My broad interests are in the application of statistical methods, stochastic models, and optimization to study extreme events and cyberphysical system resilience. In my PhD, I have developed a decisions-based modeling approach for resilience of electricity networks to extreme weather:
- The first contribution is a stochastic optimization approach for resource allocation and repair of electricity distribution networks that are disrupted by hurricanes. My approach incorporates uncertainties in locations of damage, decisions for allocation of distributed energy resources (DERs), and models of post-disruption power flows in resulting network islands. I also develop algorithmic approaches to solve the optimization problem, which enable decision-making for resilient network operations in a scalable manner.
- The second contribution is a probabilistic, physically-based modeling approach for estimating risk in infrastructure networks due to hurricanes. This approach, which incorporates Forecasts of Hurricanes using Large-Ensemble Outputs (FHLO) and Nonhomogeneous Poisson Process (NHPP) modeling, considers uncertainties in hurricane forecasts and the spatial variability of risk within infrastructure systems. I am focusing on applying this approach to prediction of outages in historical hurricanes (Hermine, 2016; Michael, 2018; Dorian, 2019).
My PhD research is motivated by the increasingly critical threat posed by extreme weather on electrical grids. The rapid modernization of the power grid into a ‘smart grid’ provides government agencies and utilities increased flexibility in handling grid damage induced by hurricanes. Unlike a traditional centralized grid, the smart grid incorporates distributed energy resources (DERs) in the form of portable microgrids, localized renewable energy, storage devices, and electric vehicles. My research leverages the instrumental role of smart grids in achieving reliable, secure, and sustainable energy supplies.
In terms of academic and computational skills, I completed academic coursework in subjects that include machine learning, Bayesian modeling, algorithms for inference, optimization, probability, and random processes engineering. I have coding experiences with Python, MATLAB, Julia, and Java.
I also have a broad interest in data sciences. During the summer of 2019, I interned as a Data Scientist at Spacemaker AI, where I worked on algorithms for object/shape classification in urban point cloud data.
My research has been supported by the National Science Foundation Graduate Research Fellowship Program (NSF GRFP) and the MIT Martin Sustainability Fellowship
Chang, D., Wells, E.M., Trump, B.D., Cegan, J., Linkov, I., & Price, P. "A COVID-19 Micro-exposure Model for Return-to-Work Decision-Making", in preparation.
Chang, D., Amin, S., & Emanuel, K. "Applying Hurricane Ensemble Forecasts for Damage-Aware Resource Allocation in Infrastructure Networks", in preparation.
Chang, D., Amin, S., & Emanuel, K. "Predicting Infrastructural Network Damage from Forecasts of Hurricanes using Large-Ensemble Outputs", working paper targeted for Reliability Engineering and System Safety.
Chang, D., Shelar, D., & Amin, S. "Leveraging DERs and Microgrids against Storm-Induced Failures for Resilient Distribution Networks", working paper targeted for IEEE Transactions of Power Systems.
Chang, D., Amin, S., & Emanuel, K. (2020). "Modeling and Parameter Estimation of Hurricane Wind Fields with Asymmetry", Journal of Applied Meteorology and Climatology.
Chang, D., Shelar, D., & Amin, S. (2020). "Stochastic Resource Allocation for Electricity Distribution Network Resilience", accepted at 2020 American Control Conference, Denver, CO, United States.
Chang, D., Shelar, D., & Amin, S. (2018). "DER Allocation and Line Repair Scheduling for Storm-induced Failures in Distribution Networks", 2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), Aalborg, Denmark.
Chang, D. (2015). "Assessing Impact of the Sulfate Aerosol First Indirect Effect on Tropical Cyclone Activity", M.S. Thesis, Massachusetts Institute of Technology.
Linkov, I., Eisenberg, D., Bates, M., Chang, D., Convertino, M., Allen, J.H., Flynn, S.E., Seager, T.P. "Measurable Resilience for Actionable Policy", Environmental Science and Technology, 2013.
"Predicting Infrastructural Network Damage from Forecasts of Hurricanes using Large-Ensemble Outputs", 10th International Conference on Climate Informatics, September 2020.
"Resource Allocation and Response Strategies for Network Resilience to Extreme Weather", MIT Pierce Research Seminar, July 2020.
"Stochastic Resource Allocation for Electricity Distribution Network Resilience", American Control Conference, July 2020.
“Algorithms for Object and Shape Classification in ALS Point Clouds”, Spacemaker AI, Cambridge, MA, August 2019.
"Uncertainty-Informed Resource Allocation and Repair in Distribution Networks", MIT Lincoln Laboratory, Lexington, MA, December 2018.
“DER Allocation and Line Repair Scheduling for Storm-induced Failures in Distribution Networks”, IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids, Aalborg, Denmark, November 2018.
“Resiliency-Improving Monitoring and Control Networks for Networked Distribution Infrastructures”, MIT Civil and Environmental Engineering Infrastructure Innovation in a Changing Environment Conference, Cambridge, MA, November 2015.
Bayesian Modeling, Machine Learning, Algorithms for Inference, Optimization Methods, Resilient Infrastructure Control, Applied Probability, Random Fields
I am currently an officer in MIT Kickboxing and MIT Kung Fu Taichi Club. My other interests include piano, foreign languages, and hosting Couchsurfers.