Next level optimization of your dynamic crossflow filter (DCF) performance with digital twins

Publisher FILTECH

M. Amthor-Stahl*, J. Rauhut; Andritz Separation GmbH, Germany

Cross-flow technology for separating substances with poor filtration properties are state of the art technology for solid-liquid separation in chemical and pharmaceutical industry. With traditional cross-flow filtration the development of the clogging layer at the surface of the filter membranes is prevented by re-circulating the feed. When handling sensitive or highly viscous products, this method is not efficient or even not applicable. With Andritz Dynamic Cross Flow filtration (DCF) recirculation can be avoided and processing of high viscous products with pasty consistency is possible.

In DCF shear forces are generated between the overlapping rotating disks. By this effect, the crossflow is generated without pumping and circulating the slurry, but just by the rotation of the membranes themselves...

For most of the players in food and chemical industries, using a DCF is an efficient way to filtrate different products but today, operations need significant improvement to gain in productivity and reduce opex. Up to now, the huge potential of Dynamic Crossflow Filter technology in industrial processes is often not utilized, because quick reaction time of the system to varying process conditions are not possible by fixed recipes. In order to further increase long-term performance of DCF and allow the handling of varying process conditions a digital twin was developed for an industrial DCF plant. The digital replica of the physical process uses asset-specific information such as actual physical dimensions and equipment characteristics and runs in parallel with operations, continuously calculating key process parameters which sometimes cannot be measured or are difficult to measure by traditional instrumentation. Through proven simulation capabilities, combined with live plant data as an input, it calculates continuously the optimal flow setpoint, and performs what-if scenario analysis.

During the harvesting season in 2022 and 2023 in total 24 industrial production batches of grape juice were analyzed in this project. After implementation of the digital twin the...

Published in: FILTECH 2024 Conference

Date of Conference: 12 November - 14 November 2024

DOI: -

Presenter's Affiliation: Andritz Separation GmbH

Publisher: FILTECH Exhibitions GmbH & Co. KG

Country: Germany

Electronic ISBN: 978-3-941655-20-1

Conference Location: Cologne, Germany

Keywords: Biotechnology, Dynamic Cross-Flow Filtration, Process Optimisation, Automation, Metris, Digital Twin, Adaptive Control