Project 4
Data Analysis of the Los Angeles Energy Grid- Victor Bogin
I conducted an analysis of the Los Angeles energy grid to examine how weather conditions influence electricity generation efficiency, energy demand, and market prices. Using publicly available data from the California ISO and the Energy Information Administration, I performed multiple regression analyses to evaluate the relationship between temperature, humidity, generation mix, and electricity prices.
I found that higher ambient temperatures were strongly associated with reduced efficiency of thermal generation units. Specifically, a 10°F increase in average daily temperature was associated with a 2.5–4.0% decline in thermal unit efficiency, while higher humidity levels amplified this effect by an additional 1–2%. During peak summer conditions, these efficiency losses coincided with average wholesale price increases of approximately 15–30% compared to mild-weather days.
Regression results showed that temperature, humidity, system load, and renewable penetration were statistically significant predictors of price outcomes, with the model explaining roughly 55–65% of the variation in hourly electricity prices. Using these inputs, I developed a predictive model to estimate electricity prices per megawatt-hour under varying weather and demand scenarios.
Based on the analysis, I proposed an optimized energy consumption pattern for a representative household of four in Los Angeles County. The model suggested that shifting discretionary usage to off-peak hours and cooler periods could reduce household electricity costs by approximately 10–18% annually, with the largest savings occurring during summer weekday afternoons.
Overall, the project highlights how weather-driven efficiency losses and demand surges affect grid performance and prices, and demonstrates how data-driven modeling can inform both system-level planning and household-level energy decisions.
