An advanced resin reaction modeling using data-driven and digital twin techniques

Abstract

Elium® resin is nowadays actively investigated to leverage its recycling ability. Thus, multiple polymerization modeling are developed and used. In this work, we investigate the polymerization of Elium®/Carbon fiber composite in a cylindrical deposition, followed by an in-oven heating. The model parameters are optimized using an active-set algorithm to match the experimental heating profiles. Moreover, the simulation efforts are coupled to an artificial intelligence modeling of the discrepancies. For instance, a surrogate model using convolution recurrent neural network is trained to reproduce the error of the simulation. Later, a digital twin of the process is built by coupling the simulation and the machine learning algorithm. The obtained results show a good match of the experimental results even on the testing sets, never used during the training of the surrogate model. Finally, the digital twin results are post-processes to investigate the resin polymerization through the thickness of the part.

Learning Data-Driven Stable Corrections of Dynamical Systems—Application to the Simulation of the Top-Oil Temperature Evolution of a Power Transformer

“…Optimal transport is very suitable in these situations where the hybrid model consists of the usual nominal model enriched from a parametric correction formulated in an optimal transport setting, as described in [16,17]; • In [18], a correction of a Mises yield function was performed from the deviation between the results predicted by using it and the measures obtained at the structure level; • Finally, when addressing processing and performances, the hybrid formulation, which can be used with different granularities, can exhibit advantages (amount of data, ability to explain, knowledge transfer, etc.) that pertain to the heart of digital twin developments [9,[19][20][21][22][23][24].…”

Learning Data-Driven Stable Corrections of Dynamical Systems. Application to the Simulation of the Top-Oil Temperature Evolution of a Power Transformer

“…Finally, when addressing processing and performances, the hybrid formulation, of different granularities, can exhibit advantages (amount of data, ability to explain, knowledge transfer, ...), at the heart of the digital twin developments [1,5,6,13,14,21,36].…”