Using machine learning to predict residence time distributions in Coiled Flow Inverter (CFI) reactors

Maria Cecilia Barrera, Aleksander Josifovic, John Robertson, Blair Johnston, Cameron Brown, Alastair Florence

Research output: Contribution to conferencePoster

25 Downloads (Pure)

Abstract

INTRODUCTION
- A CFI reactor consist of a inner tube wrapped around a circular frame. This geometry leads to the formation of secondary flow patterns (Dean vortices) known to improve radial mixing.
- Better radial mixing results in a tighter residence time distribution (𝑅𝑅𝑅𝑅𝑅𝑅). Many processes, such as virus inactivation, require a tight 𝑅𝑅𝑅𝑅𝑅𝑅 to avoid unwanted transformations or product damage.
- The tightness of the 𝑅𝑅𝑅𝑅𝑅𝑅 can be evaluated using the relative width (𝑅𝑅𝑤𝑤), the ratio between the minimum and maximum residence times. An 𝑅𝑅𝑤𝑤 value of 1 corresponds to an ideal plug flow reactor.
Original languageEnglish
Pages11-11
Number of pages1
Publication statusPublished - 16 May 2022
EventCMAC Annual Open Day 2022 - Glasgow, United Kingdom
Duration: 16 May 202218 May 2022

Conference

ConferenceCMAC Annual Open Day 2022
Country/TerritoryUnited Kingdom
CityGlasgow
Period16/05/2218/05/22

Keywords

  • Coiled Flow Inverter reactor
  • CFI reactor
  • machine learning
  • flow patterns

Fingerprint

Dive into the research topics of 'Using machine learning to predict residence time distributions in Coiled Flow Inverter (CFI) reactors'. Together they form a unique fingerprint.

Cite this