001048793 001__ 1048793
001048793 005__ 20251202203138.0
001048793 0247_ $$2doi$$a10.48550/ARXIV.2507.11640
001048793 037__ $$aFZJ-2025-04908
001048793 1001_ $$0P:(DE-HGF)0$$aTrávníková, Veronika$$b0
001048793 245__ $$aQuantifying data needs in surrogate modeling for flow fields in two-dimensional stirred tanks with physics-informed neural networks
001048793 260__ $$barXiv$$c2025
001048793 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1764687908_30803
001048793 3367_ $$2ORCID$$aWORKING_PAPER
001048793 3367_ $$028$$2EndNote$$aElectronic Article
001048793 3367_ $$2DRIVER$$apreprint
001048793 3367_ $$2BibTeX$$aARTICLE
001048793 3367_ $$2DataCite$$aOutput Types/Working Paper
001048793 520__ $$aStirred tanks are vital in chemical and biotechnological processes, particularly as bioreactors. Although computational fluid dynamics (CFD) is widely used to model the flow in stirred tanks, its high computational cost$-$especially in multi-query scenarios for process design and optimization$-$drives the need for efficient data-driven surrogate models. However, acquiring sufficiently large datasets can be costly. Physics-informed neural networks (PINNs) offer a promising solution to reduce data requirements while maintaining accuracy by embedding underlying physics into neural network (NN) training. This study quantifies the data requirements of vanilla PINNs for developing surrogate models of a flow field in a 2D stirred tank. We compare these requirements with classical supervised neural networks and boundary-informed neural networks (BINNs). Our findings demonstrate that surrogate models can achieve prediction errors around 3% across Reynolds numbers from 50 to 5000 using as few as six datapoints. Moreover, employing an approximation of the velocity profile in place of real data labels leads to prediction errors of around 2.5%. These results indicate that even with limited or approximate datasets, PINNs can be effectively trained to deliver high accuracy comparable to high-fidelity data.
001048793 536__ $$0G:(DE-HGF)POF4-2172$$a2172 - Utilization of renewable carbon and energy sources and engineering of ecosystem functions (POF4-217)$$cPOF4-217$$fPOF IV$$x0
001048793 588__ $$aDataset connected to DataCite
001048793 650_7 $$2Other$$aComputational Engineering, Finance, and Science (cs.CE)
001048793 650_7 $$2Other$$aFOS: Computer and information sciences
001048793 650_7 $$2Other$$a76-10, 68T07 (Primary) 76D05, 35Q68 (Secondary)
001048793 7001_ $$0P:(DE-Juel1)129081$$avon Lieres, Eric$$b1$$ufzj
001048793 7001_ $$0P:(DE-HGF)0$$aBehr, Marek$$b2
001048793 773__ $$a10.48550/ARXIV.2507.11640
001048793 909CO $$ooai:juser.fz-juelich.de:1048793$$pVDB
001048793 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)129081$$aForschungszentrum Jülich$$b1$$kFZJ
001048793 9131_ $$0G:(DE-HGF)POF4-217$$1G:(DE-HGF)POF4-210$$2G:(DE-HGF)POF4-200$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-2172$$aDE-HGF$$bForschungsbereich Erde und Umwelt$$lErde im Wandel – Unsere Zukunft nachhaltig gestalten$$vFür eine nachhaltige Bio-Ökonomie – von Ressourcen zu Produkten$$x0
001048793 9141_ $$y2025
001048793 9201_ $$0I:(DE-Juel1)IBG-1-20101118$$kIBG-1$$lBiotechnologie$$x0
001048793 980__ $$apreprint
001048793 980__ $$aVDB
001048793 980__ $$aI:(DE-Juel1)IBG-1-20101118
001048793 980__ $$aUNRESTRICTED