Ziqi Wang

Wang's research focuses on analyzing and understanding the reliability, risk, and resilience of structures and critical infrastructures under hazards. He is also interested in applying probabilistic methods to a broader field of science. Here are a few of the research areas Wang is currently working on:

Computational reliability and uncertainty quantification methods leveraging domain-specific knowledge of civil engineering

An optimal (efficient, accurate, interpretable, scalable, general) computational method does not exist when a wide spectrum of problems is considered. Domain knowledge should be embedded into the design of computational methods for specific problem classes. In many civil engineering applications, rather than resorting to end-to-end machine learning, we find that tuning simplified physical models with self-correcting mechanisms often yields superior and more reliable results.

• [2506.10569] A composition of simplified physics-based model with neural operator for trajectory-level seismic response predictions of structural systems
• [2402.04582] Dimensionality reduction can be used as a surrogate model for high-dimensional forward uncertainty quantification
• [2405.08006] Physics-based linear regression for high-dimensional forward uncertainty quantification (arxiv.org)
• [2310.00261] A physics and data co-driven surrogate modeling method for high-dimensional rare event simulation (arxiv.org)
• [2303.13023] Relaxation-based importance sampling for structural reliability analysis (arxiv.org)
• [2208.08991] Optimized Equivalent Linearization for Random Vibration (arxiv.org)
• [2310.10232] Efficient seismic reliability and fragility analysis of lifeline networks using subset simulation (arxiv.org)

Statistical physics of civil systems under natural hazards

Even the most complex systems can reveal striking regularities when examined through the right lens. Statistical physics provides a formal apparatus for this, allowing us to analyze how the macroscopic behavior of many-body systems emerges from component-level interactions. This approach seeks a global understanding of complex systems by analyzing the tension between unstructured random fluctuations and structured organizing forces. As rapid urbanization pushes civil infrastructure systems to unprecedented scales and complexities, applying these principles will become foundational to the future of civil engineering practice and education.

• [2602.16195] Phase Transitions in Collective Damage of Civil Structures under Natural Hazards
• [2603.29282] Social Amplification Dominates Collective Hazard Response
• [2403.11429v2] Long-range Ising model for regional-scale seismic risk analysis (arxiv.org)
• [2310.17798] Maximum entropy-based modeling of community-level hazard responses for civil infrastructures (arxiv.org)

Other research activities related to probabilistic methods 

• [2403.00283] Risk Twin: Real-time Risk Visualization and Control for Structural Systems
• [2404.07323] Surrogate modeling for probability distribution estimation:uniform or adaptive design?
• Surrogate modeling for identifying critical bridges in traffic networks under earthquake conditions - ScienceDirect
• [2605.25261v1] A Statistical Physics View of the S&P 500: Pairwise Interactions and Time-Varying Dynamics
• [2502.19549] Recorded Versus Synthetic Spectral-compatible Ground Motions: A Comparative Analysis of Structural Seismic Responses
• [2007.04136] The dynamics of entropy in the COVID-19 outbreaks (arxiv.org)
• (in Chinese) https://www.bilibili.com/video/BV1wY4y1y7aX/?spm_id_from=333.999.0.0
• Dynamics of severe accidents in the oil & gas energy sector derived from the authoritative ENergy-related severe accident database | PLOS ONE