Challenge 02

Improve rainfall forecasting and contribute to smarter decisions in hydroelectric power plant operations

Rainfall forecasting is a critical variable for the planning and operation of the National Interconnected System (SIN), especially in subsystems with high hydroelectric generation capacity. However, the global and regional models currently in use have limitations in accurately representing rainfall behavior in Brazilian basins. This misalignment compromises estimates of flow and Natural Inflow Energy, hindering water resource management and strategic decision-making.

Given this scenario, we seek solutions that improve the calculation of forecasts considering three main dimensions: spatial, temporal, and volumetric. The proposal should address two areas of action, starting with the creation of a more accurate daily rainfall forecast model that is compatible with the specific characteristics of the SIN basins, followed by the development of automation capable of identifying, based on the forecast generated, which data series available on CPFL's internal platform best represents rainfall behavior in each subsystem.

Submit your proposal and help with decision-making in the electricity sector with weather forecasts that are more in line with brazilian reality.

• Develop a daily rainfall forecast model with greater spatial, temporal, and volumetric accuracy for the SIN basins

• Automate the identification of historical scenarios that are closer to the forecasts, using CPFL's internal data

• Contribute to the improvement of flow and ENA estimates based on the country's different hydrological profiles

• Structure a platform that allows comparing, organizing, and prioritizing models based on technical and statistical criteria.