XJU
The Semantic Segmentation-Based Recognition Phase Components within the UHPC under normal and high temperature: Theory, Methodology and Application
✓ Fully Funded
🎓 Artificial Intelligence
🎓 Civil Engineering
🎓 Computer Science
🎓 Engineering
AI
deep learning
UHPC
high-temperature performance
phase components
phase-field modeling
semantic segmentation
sustainable construction materials
ultra-high-performance concrete
This PhD develops an AI-driven semantic segmentation framework to identify phase components in UHPC materials, linking microscale imaging with multiscale modeling. The project investigates UHPC performance under normal and high-temperature conditions and aims to design sustainable, fire-resistant concrete solutions.
Project Description
The research integrates AI, materials science, and civil engineering to enhance UHPC modeling and performance prediction:
Develop deep learning-based semantic segmentation technology to identify and quantify UHPC phase components from microscale digital images.
Use phase-field modeling to predict mechanical properties and failure mechanisms of UHPC under normal and elevated temperatures.
Explore sustainable UHPC materials with hybrid fibres suitable for high-temperature applications and fire-hazard environments.
The project is part of a joint programme with XJTLU, XJTU, and University of Liverpool (UoL). Students will register with XJTLU and may conduct research visits to XJTU and UoL for collaborative supervision and access to research facilities.
Entry Requirements
UK first-class or upper second-class Bachelor’s degree + Master’s with Merit (or equivalent international qualifications). Exceptional Bachelor’s-only candidates may be considered.
Strong spoken and written English (IELTS ≥6.5 or equivalent if first language is not English).
Background in AI, civil engineering, materials science, or a related field.
Strong spoken and written English (IELTS ≥6.5 or equivalent if first language is not English).
Background in AI, civil engineering, materials science, or a related field.
How to Apply
Email jun.xia@xjtlu.edu.cn
or yun.gao@xjtu.edu.cn
with:
CV
Two reference letters
Personal statement outlining interest in the project
English language certificates
Academic transcripts
Verified educational certificates
Master’s dissertation or equivalent writing sample
or yun.gao@xjtu.edu.cn
with:
CV
Two reference letters
Personal statement outlining interest in the project
English language certificates
Academic transcripts
Verified educational certificates
Master’s dissertation or equivalent writing sample
Eligibility
UK/Home
EU
International
Supervisor Profile
DJ
Dr Jun Xia
Xi’an Jiaotong-Liverpool University (XJTLU) – Design School, School of Engineering
Related Opportunities
MUM
MU
MUM
UCL