Journal of Fluid MechanicsScaling laws for caudal fin swimmers
Direct numerical simulation dataset spanning Reynolds numbers 5,000-100,000, with ongoing BAUV scaling and optimization extensions.
Ph.D. researcher · Johns Hopkins University
From flow physics to collective intelligence.
I'm Ji Zhou, and I also go by Collin. I'm a computational physicist developing high-fidelity simulations and physics-informed multi-agent models to study biological locomotion, hydrodynamic interactions, and collective behavior.
Enter to be a leading fish :)
Selected work
Fluid dynamics, collective behavior, and scientific machine learning.
Projects range from direct numerical simulation of swimming and schooling to physics-informed multi-agent modeling and active learning.
A wake-based model predicts where a trailing fish can gain up to 20% more thrust—without running a new high-fidelity simulation—turning complex schooling hydrodynamics into a practical map for efficient swimming and underwater-vehicle design.
Research role
Simulation design, geometric and kinematic model development, high-performance computing, flow analysis, and manuscript preparation.
Publication record
Read in Journal of Fluid Mechanics · Volume 1014, A32 (2025) ↗
Fish wakes do more than trail behind a school: they actively improve organization, especially when social attraction and alignment are weak. The model reveals how fluid physics can help a large group cohere even when behavior alone is not enough.
Research role
Model formulation, implementation, numerical experiments, parameter analysis, scientific visualization, and manuscript preparation.
02
Methods and directions
From resolved flow physics to interpretable models of collective behavior.
My current research centers on computational physics and collective behavior. The underlying modeling tools may also support future problems in autonomous systems and bio-inspired robotics.
Physics-resolved systems
Biological locomotion, hydrodynamic interactions, acoustics, direct numerical simulation, and scaling.
Interpretable computational models
Domain-informed features, reduced-order dynamics, active learning, and uncertainty-aware evaluation.
Autonomous and bio-inspired systems
Transferable ideas for coordination, interaction-aware modeling, and control in multi-agent systems.
Methods
Immersed-boundary direct numerical simulationPhysics-informed agent-based modelingScaling and reduced-order modelingActive-learning researchCFD → FW–H acousticsLarge-scale simulation workflowsMATLABFortran/MPI solver experiencePythonScientific visualization
03
Selected publications
Selected publications and manuscripts.
Complete record on Google Scholar ↗2026Journal of Fluid Mechanics, Volume 1030, A25Scaling laws for caudal fin swimmers incorporating hydrodynamics, kinematics, morphology and scale effectsDNS dataset generation and validation2026Proceedings of the National Academy of Sciences · under reviewPhysical Origins of Speed Control in Swimming FishesStudy-specific simulation design, execution, and analysis2025Journal of Fluid Mechanics, Volume 1014, A32Hydrodynamically beneficial school configurations in carangiform swimmersSimulation design, analysis, and manuscript preparation2025Physics of Fluids, Volume 37, 011912 · cover articleEffect of hydrodynamic wakes in dynamical models of large-scale fish schoolsModel development, numerical experiments, analysis, and manuscript preparation2024Bioinspiration & Biomimetics, Volume 19, 036015Effect of schooling on flow generated sounds from carangiform swimmersStudy design, simulations, acoustic analysis, and manuscript preparation2024Advances in Neural Information Processing Systems 37ActSort: An active-learning accelerated cell sorting algorithm for large-scale calcium imaging datasetsFeature engineering and research evaluation
04
News & reach
Research beyond the paper.
Selected editorial features, institutional coverage, public exhibitions, and reporting on my research.
Research and collaboration
Computational physics, scientific modeling, and multi-agent systems.
Ph.D. defense planned for December 2026. Exploring research and engineering roles beginning in late 2026 or early 2027.