Yongqing Jiang (江永清) — SCU-NTU Joint PhD Student in AI for Civil Engineering

👤 About Me

Hi! I’m Yongqing Jiang, a joint PhD student at Sichuan University (SCU) and Nanyang Technological University (NTU), under the supervision of Prof. Kaoshan Dai and Prof. Zhiqi Shen. My research resides at the intersection of Civil Engineering and Computer Science. Prior to my doctoral studies at SCU, I gained significant research experience at the Shandong Key Laboratory of Intelligent Building Technology.

My research interest focuses on applying artificial intelligence and deep learning to intelligent infrastructure systems, which can be divided as follows:

  • Structural Health Monitoring (SHM).
  • AI and Data Science in Engineering.
  • Large Language Models (LLMs) & Vision-Language Models (VLMs).
  • Intelligent and Resilient Infrastructures.

Email: yongqingjiang97@gmail.com

Note: I am currently open to academic opportunities. Please feel free to contact me regarding any available positions.

Total Citations
h-index
i10-index
View on Google Scholar → Auto-updated every 6 hours Last sync:

🔥 News

📝 Selective Publications

Arxiv 2026
Rethinking Scientific Modeling: Toward Physically Consistent and Simulation-Executable Programmatic Generation

Rethinking Scientific Modeling: Toward Physically Consistent and Simulation-Executable Programmatic Generation

Yongqing Jiang, Jianze Wang, Zhiqi Shen, Kaoshan Dai, Haoran Luo

Project

  • The Problem: LLMs often struggle with the "physical common sense" required for structural engineering, producing code that looks right but fails in simulation. The Solution: I developed a framework that enforces structural dynamics consistency through CivilInstruct (a domain-specific dataset) and MBEval (a verification-driven benchmark). Impact: This approach ensures simulation-ready code generation, bridging the gap between high-level AI and stringent engineering constraints.
Automation in Construction 2026
Post-earthquake structural damage assessment

Multitask Unified Large Vision-Language Model for Post-Earthquake Structural Damage Assessment of Buildings

Yongqing Jiang, Jianze Wang, Xinyi Shen, Kaoshan Dai, Qingzi Ge

Project

  • We develop a multitask unified vision-language model for post-earthquake structural damage assessment, supported by instruction-based multi-task learning, a two-stage assessment framework, and a large-scale multi-attribute image–text dataset.
Computer-Aided Civil and Infrastructure Engineering 2025
🔥 Featured Cover Large language model for post-earthquake structural damage assessment of buildings

Large language model for post-earthquake structural damage assessment of buildings

Yongqing Jiang, Xinyi Shen, Jianze Wang, Kaoshan Dai

Project

  • We developed SDA-Chat, a multimodal Large Language Model (LLM) that automates professional post-earthquake structural damage assessment through multi-round vision-language interactions, achieving 83.04% accuracy with high inference efficiency.
Engineering Structure 2025
A data-driven approach for predicting peak floor response based on visually observed rocking behaviors of freestanding NSCs

A data-driven approach for predicting peak floor response based on visually observed rocking behaviors of freestanding NSCs

Yongqing Jiang, Jianze Wang, Weiwei Chen, Kaoshan Dai

Project

  • We propose a vision-based framework leveraging machine learning to infer Peak Floor Acceleration (PFA) ranges from the seismic responses of unanchored non-structural components, providing a cost-effective alternative to traditional monitoring with 84%–94% accuracy.
Engineering Structure 2024
Video comprehension-based approach for seismic damage recognition of freestanding non-structural components

Video comprehension-based approach for seismic damage recognition of freestanding non-structural components

Yongqing Jiang, Jianze Wang, Xingquan Guan, Kaoshan Dai

Project

  • We developed TPViT-DMSR, a customized two-pathway vision transformer designed to recognize the seismic damage states of freestanding non-structural components through video analysis, achieving a mean average precision of 74.87%.
Automation in Construction 2021
A deep learning approach for fast detection and classification of concrete damage

A deep learning approach for fast detection and classification of concrete damage

Yongqing Jiang, Dandan Pang, Chengdong Li

Project

  • We developed TPViT-DMSR, a customized two-pathway vision transformer that leverages video-comprehension and trajectory attention to recognize the seismic damage states of freestanding non-structural components with a 74.87% mean average precision.
Computer-Aided Civil and Infrastructure Engineering 2023
A method of concrete damage detection and localization based on weakly supervised learning

