| Pumped storage
energy systems prioritise dispatchable, zero-emission flexibility.
Rapid response, one platform One of the most powerful features of the module is how
quickly it can respond to changing circumstances. “In terms of operational efficiency, the biggest benefit is reaction speed,” says Goodenough. She gives a practical example: “Suppose a turbine goes offline unexpectedly in a cascade system. That affects multiple other plants in the same hydrological system – when they’ll get water, how much, and what that means for generation.”
With HYDROGRID Insight, the operator can immediately reoptimise the rest of the cascade, update market trading positions, and minimise lost value. “This whole chain – detecting the issue, rebalancing generation, adjusting the power market action, possibly with intraday trading – happens within less than 60 minutes, all within one system,” says Goodenough. “With legacy setups, that could take days and require four or five different tools.” The result is a smoother, more efficient, and more
profitable operation – even under pressure. “It also gives everyone – the maintenance team, the traders, the planners – a shared view of reality. That’s crucial when you need to act fast.”
Intelligence that improves decisions The software’s machine learning capabilities also
deliver significant advantages. “In pump storage, you’re not just deciding how much to generate – you also decide when to pump,”
says Goodenough. “That adds complexity.” Plants may have separate pump and turbine units – or a combined unit that can do both, though not simultaneously. In some cases, both modes may operate briefly at the same time. “The term for this is “hydraulic short circuit,” she
explains. “The need to perform this magical feat of engineering can occur when hydro operators are delivering short-notice grid services and need to have both a pump and a turbine synchronised to the grid to react quickly in both up and down-regulation within a very short time frame of about 90 seconds, which is typical for secondary reserve products (meaning any spinning reserve with automated activation) – these grid reserve products are in most countries called ‘aFRR’ (short for ‘automatic Frequency Restoration Reserve’). So knowing whether pumping and turbine generation is in principle physically possible at the same time is important” HYDROGRID’s system allows for that level of detail. And while it leverages AI, Goodenough emphasises that it avoids generative AI. “We have always used machine learning models, which based on their complexity would now be classified as ‘agentic AI’ – intelligence that combines multiple input data sources to derive optimal actions and outputs in a way that’s testable and traceable. But we do not use large language models or other generalised AI in our product,” she says. “Because this is critical infrastructure. Safety and security are paramount.” Two examples highlight how this intelligence works in practice.
First, the system can detect deviations from
Below: Pumped storage project in HYDROGRID Insight
www.waterpowermagazine.com | July 2025 | 7
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61
