School of Science Department of Physics 63 Physics-guided Data-driven Modeling to Understand Complex Phenomena and to Solve Real-world Problems Supervisor: ZHANG Rui / PHYS Co-supervisor: LI Sai Ping / PHYS Student: CHENG Chun Hei / PHYS Course: UROP 3200, Fall Resistor networks are studied using a network science approach. We studied the change of voltages at each node in different resistor networks after removing resistors. A deep neural network is used to predict the location of removed resistors in a resistor network with the voltage changes from a few nodes as input data. A real resistor network is constructed to verify the computational results experimentally. Physics-guided Data-driven Modeling to Understand Complex Phenomena and to Solve Real-world Problems Supervisor: ZHANG Rui / PHYS Co-supervisor: LI Sai Ping / PHYS Student: LI Hung Yu / PHYS Course: UROP 2100, Fall In this report, the Vicsek model under the presence of obstacles and finite size effect of active matter are investigated. It is found that after adding the alignment collision rule, the fundamental nature of the Vicsek model is changed, the orderliness is no longer monotone decreasing with a higher noise coefficient, and it start to appear some turning points in the phase plot, while it the finite size of collision between the active matter can help to preserve the property of Vicsek model, this is because that the obstacles can help to gather the matter to form cluster under ambient noise condition and increase the alignment. Physics-guided Data-driven Modeling to Understand Complex Phenomena and to Solve Real-world Problems Supervisor: ZHANG Rui / PHYS Co-supervisor: LI Sai Ping / PHYS Student: LIU Ka In / PHYS-IRE Course: UROP 1000, Summer This report presents a mathematical model for describing the evolution of an elastic ring confined to a cylindrical surface. The equilibrium shape of the ring as well as the temporal evolution are calculated using numerical simulation. Our main conclusion is that the ring tends to elongate along the long axis of the cylindrical surface. We expect our work will be useful for examining the behavior of soft robots moving on a curved surface. Physics-guided Data-driven Modeling to Understand Complex Phenomena and to Solve Real-world Problems Supervisor: ZHANG Rui / PHYS Co-supervisor: LI Sai Ping / PHYS Student: ZHU Xiaohang / MAEC Course: UROP 1100, Fall In this report, I explored the pathway of comments and replies in a stock bar within a curtain time. Methods like SIR model and the Planar Maximally Filtered Graph (PMFG) model are being used. Using weekly data from January 2017 to September 2021, I construct a complex network inside stock bar based on the users’ comments and replies.
RkJQdWJsaXNoZXIy NDk5Njg=