Zhou Changjian zhou-changjian

Ph.D. student · The University of Melbourne
_{Faculty of Engineering and Information Technology · Melbourne, Australia}

Academic homepage ↗ · Google Scholar · ORCID · Email

I build physics-aware models that learn from simulations and adapt to real-world data.

Research interests

Physical AI & world models — learned simulators and contact dynamics.
Scientific machine learning — physics-informed learning and sim-to-real adaptation.
Computational mechanics — granular dynamics, constitutive modelling, and inverse problems.

_{01 / SIM-TO-REAL}

Fisher-guided gradient projection for physics-preserving adaptation of neural PDE surrogates.

_{Scientific ML · Neural PDEs}

_{02 / LEARNED SIMULATION}

Contact-memory graph networks for learned simulation of granular dynamics.

_{Graph networks · Computational mechanics}

Open-source contributions · DeepCopilot, a coding agent for VS Code · StructureClaw, LLM agents for structural engineering.

_{Publications, background, and collaborations → zhou-changjian.github.io}