Arctos

Research Engineer

• CGAN Development and Refinement
• Retrain and refine a conditional generative adversarial network (CGAN) to generate synthetic PMC and CMC microstructures
• Incorporate experimental optical microscopy data into the CGAN training process
• Ensure the CGAN accurately models realistic morphologies conditioned on volume fraction and fiber arrangement
• Data Integration and Preprocessing
• Process and analyze experimental optical microscopy data to prepare it for use in CGAN training
• Develop algorithms to extract relevant features (e.g., volume fraction, fiber arrangement) from experimental data
• Model Validation and Performance Evaluation
• Validate the CGAN-generated microstructures against experimental data to ensure realism and accuracy
• Develop metrics and benchmarks to evaluate the quality of synthetic microstructures
• Collaboration and Reporting
• Collaborate with AFRL researchers and other stakeholders to align CGAN outputs with experimental observations
• Document methodologies, results, and findings in technical reports and presentations
• Simulation and Analysis
• Use the refined CGAN-generated microstructures to simulate residual stress distributions and analyze their impact on material performance
• Provide insights into the relationship between microstructure morphology and mechanical properties
• Tool Development and Optimization
• Develop and optimize computational tools for microstructure generation and analysis
• Ensure scalability and efficiency of the CGAN framework for broader applications

Research Engineer – Composite Analysis