HydroForecast
/
Use cases

Hydropower generation

Maximize revenue, optimize generation and minimize spill with forecasts directly at a fleet’s assets.

Hydropower is poised to be the backbone of our nation’s future low-carbon grid due to its storage capacity, existing infrastructure, and renewable nature.

To effectively fulfill its promise as the most reliable renewable baseload, hydropower operators require the most accurate and up-to-date hydrological forecasts to make optimal operational decisions.

HydroForecast’s Short-term hourly 10-day forecasts and Seasonal 90-365 day forecasts enable:

  • Prompt flood control measures earlier while avoiding false alarms
  • Optimization of generation and release schedules generating more aggressively while safeguarding seasonal storage and lake-level targets
  • Minimize revenue loss from spillage
  • Improve regulatory compliance - maintain precise flow requirements

“HydroForecast has led to a more proactive approach to the short- and mid-term planning, an improved probabilistic view of generation and enhanced trading and operational decision-making. HydroForecast has been a truly successful project.”

Samiha Tahseen

Brookfield Renewable

Virtual series

The virtual series: Decoding AI & Hydrology for Water Management Decisions covers an introduction to AI and machine learning and how it can be used in hydrology. The series discusses strengths and limitations of the technology, and the role it can play in advancing water management science.

Tune into our recorded live Q&A sessions in which our team of experts answer your questions from the series. The discussion will center on the difference between machine learning vs. conceptual models, how it applies to streamflow forecasting, and the benefits of a theory-guided machine learning model.

Trusting the science

In the world of hydrology, a basin’s streamflow response from rain-on-snow events can be particularly difficult to predict. These can be critical events for water managers, hydropower generators and other water system operators because they can cause unexpected runoff events and indicate the start or end of the snowmelt season, both of which need to be closely managed.  

HydroForecast’s unique combination of physical and statistical modeling approaches in a neural network framework has created a model that is better able to handle the uncertainty of rain-on-snow events and provide more accurate forecasts to customers to help de-risk their operational decision-making. Read our case study here for more information.

Powering hydropower flexibility

Lo, there is another kind of energy that fills this role! Hydropower is poised to be an important component of a renewable grid because of its storage via dispatchability and the fact that it… already exists! Its fuel, instead of natural gas’ methane, is the potential or kinetic energy of water. Insight into the future of how much “fuel” (read: water) will be available in the future is crucial for hydropower’s flexibility. We’re proud to support hydropower’s role in a 100% renewable grid with HydroForecast.