Shyft

Shyft is an open-source, high-performance simulator for distributed operational hydrological forecasting, designed to integrate large datasets for improved water balance assessment in hydropower production

In storage hydropower systems, water is kept in reservoirs upstreams of the power plants. By opening the water ways, electricity can be produced on demand, as long as there is water in the reservoirs.

To maximize the value of the hydro power production, knowledge about how much and when the water flows into reservoirs from the catchment areas around them is essential. This is especially important during the snow melting season.

Expert Analytics is one of the major contributors to the open source platform Shyft!

Approach

We developed Shyft as a joint effort with Statkraft, combining data-driven and physics-based modeling to more accurately predict the water flow in the catchment areas. Shyft is able to fully utilize distributed weather and terrain data, to create an accurate hydrological state. This makes it possible to model snow smelting and to adapt new weather patterns, e.g., caused by climate changes. To maximize performance, Shyft is implemented it as a hybrid system (C++/Python).

Impact

Shyft is released as an Open Source Software, under the LGPL license. Both the Norwegian hydro power industry and academia use it extensively and contribute to its further development.

Resources

Github Source Code

Documentation