Dissertations and Theses
Date of Award
2024
Document Type
Dissertation
Department
Mechanical Engineering
First Advisor
Jorge E. Gonzalez-Cruz
Second Advisor
Yang Liu
Keywords
Resiliency, Hurricanes, Power Infrastructure, Machine Learning
Abstract
This doctoral dissertation focuses on the resilience of power transmission infrastructure in tropical coastal environments, particularly in the face of extreme weather events such as hurricanes. The research is anchored on the case of the passage of Hurricane Maria in the Island of Puerto Rico in September of 2017 which caused the largest damage on the power infrastructure in US history. As such, the research investigates the interaction between extreme winds and power transmission infrastructure in complex terrain, aiming to quantify and predict power loss and damage to the infrastructure during such events. The study employs a comprehensive approach combining numerical weather prediction models, machine learning techniques, satellite-based observations, computational fluid dynamics simulations, wind tunnel experiments, and socio-technical vulnerability assessments.
The research examines the historical impact of extreme weather events on the power systems, highlighting the vulnerabilities and the need for robust resilience planning. It introduces a novel methodology for forecasting power loss using satellite-based nighttime lights data, offering a valuable tool for regions where traditional outage information is limited. The study also quantifies the damage to transmission lines caused by Hurricane Maria and develops a methodology to assess the effectiveness of various hardening strategies.
Furthermore, this study investigates the intricate wind patterns in mountainous terrain, employing high-resolution Large Eddy Simulations and wind tunnel experiments to estimate mechanical drag effects on power towers and incorporate these findings into a predictive model. Additionally, a socio-technical analysis is conducted to evaluate the impact of critical and interconnected infrastructure upgrades on vulnerable communities using Western Puerto Rico as testing case, providing insights into the broader implications of such improvements.
This research contributes to a deeper understanding of extreme weather-structure interactions in complex terrain, especially in hurricane-prone areas. The findings provide valuable insights and tools to improve the resilience of power transmission systems, making them better equipped to handle the challenges of increasingly frequent and severe weather events in tropical coastal regions due to a warming global and regional climate.
Recommended Citation
Montoya Rincon, Juan P., "Understanding the Impacts of Extreme Weather on the Power Transmission Infrastructure: A Machine Learning Approach to Quantifying Risks and Enhancing Grid Resilience" (2024). CUNY Academic Works.
https://academicworks.cuny.edu/cc_etds_theses/1294
