Prediction of porosity in metal-based additive manufacturing using spatial Gaussian process models

Tapia, Gustavo; Elwany, Alaa; Sang, Huiyan

doi:10.1016/j.addma.2016.05.009

Cited by 139 publications

(82 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[78,79] Tapia et al applied a GP model to predict the porosity in metallic parts produced by selective laser melting, which is a laser-based additive manufacturing process (Table I). [80] In the authors' work, the input variables are the laser power and laser scanning speed, which are two of the most influential processing parameters. The output variable is the porosity of specimens (17-4 PH stainless steel).…”

Section: Gaussian Processmentioning

confidence: 99%

Machine learning for composite materials

2019

View full text Add to dashboard Cite

show abstract

Section: Gaussian Processmentioning

confidence: 99%

Machine learning for composite materials

2019

View full text Add to dashboard Cite

show abstract

“…where P is the laser power, v is the scan speed, t is the layer thickness and h is the hatch distance. Consequently, the studies [10,20,21,22] focus on the influence of these parameters on build quality. Arisoy et al [21] discusses the effect of scan strategy, laser power, scan speed and hatch distance on grain sizes and showed that increasing the energy density results in larger grain sizes.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Arisoy et al [21] discusses the effect of scan strategy, laser power, scan speed and hatch distance on grain sizes and showed that increasing the energy density results in larger grain sizes. The work described in [22] has used scan speed and laser power to predict the porosity of metallic parts produced using L-PBF.…”

Section: Literature Reviewmentioning

confidence: 99%

Automatic Quality Assessments of Laser Powder Bed Fusion Builds from Photodiode Sensor Measurements

Jayasinghe¹,

Paoletti²,

Sutcliffe³

et al. 2020

Preprint

View full text Add to dashboard Cite

This study evaluates whether a combination of photodiode sensor measurements, taken during laser powder bed fusion (L-PBF) builds, can be used to predict the resulting build quality via a purely data-based approach. We analyse the relationship between build density and features that are extracted from sensor data collected from three diﬀerent photodiodes. The study uses a Singular Value Decomposition to extract lower-dimensional features from photodiode measurements, which are then fed into machine learning algorithms. Several unsupervised learning methods are then employed to classify low density (< 99% part density) and high density (≥ 99% part density) specimens. Subsequently, a supervised learning method (Gaussian Process regression) is used to directly predict build density. Using the unsupervised clustering approaches, applied to features extracted from both photodiode sensor data as well as observations relating to the energy transferred to the material, build density was predicted with up to 93.54% accuracy. With regard to the supervised regression approach, a Gaussian Process algorithm was capable of predicting the build density with a RMS error of 3.65%. The study shows, therefore, that there is potential for machine learning algorithms to predict indicators of L-PBF build quality from photodiode build-measurements. Moreover, the work herein describes approaches that are predominantly probabilistic, thus facilitating uncertainty quantiﬁcation in machine-learnt predictions of L-PBF build quality.

show abstract

“…Gaussian processes have a long history of application in mining, agriculture, and forestry since 1920 [41]. Due to their versatility, they have experienced significant growth over the past three decades, especially with the advances in low-cost high-speed computing [42].…”

Section: Literature Reviewmentioning

confidence: 99%

Bayesian Calibration and Uncertainty Quantification for a Physics-Based Precipitation Model of Nickel–Titanium Shape-Memory Alloys

Tapia

Johnson

Franco

et al. 2017

Journal of Manufacturing Science and Engineering

View full text Add to dashboard Cite

Uncertainty quantification (UQ) is an emerging field that focuses on characterizing, quantifying, and potentially reducing, the uncertainties associated with computer simulation models used in a wide range of applications. Although it has been successfully applied to computer simulation models in areas such as structural engineering, climate forecasting, and medical sciences, this powerful research area is still lagging behind in materials simulation models. These are broadly defined as physics-based predictive models developed to predict material behavior, i.e., processing-microstructure-property relations and have recently received considerable interest with the advent of emerging concepts such as Integrated Computational Materials Engineering (ICME). The need of effective tools for quantifying the uncertainties associated with materials simulation models has been identified as a high priority research area in most recent roadmapping efforts in the field. In this paper, we present one of the first efforts in conducting systematic UQ of a physics-based materials simulation model used for predicting the evolution of precipitates in advanced nickel–titanium shape-memory alloys (SMAs) subject to heat treatment. Specifically, a Bayesian calibration approach is used to conduct calibration of the precipitation model using a synthesis of experimental and computer simulation data. We focus on constructing a Gaussian process-based surrogate modeling approach for achieving this task, and then benchmark the predictive accuracy of the calibrated model with that of the model calibrated using traditional Markov chain Monte Carlo (MCMC) methods.

show abstract

Prediction of porosity in metal-based additive manufacturing using spatial Gaussian process models

Cited by 139 publications

References 63 publications

Machine learning for composite materials

Machine learning for composite materials

Automatic Quality Assessments of Laser Powder Bed Fusion Builds from Photodiode Sensor Measurements

Bayesian Calibration and Uncertainty Quantification for a Physics-Based Precipitation Model of Nickel–Titanium Shape-Memory Alloys

Contact Info

Product

Resources

About