Boundary Conditions for Hydrothermal Operation Planning Problems: The Infinite Horizon Approach

Multistage stochastic optimization models are used in the energy sector for a variety of planning and operational decisions. Traditional models optimize over a finite planning horizon, which requires setting up ad-hoc terminal conditions, in many cases given as a terminal cost function which induces the water value along the planning horizon. We propose the use of periodic, infinite horizon models, which are better approximations of the long-term behaviour of the system, as a framework to calculate such terminal cost functions. The periodic nature of the model allows for a more efficient algorithm, and leads to a method that is naturally adapted to the system configuration. This is especially useful for the operation planning of hydrothermal power systems, where the water value is a key component of the operation strategy, as well as for the expansion planning of the system.

Recommended citation: BFP da Costa et al. "Boundary Conditions for Hydrothermal Operation Planning Problems: The Infinite Horizon Approach". XLIII Congresso Nacional de Matemática Aplicada e Computacional, Porto de Galinhas, PE, 16–20 Setembro 2024.
Download Paper