Options
Uncertainty quantification and statistical modeling of selective laser sintering process using polynomial chaos based response surface method
Journal
Journal of Manufacturing Processes
ISSN
15266125
Date Issued
2022-09-01
Author(s)
Thakre, Utkarsh
Mote, Rakesh G.
Abstract
Additive manufacturing processes like Selective Laser Sintering (SLS) are rapidly evolving and posing to replace conventional manufacturing routes. Parametric sensitivity analysis is vital in order to attain robust performance of Additive Manufacturing (AM) processes. This is to yield desired properties of the components fabricated given the complex process dynamics involved. Stochastic analysis of Selective Laser Sintering (SLS) process based on Monte Carlo simulation of the physics model as well as the surrogate models is found to be cumbersome or inefficient. In this work, Polynomial Chaos (PC) based response surface is utilized for stochastic analysis. The SLS physics is approximated using response surfaces utilizing lower-order polynomials. Unlike other surrogate models, sensitivity indices can be evaluated without additional computational expense. Thus, the framework is not only simple in implementation but also proven to be robust, accurate and computationally efficient with lower-order polynomials. Multiple key response properties are analyzed simultaneously for effects of input uncertainties. This was for the first-time stochastic analysis of SLS was performed with the complete and exhaustive treatment given to response data. Thus, a detailed study of aspects like parameter correlation, sensitivity analysis, percentile ranges, major distribution fit analysis and higher-order moments was performed.
Subjects