ProgramMaster Logo
Conference Tools for 2024 TMS Annual Meeting & Exhibition
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools
About this Abstract
Meeting 2024 TMS Annual Meeting & Exhibition
Symposium Additive Manufacturing Modeling, Simulation and Machine Learning
Presentation Title Hardness Predictions of Additively Manufactured Components Using Convolutional Neural Networks on Backscattered Electron Images
Author(s) Dillon Watring, Patrick G Callahan, David J Rowenhorst
On-Site Speaker (Planned) Dillon Watring
Abstract Scope Additive manufactured (AM) components have heterogenous microstructure due to variations in thermal histories. Understanding the correlation between these microstructures and mechanical properties are important to qualification and critical use. While hardness tests are a standard technique used to rapidly assess the mechanical properties of a material, many tests are needed to capture the variations in properties due to the fluctuations within the processing/microstructure. Convolutional neural networks (CNNs) allow unique feature identification and extraction from images that are difficult for trained experts to detect, which allows prediction of regression values based solely on images. This work provides a non-destructive alternative for hardness prediction by leveraging CNNs to predict hardness from backscattered electron (BSE) images. The study evaluates the CNN model’s predictive performance using nickel aluminum bronze (NAB) samples, and investigates and identifies microstructural feature importance in governing hardness across different length scales.
Proceedings Inclusion? Planned:
Keywords Additive Manufacturing, Machine Learning, Mechanical Properties

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

3D Deep Learning for Pore Stress Concentration Analysis in Additive Manufacturing
A CALPHAD Model of Dendritic Growth for the Design of Printable Industrial Alloys
A Mixed Sharp and Diffusive Interface Approach for Multi-physics Modeling of Metal Additive Manufacturing
A Model Experiment of Meltpool Dynamics in Additive Manufacturing with Magnetic Fields
A Multi-physics Model for Melt Pool and Keyhole Dynamics in Laser Powder Bed Fusion Process
A Robust Model for Estimating the Metal Evaporation during Laser Powder Bed Fusion with Inputs from CALPHAD Approach
Additive Manufacturing Process Optimization and Property Prediction with Integrated Computational Materials Design
An ICME Workflow to Identify the Root Cause of Properties Variations of AM Parts
Analyzing Debinding and Carbide Pickup for Quality Control of Binder Jet Printed SS 316L Using Computer Vision
Analyzing Micro-macro Transitional Length Scale in 3D Printed Chopped Fiber Reinforced Polymers
Application of Computer Vision to Mapping of Process Parameters to Material Structure of AM Carbon Fiber Composites
Compensating for Sintering Distortion in Additively Manufactured Copper using Physics-Informed Gaussian Process Regression
Computational Modeling and Experimental Investigation of Additively Manufactured Fused Deposition Modeling Samples with In-Built Porosity
Computationally Derived Additively Manufactured Microstructure-property Correlations for Nickel-based Superalloys
Coupling of Microscopy and Thermomechanical Models to Explain the Extent and Location of TRIP Product in Simulated PBF-LB of Ti-1023
Critical Velocity and Deposition Efficiency in Cold Spray: A Reduced-order Model and Experimental Validation
Crystal Plasticity Modeling for Prediction of Fatigue Crack Initiation in Defect-containing Additively Manufactured Al-10Si-0.4Mg Alloys
Crystal Plasticity Modeling of Thermo-elastic-plastic Deformation during Laser-based Additive Manufacturing
Data Bridging: A Novel Pipeline for Efficient Statistical Exploitation Across Multiple Data Populations
Deep Neural Network for Image Segmentation and Feature Quantification during Laser Powder Bed Fusion Additive Manufacturing
Development of a Multi Scale and Multi Physics Modeling Framework to Study the Defect Evolution in The Laser-based Powder Bed Fusion Process
Development of Process-Microstructure Relationships in Laser Powder Bed Fusion of IN718
Development of Simulation-based Qualification Data for Laser Powder Bed Fusion Using Modeling and Uncertainty Quantification
Digital Twin Framework for Identifying Microstructure Heterogeneity in an As-built Powder Bed Fusion Part
Domain Stitching: A Technique for Large-scale Microstructural Studies in Laser Powder Bed Fusion
Effect of Nonuniform Void Distributions on the Yield Strength of Metals
Efficient Computational Framework for Image-based Micromechanical Analysis of Additively Manufactured Ti-6Al-4V Alloys
Efficient Process Control Model for Laser Powder Bed Fusion Using an Experimentally Validated Heat Source
Exascale Simulation for Additive Manufacturing
F-13: A Machine Learning Based Approach for Accelerated Textured Microstructure Generation
F-14: A Multi-scale Modelling Approach for Wire-based Laser Metal Deposition Process
F-15: An Accurate Machine Learning Approach for Process Optimization in Directed Energy Deposition
