(185f) Machine Learning-Based Ethylene Concentration Estimation, Real-Time Optimization and Feedback Control of an Experimental Electrochemical Reactor

With the increase in electricity supply from clean energy sources, electrochemical reduction of carbon dioxide (CO₂) has received increasing attention as an alternative source of carbon-based fuels [1, 2]. As CO₂ reduction is becoming a stronger alternative for the clean production of chemicals, the necessity to model, optimize and control the electrochemical reduction of CO₂ process is becoming inevitable. However, on the one hand, a first-principles model to represent the electrochemical CO₂ reduction has not been fully developed yet because of the complexity of its reaction mechanism [3], which makes it challenging to define a precise state-space model for the control system. On the other hand, the unavailability of efficient concentration measurement sensors continues to challenge our ability to develop feedback control systems. Gas chromatography (GC) is the most common equipment to monitor the product composition, but it requires a period of time to analyze the sample, which means the GC can only provide delayed, real-time measurements during the operation. Also, the electrochemical CO₂ reduction process is catalyzed by a fast-decaying copper catalyst and undergoes a selectivity shift from the product-of-interest at the later stages of experiments, challenging conventional control methods. To this end, machine learning (ML) techniques provide a potential approach to overcome those difficulties due to their demonstrated capability to capture the dynamic behavior of a chemical process from data [5].

Motivated by the above considerations, this work proposes a machine-learning-based modeling methodology that integrates support vector regression and first-principles modeling to capture the dynamics behavior of an experimental electrochemical reactor; this model, together with limited gas chromatography measurements, is employed to predict the evolution of gas-phase ethylene concentration. The model prediction is directly used in a proportional-integral controller that manipulates the applied potential to regulate the gas-phase ethylene concentration at energy-optimal set-point values computed by a real-time process optimizer (RTO). Specifically, the RTO optimizes the operation set-point by solving an optimization problem to maximize the economic benefit of the reactor. Lastly, suitable compensation methods are introduced to further account for the experimental uncertainties and handle catalyst de-activation. The performance of the proposed modeling, optimization and control approaches are demonstrated by results from a series of experiments.

References:

[1] Morales-Guio, C.G.; Cave, E.R.; Nitopi, S.A.; et al. Improved CO2 reduction activity towards C2+ alcohols on a tandem gold on copper electrocatalyst. Nat Catal 1, 764–771 (2018).

[2] Jang, J.; Shen, K.; Morales-Guio, C.G.; Electrochemical direct partial oxidation of methane to methanol. Joule. 2019, 3(11), 2589-93.

[3] Nitopi, S.; Bertheussen, E.; Scott, S.; Liu, X.; Engstfeld, A.; Horch, S.; Seger, B.; Stephens, I.;
Chan, K.; Hahn, C.; et al., Progress and perspectives of electrochemical CO2 reduction on copper in aqueous electrolyte. Chemical reviews 2019, 119, 7610–7672.

[4] Wu, Z.; Tran, A.; Rincon, D.; Christofides, P. D. Machine learning-based predictive control of nonlinear processes. Part I: Theory. AIChE J. 2019, 65, e16729

[5] Luo, J.; Canuso, V.; Jang, J.; Wu, Z.; Morales-Guio, G.C.; Christofides, P. D. Machine Learning-Based Operational Modeling of an Electrochemical Reactor: Handling Data Variability and Improving Empirical Models. Ind. Eng. Chem. Res. 2022, Accepted.