This project is supported by the U.S. Department of Energy (DOE).

We collaborate with the University of Arizona team (led by Prof. Neng Fan and Dr. Zhong) and Portland General Electric (PGE) to assist cascaded hydropower fleets in achieving greater operational efficiency in modern power systems.

The project uses PGE’s Pelton Round Butte system as the real-world case.

Alt text

Illustration of the Pelton Round Butte Project.

A key deliverable of this project is an analytical future value function that quantifies the hydropower generation (MWh) that water storage (Mm³) can provide in the future, known as future value. This function is user-friendly and interpretable. Given this function, hydropower operators can view the future value surface and access specific results with a simple click.

Alt text

Illustration of the visualized future value function.

Another key deliverable is a medium-term planning approach designed for renewable-integrated cascaded hydropower systems.

Alt text

Illustration of the Pelton Round Butte Project integrated with variable renewables.

The medium-term planning method is built upon deep reinforcement learning, offering two key advantages:

  1. It leverages short-term contextual information, such as day-ahead electricity prices and renewable forecasts, to determine the medium-term planning strategy.
  2. The quality of the medium-term planning strategy is measured by its ability to improve profits in short-term operations. Obviously, it involves the closed-loop predict-then-optimize idea.

Alt text

The existing open-loop relationship between mid-term planning and short-term operations.

Alt text

The presented closed-loop relationship between mid-term planning and short-term operations.

Related Papers

  • [1] Xianbang Chen, Yikui Liu, Zhiming Zhong, Neng Fan, Zhechong Zhao, Lei Wu. "A carryover storage valuation framework for medium-term cascaded hydropower planning: A Portland General Electric system study," IEEE Transactions on Sustainable Energy, 2025. [PDF »]
  • [2] Xianbang Chen, Yikui Liu, Neng Fan, Lei Wu. "An operating profit-oriented medium-term planning method for renewable-integrated cascaded hydropower," Energy, 2025. [PDF »]
  • [3] Yikui Liu, Xianbang Chen, Neng Fan, Zhechong Zhao, Lei Wu. "Stochastic day-ahead operation of cascaded hydropower systems with Bayesian neural network-based scenario generation: A Portland General Electric system study," International Journal of Electrical Power & Energy Systems, 2023.
  • [4] Zhiming Zhong, Neng Fan, Lei Wu. "Multistage stochastic optimization for mid-term integrated generation and maintenance scheduling of cascaded hydroelectric system with renewable energy uncertainty," European Journal of Operational Research, 2024.
  • [5] Zhiming Zhong, Neng Fan, Lei Wu. "Multistage robust optimization for the day-ahead scheduling of hybrid thermal-hydro-wind-solar systems," Journal of Global Optimization, 2024.
  • [6] Zhiming Zhong, Neng Fan, Lei Wu. "A hybrid robust-stochastic optimization approach for day-ahead scheduling of cascaded hydroelectric system in restructured electricity market," European Journal of Operational Research, 2023.