 |
|||
| Url | https://cimne.com/sgp/rtd/Project.aspx?id=931 | ||
|
|||
| Acronym | DIDRO | ||
| Project title | Towards building of Digital Twins for manufacturing processes based on drop-on-demand printing | ||
| Reference | TED2021-130471B-I00 | ||
| Principal investigator |
Pavel RYZHAKOV - pryzhakov@cimne.upc.edu
Riccardo ROSSI BERNECOLI - rrossi@cimne.upc.edu |
||
| Start date | 01/12/2022 | End date | 31/05/2025 |
| Coordinator | CIMNE | ||
| Consortium members |
|
||
| Program | P.E. para Impulsar la Investigación Científico-Técnica y su Transferencia | Call | Proyectos Estratégicos Orientados a la Transición Ecológica y a la Transición Digital» 2021 |
| Subprogram | Subprograma Estatal de Generación de Conocimiento | Category | Nacional |
| Funding body(ies) | MICINN | Grant | $143,750.00 |
| Abstract | Drop-on-demand (DOD) inkjet printing is interesting to multiple industries due to its precision and high cost-effectiveness. Products ranging from optoelectronic devices to catalyst layers in energy conversion devices are currently manufactured using such an approach. For the existing applications, the challenge consists in finding a way for achieving maximum printing precision and, simultaneously, minimizing ink expenditure (particularly, when the application requires using inks containing costly materials, such as platinum or gold). Additionally, practitioners seek techniques to design optimal ink composition for a given application. In the framework of the fourth industrial revolution, the ideal tools that can enable such optimizations are the digital twins, capable of analyzing, monitoring, and simulating the intelligent manufacturing processes in real-time. The overall objective of DIDRO project is to make inroads towards developing a digital twin for intelligent DOD-based manufacturing. This will be done by devising and implementing a highly efficient computational tool for drop-on-demand inkjet printing process. The core of the tool will be a high-fidelity computational fluid dynamics (CFD) model accounting for the formation and detachment of the ink droplet/jet at the nozzle, its transport through the surrounding air and its spreading over the solid substrate. The model will rely on an enhanced two-phase flow solver incorporating, among others, non-Newtonian behavior of the liquid phase, and sophisticated treatment of the liquid-solid contact. Corresponding implementations will be done in an Open Source numerical framework with the aim of eventually enhancing the impact of the research onto the community. Efficient computational algorithms will also be devised for solving the coupled and non-linear physical system. Once developed and implemented, the model will be combined with the datadriven approach in order to facilitate nearly real-time feedback. This will be done by employing a machine learning (ML) cataloguing" approach, where the ML model will be trained by using detailed computational simulations on a set of typical scenarios, with the ultimate goal of providing quick, nearly real-time feedback regarding the inkjet behavior. This will result in a hybrid CFD-ML methodology for the simulation of DOD-printing, serving as a nucleus of the corresponding digital twin. The computational model will allow adjusting the optimization of DOD-printer configuration, finding printing operation regimes and optimal ink properties for a given industrial application. It will help answer the fundamental question of how different components of the DOD-printing process should be changed/adjusted so as to obtain the best possible quality of the given final product, providing powerful assistance in the digitalization of the product development chain in the manufacturing community | ||
| Proyecto TED2021-130471B-I00 de investigación financiado por MICIU/AEI /10.13039/501100011033 y por la Unión Europea Next GenerationEU/ PRTR | |||