Please use this identifier to cite or link to this item: http://hdl.handle.net/2445/173450
Title: Window of deposition description and prediction of deposition efficiency via machine learning techniques in cold spraying
Author: Canales, Horacio
García Cano, Irene
Dosta Parras, Sergi
Keywords: Deposició (Metal·lúrgia)
Coure
Alumini
Aprenentatge automàtic
Plating
Copper
Aluminum
Machine learning
Issue Date: 1-Aug-2020
Publisher: Elsevier B.V.
Abstract: In this work we describe an energy-based window of deposition, and predict the deposition efficiency for different cold-sprayed powder/substrate systems, using machine learning techniques. We implement several machine learning models to predict whether particles adhere or bounce off during cold spraying. The models are trained using data extracted from several experimental runs taking into account the cumulative particle size distribution and the deposition efficiency of the process. The classification models infer a critical total energy threshold above which deposition occurs. Based on this threshold, we describe an energy-based window of deposition for the powder/substrate systems studied. These models predict the deposition efficiency of different spraying operations for different powder materials with acceptable accuracy. Machine learning techniques provide better understanding of the particle deposition process and enable a more comprehensive exploration of the scope of cold spraying. The use of these techniques opens up new possibilities for the pursuit of links between the spraying process, the structure and different properties for novel cold-sprayed materials.
Note: Versió postprint del document publicat a: https://doi.org/10.1016/j.surfcoat.2020.126143
It is part of: Surface & Coatings Technology, 2020, vol. 401, p. 126143
URI: http://hdl.handle.net/2445/173450
Related resource: https://doi.org/10.1016/j.surfcoat.2020.126143
ISSN: 0257-8972
Appears in Collections:Articles publicats en revistes (Ciència dels Materials i Química Física)

Files in This Item:
File Description SizeFormat 
704870.pdf22.38 MBAdobe PDFView/Open    Request a copy


Embargat   Document embargat fins el 1-8-2022


This item is licensed under a Creative Commons License Creative Commons