F-16: Controlling Microstructures and Mechanical Properties of Nickel-based Superalloy Based on Multiscale Finite Element Thermal Analysis in Laser Powder Bed Fusion
F-17: Coupling In-situ Monitoring and Machine Learning Towards Faster Laser-based Powder Bed Fusion Process Qualification
F-18: Efficient Process Parameter Optimization for Titanium Alloys in Additive Manufacturing
F-19: High-strain Rate and High-temperature Properties of Additively Manufactured Alloy 718
F-65: Microstructure Control via Coordinated Dual-beam Laser Scanning
F-66: Numerical Analysis of Heat Accumulation during Wire Arc Additive Manufacturing
F-67: Particle Tracking in a Simulated Melt Pool of Laser Powder Bed Fusion
F-68: Reducing Validation to Days, Enabling Rapidly Deployable Additive Manufacturing at the Front Line
F-70: RLTube: Optimizing Path Planning in Wire Arc Additive Manufacturing for Customized Bent Tubes
F-74: Crystal Plasticity Modeling for the Prediction of Mechanical Properties of Laser Powder Bed Fusion AlSi10Mg Parts
F-75: Some Guidelines for the Use of Machine Learning in Metal AM Process Parameter Development
Fast and Scalable Method to Generate Reduced Order Models of Metal-based Additive Manufacturing Simulations Using a Hypercomplex-based Automatic Differentiation Finite Element Method
Grain Structure Control Through Modeling of Laser Beam Shaping and Multibeam Solidification
Hardness Predictions of Additively Manufactured Components Using Convolutional Neural Networks on Backscattered Electron Images
Hybrid Model Guided Additively Manufactured Aerospace Heat Exchanger Development
ICME-based Integrated Modelling Framework for Additively Manufactured Ni-based Superalloys
In-situ Monitoring and Numerical Simulation of Shrinkage during Sintering in Metal Binder Jetted Parts
JIMM Young Leader International Scholar Award Lecture: Machine-learning Approaches to Control the Microstructure and Properties of Laser Powder Bed Fused Metallic Components
Leveraging Convolutional Neural Networks for the Prediction of Enhanced Plume and Coating Quality in Atmospheric Plasma Spraying
Machine Learning-based Prediction of Evolution of Thermal Profiles During Additive Manufacturing
Machine Learning Guided Prediction of Jetting Behavior during Electrohydrodynamic (EHD) Printing
Micro-mechanical Computational Modeling of Dislocation Cell Structures
Modeling Inherent Anisotropic Deformation Behavior of Laser Powder Bed Fusion (LPBF) Manufactured Metals for Different Laser Beam Shapes
Modeling the Hardening and Damage Evolution of Additively Manufactured Metal Matrix Composites Using a Large-strain Elasto-viscoplastic FFT-based Framework
Multi-beam Process Modeling for Optimization of Melt Pool Shape and Build Rate for Laser Powder Bed Fusion
Multi-Information Source Thermal Modeling for Design of Printable Refractory Alloys
Multi-physics Modeling of Melt Pool with Ray-tracing in the Open-source MALAMUTE Software
Multiscale and Machine Learning Modeling for Texture Prediction during Additive Manufacturing
Numerical Model to Unravel Thermal Evolution and Material Flow Behavior in Additive Friction Stir Deposition of Mg-alloy
On Predicting the Fatigue Behavior of Direct Aged L-PBF IN718 Using Machine Learning Informed by μXCT and EBSD
On the Applicability of CALPHAD and Process Models to Predict Solidification Cracking
Physics-constrained, Inverse Design of High-temperature Strength Printable Aluminum Alloys with Low Cost and CO2 Emissions for High Demand Industries
Planning and Adaptive Control of AM Processes via In Situ Characterization, Faster-than-real-time Simulations, and AI/ML Methods
Prediction of Process Maps and Location Specific Properties for Additive Manufacturing through CALPHAD.
Revealing the Role of Volumetric Defect’s Geometry on Fatigue Crack Initiation in Additively Manufactured Materials
Scanning Strategies Optimization for Gluing of Kraft Paper Using Laser in Laminated Additive Manufacturing
Shape Distortion in Sintering-based Additive Manufacturing Results from Nonhomogeneous Temperature Activating a Long-range Mass Transport
Simulation of Residual Stresses, Deformations, and the Effect of Support Removal in Additively Manufactured Thin Plates
Softening Mechanisms in Additively Manufactured 420 Stainless Steel at Elevated Temperatures
Solidification Kinetics in Ternary Alloys: Insights from Phase Field Modeling
The Development of Grain Structure During Additive Manufacturing: A Comparison Between Experiment and Simulation
The Impact of Beam Shaping on Grain Morphology and Mechanical Response of Additively Manufactured Microstructures
Towards Adaptive Metal Additive Manufacturing: The Role of Modeling in Real-time Process Control
Towards Large-scale Grain Growth Modeling in Powder Bed Fusion
Transfer Learning Based Prediction of Part Quality in Additive Manufacturing
Utilizing Additive Manufacturing and CFD Simulation to Enhance Investment Casting of Aluminum Alloys
Validation of a Transient Heat Source Model for Laser Powder Bed Fusion
Questions about ProgramMaster? Contact programming@programmaster.org