions of the gas and steam turbine cycles, while still keeping the high net plant efficiency of 55 [%]. The current research results open new avenues for operability future research, such as: i) formulation of an integrated framework for the design, intensification and control of modular energy systems; ii) incorporation of dynamic operability ideas (Georgakis et al., 2003) for intensification and drug administration; iii) development of new operability indicators; and iv) employment of stochastic tools to consider design and control under uncertainties.


Carrasco J. C. and Lima F. V. “Novel operability-based approach for process design and intensification: Application to a membrane reactor for DMA”. AIChE J. 2017a; 63(3): 975-983.

Carrasco J. C. and Lima F. V. “An optimization-based operability framework for process design and intensification of modular gas utilization systems”. Accepted for publication in Comput. Chem. Eng. 2017b; DOI: CACE-5645.

Carrasco J. C. and Lima F. V. “Operability-based approach for process design, intensification, and control: application to high-dimensional and nonlinear membrane reactors”. In proceedings of the FOCAPO/CPC 2017c.

Georgakis C., Uzturk D., Subramanian S., and Vinson D. R. "On the operability of continuous processes". Control Eng. Prac. 2003; 11(8): 859–869.

Lima F. V., Georgakis C., Smith J. F., Schnelle P. D., and Vinson D. R. "Operability-based determination of feasible control constraints for several high-dimensional nonsquare industrial processes". AIChE J. 2010; 56(5): 1249-1261.

Vinson D. R. and Georgakis C. "A new measure of process output controllability". J. Proc. Cont. 2000; 10(2-3): 185-194.