Predictive modeling approaches in laser-based material processing

Predictive modeling represents an emerging field that combines existing and novel methodologies aimed to rapidly understand physical mechanisms and concurrently develop new materials, processes, and structures. In the current study, previously unexplored predictive modeling in a key-enabled technolo...

Full description

Saved in:
Bibliographic Details
Published in:Journal of applied physics 2020-11, Vol.128 (18)
Main Authors: Velli, Maria-Christina, Tsibidis, George D., Mimidis, Alexandros, Skoulas, Evangelos, Pantazis, Yannis, Stratakis, Emmanuel
Format: Article
Language:English
Subjects:
Algorithms
Data integration
Information theory
Laser applications
Laser processing
Lasers
Machine learning
Modelling
Performance prediction
Prediction models
Production costs
Uncertainty
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Predictive modeling represents an emerging field that combines existing and novel methodologies aimed to rapidly understand physical mechanisms and concurrently develop new materials, processes, and structures. In the current study, previously unexplored predictive modeling in a key-enabled technology, the laser-based manufacturing, aims to automate and forecast the effect of laser processing on material structures. The focus is centered on the performance of representative statistical and machine learning algorithms in predicting the outcome of laser processing on a range of materials. Results on experimental data showed that predictive models were able to satisfactorily learn the mapping between the laser’s input variables and the observed material structure. These results are further integrated with simulation data aiming to elucidate the multiscale physical processes upon laser–material interaction. As a consequence, we augmented the adjusted simulated data to the experiment and substantially improved the predictive performance due to the availability of an increased number of sampling points. In parallel, an information-theoretic metric, which identifies and quantifies the regions with high predictive uncertainty, is presented, revealing that high uncertainty occurs around the transition boundaries. Our results can set the basis for a systematic methodology toward reducing material design, testing, and production cost via the replacement of expensive trial-and-error based manufacturing procedures with a precise pre-fabrication predictive tool.
ISSN:0021-8979
1089-7550
DOI:10.1063/5.0018235