Hydropower operators in the US may soon have a more reliable way to forecast revenue and plan energy storage. Researchers at Idaho National Laboratory (INL) and Argonne National Laboratory (ANL), with funding from the Department of Energy’s Water Power Technologies Office, have developed HydroBoost, an optimisation software tool designed to make forecasting more realistic.
Hydropower plants earn much of their revenue in the day-ahead electricity market, where utilities purchase power based on expected demand and producers offer electricity at forecasted prices. But predicting water flow and power output is difficult, and existing tools often provide forecasts that are overly optimistic.
“We did a study for Pacific Gas & Electric where we looked at a hydropower plant to see if it was economically viable for them to include a battery,” said Hill Balliet, technical relationships and program development manager at INL. “We got to the end and gave them the results; the analysis said they should build the battery, but the tool we used was overly optimistic about the profits they’d make and assumed they had perfect information about future trends. That unrealistic optimism left too much uncertainty, and they decided not to add batteries to the hydropower plant. That was a big part of the motivation to build HydroBoost. We wanted to build a more realistic and useful tool that was based on how operators actually think and work.”
HydroBoost is the first tool to integrate a seven-day forecasting horizon in its optimisation process. It can be used for both run-of-river and reservoir storage sites. Future updates will include support for cascaded sites and pumped storage plants.
The planned upgrades will allow analysis of factors such as battery size and charge capacity, efficiency, charging rates, water inflows, reservoir levels, and the financial costs of operation. The tool is intended for use by consultants, operators, utility experts, and investors, and was designed to meet technical and financial review standards.
“Operators are missing out on opportunities to improve energy security,” said Balliet. “Batteries, if viable, would decrease prices and improve the reliability of the systems. We saw a gap that needed to be filled and worked with ANL to get it done.”
According to developers, HydroBoost is built to mirror how operators make decisions, offering guidance that reflects realistic scenarios in the day-ahead market.
“HydroBoost’s biggest advantage is its versatility in evaluating different types of planning and decision-making approaches,” said Matt Mahalik, a principal software developer at ANL. “Traditional optimization tools aren’t practical for developing realistic expectations for investment.”
Mahalik added that HydroBoost can reduce the complexity of planning while producing forecasts that match real-world conditions. “HydroBoost can model true-to-life planning approaches with different horizons and uncertainties, mimicking real-world outcomes expected from human schedulers. It provides evaluations of all these approaches with minimal user effort, offering the benefits of multiple models in one.”
Source: Idaho National Laboratory
