Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems
The population balance equation (PBE) is today a convenient tool to describe the evolution of multiphase systems, such as bubble plumes or columns, mixing tanks, solvent extraction columns, where they are used to infer the particle size distribution required to quantify buoyancy forces or the interf...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Chemical Engineering Journal Advances |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666821125001267 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849225262300921856 |
|---|---|
| author | Grégory Bana Fabrice Lamadie Sophie Charton Didier Lucor Nida Sheibat-Othman |
| author_facet | Grégory Bana Fabrice Lamadie Sophie Charton Didier Lucor Nida Sheibat-Othman |
| author_sort | Grégory Bana |
| collection | DOAJ |
| description | The population balance equation (PBE) is today a convenient tool to describe the evolution of multiphase systems, such as bubble plumes or columns, mixing tanks, solvent extraction columns, where they are used to infer the particle size distribution required to quantify buoyancy forces or the interfacial exchange area. However, the PBE often rely on semi-empirical models (or kernels) to predict the events likely to modify the population properties, such as breakage and coalescence. These kernels contain parameters that are tailored to specific fluid properties and operating conditions, thus limiting their general applicability (i.e. transposition to other operating conditions or processes than those used during model development). As a consequence, accurately predicting the frequency of events that modify the population of droplets remains challenging. In this study, we focus on the breakage phenomenon. Recent advances in machine learning, particularly Artificial Neural Networks (ANNs), present new opportunities for predicting the breakage frequencies. However, ANN training requires a large and high-fidelity dataset, making it time-consuming and error-prone. Physics-Informed Neural Networks (PINNs) may address this challenge by embedding physical laws into the learning process, ensuring physically consistent PBE predictions. This paper introduces a novel PINN-based algorithm that is trained to infer the droplet breakage frequencies in a turbulent agitated vessel without prior knowledge of the breakage kernel. It uses only the PBE discretized structure and a reasonable number of measured droplet size distributions obtained by in situ imaging. The methodology, we have named DROP-PINN, is first deployed and evaluated through controlled simulations, then experimentally over a wide range of dispersed phase viscosity, interfacial tension and agitation speed. |
| format | Article |
| id | doaj-art-7f5ea2199b8b4b91a3695bf0a8a6dbb1 |
| institution | Kabale University |
| issn | 2666-8211 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Chemical Engineering Journal Advances |
| spelling | doaj-art-7f5ea2199b8b4b91a3695bf0a8a6dbb12025-08-25T04:14:52ZengElsevierChemical Engineering Journal Advances2666-82112025-08-012310082910.1016/j.ceja.2025.100829Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systemsGrégory Bana0Fabrice Lamadie1Sophie Charton2Didier Lucor3Nida Sheibat-Othman4CEA, DES, ISEC, DMRC, Université de Montpellier, Marcoule, Bagnols-sur-Cèze, 30200, FranceCEA, DES, ISEC, DMRC, Université de Montpellier, Marcoule, Bagnols-sur-Cèze, 30200, FranceCEA, DES, ISEC, DMRC, Université de Montpellier, Marcoule, Bagnols-sur-Cèze, 30200, FranceUniversité Paris-Saclay, CNRS, Laboratoire Interdisciplinaire des Sciences du Numérique, bâtiment 507, Rue du Belvédère, Orsay, 91405, FranceUniversite Claude Bernard Lyon 1, LAGEPP, UMR 5007 CNRS, 43 Boulevard du 11 Novembre 1918, Villeurbanne, 69100, France; Corresponding author.The population balance equation (PBE) is today a convenient tool to describe the evolution of multiphase systems, such as bubble plumes or columns, mixing tanks, solvent extraction columns, where they are used to infer the particle size distribution required to quantify buoyancy forces or the interfacial exchange area. However, the PBE often rely on semi-empirical models (or kernels) to predict the events likely to modify the population properties, such as breakage and coalescence. These kernels contain parameters that are tailored to specific fluid properties and operating conditions, thus limiting their general applicability (i.e. transposition to other operating conditions or processes than those used during model development). As a consequence, accurately predicting the frequency of events that modify the population of droplets remains challenging. In this study, we focus on the breakage phenomenon. Recent advances in machine learning, particularly Artificial Neural Networks (ANNs), present new opportunities for predicting the breakage frequencies. However, ANN training requires a large and high-fidelity dataset, making it time-consuming and error-prone. Physics-Informed Neural Networks (PINNs) may address this challenge by embedding physical laws into the learning process, ensuring physically consistent PBE predictions. This paper introduces a novel PINN-based algorithm that is trained to infer the droplet breakage frequencies in a turbulent agitated vessel without prior knowledge of the breakage kernel. It uses only the PBE discretized structure and a reasonable number of measured droplet size distributions obtained by in situ imaging. The methodology, we have named DROP-PINN, is first deployed and evaluated through controlled simulations, then experimentally over a wide range of dispersed phase viscosity, interfacial tension and agitation speed.http://www.sciencedirect.com/science/article/pii/S2666821125001267Population balance equationPhysic informed neural networksBreakage frequencyLiquid–liquid extraction |
| spellingShingle | Grégory Bana Fabrice Lamadie Sophie Charton Didier Lucor Nida Sheibat-Othman Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems Chemical Engineering Journal Advances Population balance equation Physic informed neural networks Breakage frequency Liquid–liquid extraction |
| title | Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems |
| title_full | Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems |
| title_fullStr | Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems |
| title_full_unstemmed | Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems |
| title_short | Bridging experiments and models, towards a new paradigm: DROP-PINN, a physics-informed neural network to predict droplet rupture in multiphase systems |
| title_sort | bridging experiments and models towards a new paradigm drop pinn a physics informed neural network to predict droplet rupture in multiphase systems |
| topic | Population balance equation Physic informed neural networks Breakage frequency Liquid–liquid extraction |
| url | http://www.sciencedirect.com/science/article/pii/S2666821125001267 |
| work_keys_str_mv | AT gregorybana bridgingexperimentsandmodelstowardsanewparadigmdroppinnaphysicsinformedneuralnetworktopredictdropletruptureinmultiphasesystems AT fabricelamadie bridgingexperimentsandmodelstowardsanewparadigmdroppinnaphysicsinformedneuralnetworktopredictdropletruptureinmultiphasesystems AT sophiecharton bridgingexperimentsandmodelstowardsanewparadigmdroppinnaphysicsinformedneuralnetworktopredictdropletruptureinmultiphasesystems AT didierlucor bridgingexperimentsandmodelstowardsanewparadigmdroppinnaphysicsinformedneuralnetworktopredictdropletruptureinmultiphasesystems AT nidasheibatothman bridgingexperimentsandmodelstowardsanewparadigmdroppinnaphysicsinformedneuralnetworktopredictdropletruptureinmultiphasesystems |