Herein, we investigate optimal flexible operations of an electrified biodiesel production process through both computer-aided approaches and experimental validation in a pilot plant. We develop a digital twin framework presenting modeling- and optimization-based methods and tools that guide the process system through key stages toward this objective.
We formulate offline dynamic optimization problems that incorporate flexibility-oriented process designs. Various process configurations with buffer tanks for storing intermediate and final products are examined and generalized as flexibility approaches for chemical plants targeting optimal dynamic operation. We demonstrate the dual role of buffer tanks, which not only enhance operational flexibility but also enable system decomposition for distributed optimization by decoupling process dynamics [4]. Building on these offline studies and flexibility-oriented process configurations, we implement real-time control applications. We leverage the process configuration that supports distributed optimization to develop and apply distributed economic nonlinear model predictive control. Our distributed control strategies incorporate both sequential and iterative communication architectures [5, 6], as well as compensation schemes for computational delays [7].
Furthermore, we employ advanced process analytical technologies—such as Raman spectroscopy [8]—for real-time monitoring and concentration measurements of key species, assessing their potential to support flexibility in chemical process operation.
We conduct experimental studies on a pilot plant that produces biodiesel via alkali-catalyzed transesterification of vegetable oil. To enable the implementation of the model-based advanced automation methods, we integrate tailored hardware and software solutions into the pilot plant.
Through the digital twin framework and pilot plant studies, our objective is to advance model-based control and inline process monitoring technologies toward industrial maturity. We aim to deliver broadly applicable methods and tools to support optimal flexible operations of chemical processes.
References
[1] Q. Zhang and I. E. Grossmann. Planning and Scheduling for Industrial Demand Side Management: Advances and Challenges. In M. Martín, editor, Alternative Energy Sources and Technologies: Process Design and Operation, pages 383–414. Springer International Publishing, Cham, 2016.
[2] A. Mitsos, N. Asprion, C. A. Floudas, M. Bortz, M. Baldea, D. Bonvin, A. Caspari, and P. Schäfer. Challenges in process optimization for new feedstocks and energy sources. Computers & Chemical Engineering, 113:209–221, 2018.
[3] D. Dering and C. L. Swartz. A scenario-based framework for the integration of scheduling and control under multiple uncertainties. Journal of Process Control, 129:103055, 2023.
[4] M. El Wajeh, A. Mhamdi, and A. Mitsos. Optimal Design and Flexible Operation of a Fully Electrified Biodiesel Production Process. Industrial & Engineering Chemistry Research, 63(3):1487–1500, 2024.
[5] R. Scattolini. Architectures for distributed and hierarchical Model Predictive Control – A review. Journal of Process Control, 19(5):723–731, 2009.
[6] M. El Wajeh, M. Granderath, A. Mitsos, and A. Mhamdi. Distributed Economic Nonlinear Model Predictive Control for Flexible Electrified Biodiesel Production— Part II: Sequential and Iterative Architectures with Computational Delay Compensation. Industrial & Engineering Chemistry Research, 63(42):18013–18026, 2024.
[7] R. Findeisen and F. Allgöwer. Computational Delay in Nonlinear Model Predictive Control. IFAC Proceedings Volumes, 37(1):427–432, 2004.
[8] A. Echtermeyer, C. Marks, A. Mitsos, and J. Viell. Inline Raman spectroscopy and Indirect Hard Modeling for concentration monitoring of dissociated acid species. Applied Spectroscopy, 2021, 75(5), 506–519.