Distinguished Professor University of California Los Angeles
The Brazilian interconnected power system meets more than 99% of the country's energy demand. From 2002 to 2012, on average, 91% of Brazilian effective electric generation was produced by hydropower, 6% by thermal and 3% by nuclear. Management and operation of the power system changed significantly over the last 11 years due to demand increase, sustained severe and extreme droughts, and impacts from the COVID-19 pandemic. In the last 11 years, there has been an increase in participation of thermal, wind and solar power plants. Very low storage levels in reservoirs lasted from 2012 to 2022 but recovered in 2023. In 2015, wind power production surpassed nuclear. In 2022-2023, wind power production (14%) surpassed thermal (10%) while solar production (3%) surpassed nuclear (2%). Hydropower continues to be the predominant source of energy generation in a more diverse power system. Newly built hydropower plants are mostly of the run-of-the-river type. There has been no expansion in reservoir storage, and operation of the reservoir system is confronted with increasing uncertainties. What lessons can we learn from the many concurrent changes in the system? What are the best planning strategies for the future of the system? To seek answers of some of these questions we use the HIDROTERM model for simulation and optimization. The model is a nonlinear programming optimization model previously developed for planning the operation of the Brazilian hydrothermal system. We compare the results from simulation and optimization runs, using different demand patterns and hydrological input scenarios.
