Impact of Natural Disasters on Solar Energy Farms in California

• Tools & Technologies : Python, Pandas, Scikit-Learn, XGBoost, Random Forest, Decision Tree, SVR, Feature Engineering, Data Integration Pipelines
• Role: Data Scientist
• Github URL: Project Link

Description:

Developed an end-to-end predictive modeling framework to quantify how extreme natural disasters—such as wildfires, flooding, and high-wind events—affect the operational efficiency of solar energy farms. Integrated geospatial hazard data, meteorological time-series, and system output records. Executed large-scale preprocessing, advanced feature engineering, and benchmarking of multiple regression and ensemble models. Studied disaster-lag effects and structural performance degradation across varied geographic zones.

Outcome:

Uncovered statistically significant correlations between specific disaster categories and solar-power output drops. Produced a predictive tool for disaster-driven performance forecasting and delivered risk mitigation strategies with applications in energy planning, infrastructure reinforcement, and climate resilience.