Identification of pressure points in modern power systems using transfer entropy

Integration of variable energy resources—e.g., solar, wind, and hydro—and end-use electrification increase the weather dependence of modern energy systems. Understanding how weather-driven stress propagates through interconnected generation, transmission, and demand is therefore essential for understanding risks and guiding adaptation. We use transfer entropy to identify predictive pressure points: grid components whose utilization dynamics provide predictive information about downstream power shortages. Using simulations of New York State's proposed future power grid across a large ensemble of meteorological and technological scenarios, we show that predictive pressure points frequently emerge from complex, system-wide interactions between generation, transmission, and demand. Importantly, these predictive pressure points vary across scenarios and are often difficult to anticipate from high-level scenario features alone. While transfer entropy cannot establish causality or prescribe optimal infrastructure investments, it offers a complementary, data-driven diagnostic perspective on reliability by revealing how and where system stress propagates under realistic operating dynamics.