● Led a team of three undergraduate research assistants to analyze power distribution systems and operators, collecting key parameters for five energy storage technologies (e.g., power capacity, charge time, self-discharge rate, round-trip efficiency.)● Utilized a linear programming-based revenue maximization model to quantify the revenue potential of ESS in electricity markets, performing arbitrage and frequent regulation services. Demonstrated a negative impact of COVID-19 on ESS revenues, particularly in densely populated areas, with findings supporting industry recommendations for post-pandemic market recovery. add● Applied the Kolmogorov-Arnold Networks (KAN) model for regression to predict Lithium-ion battery capacity across charge cycles and for classification to assess electric vehicle performance under varying external conditions, enhancing predictive accuracy in both domains.● Achieved higher performance with KAN compared to traditional MLP models, demonstrating improvements in MSE, MAPE, accuracy, parameter efficiency, and reduced training data requirements, validating KAN’s advantages in real-world applications.

● Processed 1-Hz GPS data from 211 school buses, totaling over 36,000 tracking hours, using the Polars package in Python. Constructed datasets of micro-trips and trips, calculating travel distances for each segment to generate insights into electric school bus adoption and operational efficiency.● Calculated the energy requirements of electric school buses based on GPS data and external conditions to assess the impact of ESB adoption on power demand, battery range requirements, daily charging opportunities, and necessary charging infrastructure.