Conference Logo ProgramMaster Logo
Conference Tools for 2027 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 2027 TMS Annual Meeting & Exhibition
Symposium AI-Enabled Materials Processing: Integrating Accelerated Experimental Workflows and Processing-Aware Machine Learning
Presentation Title Deep Learning-Based Segmentation for Porosity and Dendrite Quantification in Aluminum-Silicon Castings
Author(s) Edgar Omar Tarin, Jorge Acosta-Cordero, Jorge Gonzalez Vaquez, Norma Sanchez, Christopher Bradley, Brian Schuster
On-Site Speaker (Planned) Edgar Omar Tarin
Abstract Scope This study investigates the effect of processing conditions on the microstructure of aluminum silicon casting alloys using a deep learning pipeline. Solidification quality of samples is characterized and compared across preheating conditions and pouring temperatures. Microstructure analysis is quantified using a U-Net convolutional neural network to classify porosity and dendritic features, replacing manual and threshold-based methods that cannot reliably separate porosity from dendritic regions. This work analyzes the effects of processing parameters on pore formation, dendritic structure, secondary dendrite arm spacing, and grain size using features extracted from microstructural images. Porosity is measured as a function of height along the casting axis, and secondary dendrite arm spacing is correlated with measured cooling rate. By correlating processing parameters with porosity and dendritic microstructure, this work aims to clarify how processing factors influence pore and dendrite formation, supporting impurity-reduction studies that rely on porosity as an outcome measure.
Proceedings Inclusion? Planned:

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

A Benchtop Self-Driving Laboratory for Aqueous Chemistry Built on the CubXL Platform
A Cost-Aware Contextual Bandit Framework for Autonomous Optimization of Titanium Leaching from Iron-Depleted Red Mud Slag
A Deep Learning-Based Framework for Three-Dimensional Temperature Field Prediction and Intelligent Secondary Cooling Optimization in Continuous Casting
A Domain-Grounded Agentic AI Framework for Autonomous Manufacturing Process Data Analysis and Quality Prediction
A physics informed digital twin for powder bed fusion additive manufacturing via data assimilation and AI surrogates
A Processing-Aware Multimodal Data Pipeline for Porosity Quantification in High-Pressure Die Casting
A Programmable Powder Doser with 15+ Reservoirs and Automated Auger Swapping for AI-Enabled Alloy-Development Workflows
AlloyBot Automated SEM–XRD Pipeline for High-Throughput Microstructure Characterization
AlloyBot Self-Driving Laboratory to Develop High Temperature Alloys 100x Faster
AlloyBot SmartMelting: Closed-Loop Process Control for Reliable Automatic Arc-Melting Synthesis
An Autonomous Agentic System Enabled by Harness Engineering for Active Learning in a Self-Driving Laboratory for Materials Spallation
An Autonomous Platform for Machine Learning-Guided Electrodeposition
Asynchronous, Autonomous Materials Aging Characterization
Automating Rietveld Refinement like a Human Expert in GSAS-II
Autonomous Arc Melting for High-Throughput Alloy Synthesis
Autonomous High-Throughput Compression Testing Enabled by Machine Vision and AI Agents
Causal Intervention Mapping from Process Intensification to Pellet Architecture Design in Hydrogen Based Direct Reduction for Green Steelmaking
Closed-Loop Bayesian Optimization of Multi-Material 3D-Printed Tensegrity-Inspired Energy Absorbers
Closed-loop machine learning guided exploration of processing pathways reveals optimal microstructures
Data-Driven Multi-Objective Optimization of Ultrashort-Pulse Laser Cutting of Magnesium Alloy Microtubes for Bioabsorbable Stent Manufacturing
Deep Learning-Based Segmentation for Porosity and Dendrite Quantification in Aluminum-Silicon Castings
Development of an automated heat-treatment and evaluation system for autonomous exploration of process–structure–property relationships in structural metallic materials
Diffusion-Based Discovery of Ordered Superconductors
Enabling Autonomous Research Workflow Design, Scheduling, and Optimization at a Self-Driving Facility
Framework Towards Design of Workflows for Self-driving Laboratories (Application in Additive Manufacturing)
From Automated Materials Experiments to Trusted Autonomous Discovery in AIMD-L
From process observation to parameter selection in Directed Energy Deposition: An Expertise-Guided, Bead-to-Build Methodology
Low-Cost Open-Source Camera Modules for Continuous Monitoring of Self-Driving Laboratories
Machine Learning for Thermographic Feature Extraction in Multi-Material Directed-Energy Deposition
Machine learning in sintering and ceramic processing: from gradient descent optimizations to hybrid methods with finite element simulations.
Online Wire-arc Additive Manufacturing Process Optimization Using Active Learning
Optimizing Bead Geometry during Laser Wire-Powder Directed Energy Deposition using a Multi-objective Machine Learning Approach
Real-time digital twins for open-die cold forging using machine learning surrogate models
Reexamining Hall–Petch and the Tabor Relation in Concentrated FCC Alloys via BO-Driven Alloy Development
Resolving hidden subgroups in heterogeneous alloy datasets
The Agility Forge as a Tool for Shaping, Material Optimization and Accelerated Experimentation
The Optimization of Tar Sand Digestion by a Locally Fabricated Tar Sand Digester
Uncertainty-Aware Processing and Heat Treatment Design of Laser Powder Additive Manufactured Alloys
Uncertainty-Propagating Active Learning Accelerates High-Entropy Oxide Discovery via Pulsed Electrodeposition for Hydrogen Evolution
Using Machine Learning for Alloys Design and Optimisation in Additive Manufacturing
Questions about ProgramMaster? Contact programming@programmaster.org | TMS Privacy Policy | Accessibility Statement