Dynamic flow experiments for Bayesian optimization of a single process objective
DynO guides an experimental optimization campaign by suggesting the conditions to use in dynamic flow experiments. DynO is supported by a Gaussian process and stopping criteria, efficiently combining dynamic experiments and Bayesian optimization.
Authors
- Cameron Armstrong (ORCID: https://orcid.org/0000-0001-6342-0860)
- Federico Florit (ORCID: https://orcid.org/0000-0002-6484-4953)
- Klavs F. Jensen (ORCID: https://orcid.org/0000-0001-7192-580X)
- Jason Mustakis (ORCID: https://orcid.org/0000-0002-8683-0691)
- Angel R. Diaz
- Steven M. Guinness (ORCID: https://orcid.org/0000-0003-4719-313X)
- Kakasaheb Y. Nandiwale (ORCID: https://orcid.org/0000-0002-0754-7362)
- Katharina Grohowalski
Institutions
- Pfizer (United States) (US)
- Massachusetts Institute of Technology (US)
Publication Details
- Journal
- Reaction Chemistry & Engineering
- Published
- 2024-12-11
- DOI
- https://doi.org/10.1039/d4re00543k
- Citations
- 5
- Primary Topic
- Innovative Microfluidic and Catalytic Techniques Innovation
- Type
- article
- Field-Weighted Citation Impact
- 0.92
Funders
- Pfizer