A method of concrete damage detection and localization based on weakly supervised learning

Yongqing Jiang, Dandan Pang, Chengdong Li, Jianze Wang

Project

  • We developed a CNN-based framework for automated concrete defect detection and sensor-free geographical localization, leveraging projection loss and feature vectorization to achieve pixel-level precision and 83.69% localization accuracy.
Engineering Structures 2023
An adapted LSTM-DRRNet approach for predicting floor acceleration response spectrum

An adapted LSTM-DRRNet approach for predicting floor acceleration response spectrum

Jianze Wang, Yongqing Jiang, Qinyong Huang, Xingquan Guan, Kaoshan Dai

Project

  • We developed a novel deep-learning framework integrating ACN-BiLSTM and DRRNet to provide generalized and efficient estimations of nonlinear floor response spectra with 97.29% accuracy, significantly advancing the seismic design and safety of non-structural components.
Engineering Applications of Artificial 2024
🔥 ESI Highly Cited
Improved YOLOv8-GD deep learning model for defect detection in electroluminescence images of solar photovoltaic modules

Improved YOLOv8-GD deep learning model for defect detection in electroluminescence images of solar photovoltaic modules

Yukang Cao, Dandan Pang, Qianchuan Zhao, Yi Yan, Yongqing Jiang, Chongyi Tian, Fan Wang, Junlin Li

Project

  • We propose YOLOv8-GD, a lightweight and accurate defect detection algorithm that optimizes the backbone with DW-Conv and GSConv, and incorporates a BiFPN structure, achieving a 16.7% reduction in model size while improving mAP@0.5 by 4.2%.

🎖 Honors and Awards

  • 2025 China Scholarship Council (CSC) Scholarship for Joint Ph.D. Students (Fully Funded / National Level)
  • 2025 Nominee for the Young Elite Scientist Sponsorship Program (PhD Student Special Track)
  • 2025 Sichuan University PhD Innovation Scholarship (Top 1%)
  • 2024 Sichuan University PhD Innovation Scholarship (Top 1%)
  • 2024 Outstanding Graduate of Sichuan University (Top 1%)
  • 2024 Sichuan University First-Class Scholarship (Ranked 1st in Major)
  • 2022 National Scholarship by Ministry of Education of China (Top 1%)
  • 2021 Outstanding Graduate of Shandong Province (Top 1%)
  • 2021 First Prize of Outstanding Graduate Achievement Award in Shandong Province (Top 3%)

🏆 Competitions

💬 Invited Talks

  • 2024.04 "A data-driven approach for estimating peak floor response based on damage of non-structural components." The 4th International Conference on Vulnerability and Risk Analysis and Management (ICVRAM-ISUMA 2024), Tongji University, Shanghai, China.
  • 2023.08 "The 2nd Workshop on Big Data in Civil Engineering." Zhejiang University, Hangzhou, China.
  • 2023.05 "The 3rd Academic Conference on Computational and Simulation Technologies in Civil Engineering." Guangxi University, Nanning, China.

🚀 Projects

  • NSFC Intelligent Recognition of Indoor Seismic Damage Characteristics and Building Resilience Assessment Based on Video Understanding — Key Researcher
  • NSFC Mechanisms and Applications of Fiber Bragg Grating–Based Acoustic Emission Sensing for Concrete Durability Damage Monitoring — Key Researcher
  • NSFC Multi-Model Ensemble–Driven Precise Profiling of Office Building Operational States and Integrated Energy-Efficiency Optimization — Key Researcher
  • NSFC Swarm-Intelligence–Based Bio-Inspired Control of Robotic Fish Schools for Navigation in Unknown Aquatic Environments — Key Researcher
  • NSFC Coupled Optimization and Dual Control of Vibration and Wind-Induced Responses for Novel Structural Systems of Ultra-High Wind Power Towers — Key Researcher
  • Provincial Outstanding Young Innovation Team in Artificial Intelligence and Building Intelligence, Shandong Provincial Higher Education Institutions — Key Researcher

🌍 Visitors

Readers from around the world