Research

My research focuses on AI for Resilient Infrastructure and Wind Engineering, bridging cutting-edge artificial intelligence with practical structural engineering applications.

Research Interests

• Machine Learning Applications to Structural Engineering: Developing AI-based methods for structural health monitoring, defect detection, and performance prediction
• Wind Engineering: Wind load estimation, wind pressure coefficient prediction, and remote sensing applications
• Smart Infrastructure: IoT-based monitoring systems, automated inspection methods, and digital twin technologies
• Large Language Models for Engineering: Specialized LLM frameworks for building design codes and engineering standards

Current Research Projects

1. Hyper-converged Forensic Research Center for Infrastructure (2021-2027)

• Funding: NRF Korea, MSIT
• Role: Graduate Research Assistant
• Description: Development of AI-based forensic analysis methods for infrastructure assessment and failure prediction

2. Smart City Innovative Technology Demonstration Project (2024-2025)

• Funding: KAIA, MOLIT
• Role: Graduate Research Assistant
• Description: Implementation of smart city technologies including IoT sensors, AI-based monitoring systems, and real-time data analysis

3. AI for Construction Research Project (2024)

• Funding: SNU, Sejong Construction
• Role: Graduate Research Assistant
• Description: Application of machine learning and AI technologies to construction industry challenges

4. Outstanding Patent Prototype Development Project (2023)

• Funding: SNU
• Role: Graduate Research Assistant
• Description: Development and prototyping of AI-based concrete defect detection system

Research Methods and Tools

Machine Learning & Deep Learning

• Supervised Learning: Random Forest, Support Vector Machines, Neural Networks
• Deep Learning: Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Deep SVDD
• Representation Learning: Feature extraction using HOG, LBP, and learned representations
• Transfer Learning: Pre-trained models for engineering applications

Structural Engineering

• Non-destructive Testing: Impact-echo testing, acoustic emission, vibration analysis
• Wind Engineering: Wind tunnel testing, computational fluid dynamics (CFD), pressure coefficient analysis
• Structural Health Monitoring: Real-time monitoring systems, anomaly detection, damage assessment

Software & Tools

• Programming: Python, MATLAB, C/C++, JavaScript
• ML Frameworks: PyTorch, TensorFlow, Scikit-learn
• IoT & Hardware: Arduino, ESP32, STM32, Teensy
• Engineering Software: ANSYS, AutoCAD, ECO2

Key Research Contributions

1. Automated Impact-Echo Testing

• Developed machine learning models for automatic classification of impact-echo test results
• Achieved reliable detection of shallow delamination in concrete structures
• Published in Journal of Nondestructive Evaluation (2025)

2. Structural Health Monitoring with Deep SVDD

• Introduced multiclass Deep Support Vector Data Description for anomaly detection
• Achieved 88.11% accuracy using only normal condition data (vs. 58.68% with traditional methods)
• Published in ASCE Journal of Structural Engineering (2025)

3. Lightweight Crack Detection Models

• Developed resource-efficient ML models for on-site concrete crack classification
• Balanced accuracy with computational efficiency for edge device deployment
• Published in ACI Structural Journal (2025)

4. Wind Load Estimation

• LSTM-based prediction of wind pressure coefficients for sensor reduction
• Dynamic time warping for wind pressure tap clustering
• Published in multiple conferences and journals

5. LLM for Engineering Standards

• Developed specialized large language model framework for building design codes
• Retrieval augmented generation for wind load design code assistance
• Presented at APCWE10 (2025)

Collaborations

National Weather Center, University of Oklahoma (2025.10-2025.11)

• Lab: Hydrometeorology and Remote Sensing Lab
• Advisor: Prof. Yang Hong
• Focus: Machine learning applications to remote sensing and wind engineering

Hi-performance Structural Engineering Lab, SNU (2021-Present)

• Advisor: Prof. Thomas H.-K. Kang
• Focus: AI applications to structural engineering, concrete structures, wind engineering

Patents

1. AI-Based Internal Defect Detection System for Concrete Members (2023)
   • Applied to PCT, US patent, EU patent
   • Registered in Korea
2. AI-Based Wind Load Estimation System (2022)
   • Applied to PCT
   • Registered in Korea

Research Philosophy

I believe in bridging the gap between cutting-edge AI research and practical engineering applications. My work emphasizes:

• Practical Applicability: Developing solutions that can be deployed in real-world scenarios
• Robustness: Creating models that work across different conditions and datasets
• Efficiency: Balancing accuracy with computational and resource constraints
• Interdisciplinary Collaboration: Working across AI, structural engineering, and wind engineering domains