School of Science Department of Physics 70 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: LAW, Suet Yiu / SSCI Course: UROP1000, Summer Active solids like biofilms formed by bacteria are found to have unique properties, which are not found in ordinary elastic materials. From previous studies, bacteria in biofilms will move themselves, resulting in an organized global oscillatory motion. To further investigate the formation and steadiness of this phenomenon, more simulations were carried out. The aim of this project is to find out the effect of boundary shape and variations in mass element activity on oscillatory motion modes. In the simulations, the oscillation modes emerged as expected in previous findings, there were only some minor differences in their motion patterns. These investigations might help to make practical applications on active solids. 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: UROP1100, Spring This report will mainly study the active matter concentrated in fish schooling, it is found that the fish are able to adjust their movement and swim together in a large cluster. This report will study how the modeled fishes moved under Vicsek model and complex network model in a given area. We present the evolution under the condition for (i) no obstacles, (ii) with repulsive obstacles, (iii) with rigid obstacles in fixed boundary and periodic boundary. It is found that the Vicsek model and network model behave similarly in the no obstacles cases, but the network model shows a stronger alignment power than the Vicsek model with the presence of obstacles. It is believed that the Vicsek model connect the active matter locally but network model can connect the matter globally to have such differences. 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, Yifan / SENG Course: UROP1000, Summer In this work, we present a comprehensive study of decade-long (2011-2021) wildfire data in Alberta Province, Canada. By replicating a previous analysis of Alberta’s wildfire and utilizing their Cellular Automaton modeling methodology, we verify the scale-free properties of wildfire spread areas. From 1960 to 2010, Alberta’s wildfire area exhibited a power law distribution. Our findings for the 2011-2021 period align closely with the outcomes of the earlier analysis. However, in the section on model replication, discrepancies arise when compared to the previous research. Reasons and suggestions for improvement are provided in the discussion section. Linked to this UROP project and inspired by complex network theory, we contribute a discussion of the intricate interplay between wildfire dynamics and network characteristics in environmental phenomena.
RkJQdWJsaXNoZXIy NDk5Njg=