Hydro-economics tradeoff surfaces to guide unit commitment in production cost models
Quentin Ploussard, T.D. Veselka, Konstantinos Oikonomou, Nathalie Voisin
Abstract
Production cost models simulate optimal operations that reliably balance load and generation at the power system scale. These models typically minimize total system production costs under several grid and generation constraints. In regions where hydropower is not a significant share of the overall generation portfolio, it is most often represented as a no-fuel-cost technology with monthly hydropower targets but no direct modeling of sub-monthly water operations. On the other hand, hydropower plant operators generally rely on hydro scheduling tools (also called “hydro schedulers”) that optimize day-ahead unit commitment to maximize revenues at the power plant scale while satisfying complex hydraulic and environmental operating criteria. Integrating hydro scheduler tools in production cost models creates a computational challenge as well as a data challenge. We propose a new method that relies on hydro-economic surfaces (or bivariate functions) derived from a hydro scheduler tool to guide power dispatch in production cost models and to address those challenges. Surfaces are constructed under many hydrologic conditions (around 100 runs per surface). These parameter surfaces inherently include all the detailed water considerations that are endogenously incorporated into the hydro-scheduling tool. For an integrated set of hydropower plants located in the Upper Colorado River Basin, we demonstrate that the use of parameter surfaces improves the accuracy of dispatch and revenue calculations in production cost models. The approach is expected to support the industry in more accurately communicating the potential needs from new hydropower operations during energy transitions and accordingly designing markets to incentivize those needed operations.