(362ab) Digitalization of an Experimental Electrochemical Reactor Via the Smart Manufacturing Innovation Platform

The exponential increase of the amount of data produced in the last two decades has revolutionized the way we collect, store, consume, process, analyze, and interpret information to design energy-efficient, economic, and sustainable processes as well as to improve profitability in every industry including manufacturing. However, the digital transformation of data collected with advanced sensors, by developing practical data-driven models, is a complex and labor-intensive process. To this end, the Clean Energy Smart Manufacturing Innovation Institute (CESMII), a US Government funded public-private partnership has undertaken the responsibility to ease the digitization and transmission of data in numerous contexts, such as laboratory research on clean energy or large scale industrial processes, and built the Smart Manufacturing Innovation Platform (SMIP) [1] to enable efficient data storage, accelerate machine-learning model building, data visualization and insight extraction through data ingestion, transformation and orchestration. On top of this, advanced data-driven models can in turn be used in advanced feedback control schemes to improve productivity.

The present work demonstrates the application of the SMIP in the operation of an experimental electrochemical reactor that reduces CO₂ gas to multiple valuable liquid and gas chemicals, such as methane (natural gas) and hydrocarbons [2, 3]. Specifically, the use of SMIP involves transmitting the real-time sensor measurements over to a cloud resource (HTTPS), which subsequently distributes those operational data to all model building experts. For example, first principal models are under development for this electrochemical reactor and data-driven machine learning models have emerged as a valuable alternative [4] to represent the process operation. Every piece of equipment in SMIP has its own Smart Manufacturing profile (SM Profile) so that producers and consumers of the data have a clear understanding of the expected data structure and contents. Furthermore, the entire data collection and transmission process is fully automated through the back-end script written in Python, which effectively relieves the impact of human error. In addition, an operating system is developed with LabVIEW to control the electrochemical reactor and monitor the data flow with a single interface that can be obtained from the SMIP. Finally, all the software library application packages, algorithm, and user interface related to the demonstrated work is packed in Docker images for reproducibility and easy collaboration among researchers.

References:

[1] Davis, J.F.; Malkani, H.; Dyck, J.; Korambath, P.; Wise, J.; Cyberinfrastructure for the Democratization of Smart Manufacturing, book chapter, Smart Manufacturing: Concepts and Methods, in publication, 2020

[2] 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).

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

[4] Luo, J., Canuso, V., Jang, J. B., Wu, Z., Morales-Guio, C. G., Christofides, P. D., 2022. Machine
learning-based operational modeling of an electrochemical reactor: Handling data variability and improving empirical models. Industrial & Engineering Chemistry Research, in press.