Published in the International Journal on Hydropower and Dams, Vol. 32 - Issue 3, 2025.
Advancing hydropower with AI: Probabilistic inflow forecasting at Cahora Bassa reservoir
Probabilistic inflow forecasting to optimise hydropower operations under uncertainty
Authors : J.P. Matos, Instituto Superior Técnico, University of Lisbon, Portugal; M. Mahunguana, Hidroeléctrica de Cahora Bassa, Mozambique; and, F. Zeimetz and M. Leite Ribeiro, Gruner Stucky SA, Switzerland
Inflow forecasts are essential for the optimal management of reservoirs. Small run-of-the-river plants can benefit from forecasts to make the most of spot energy markets on a short-term scale. For large storage infrastructure, long-term information can be provided to help reap the most benefits of seasonal water transfers. Regardless of the desired time scale, forecasts and uncertainty go hand-in-hand.
Cahora Bassa is located in the Lower Zambezi River, in Mozambique. It has an installed capacity of 2075 MW and holds a massive reservoir, whose active storage is 52 km3. The region is prone to multi-yearly spells of “extreme weather” and has a complex hydrological response, leading to droughts and floods. Using the case study of Cahora Bassa, this paper explores how seasonal inflow forecasting based on artificial intelligence is capable of providing reliable probabilistic forecasts upon which reservoir management decisions can be taken. Despite advances in hydrological modelling capabilities, what can be inferred about streamflow with lead times of several months is limited by the complexity of meteorological patters which, for the time being, meteorological models struggle to capture. In such a setting, the best decisions are those that acknowledge unknowns.
For further information, please contact the following person:
You can also contact the Publisher: