Grid tech company envelio has developed a GPU-based power flow solver that it says cuts the time required for annual time-series simulations from days or weeks to less than 30 seconds.

This achievement is noteworthy because it is up to 20,000 times faster than industry-standard processes relying on desktop grid calculation tools, the company said. Utilities can now run continuous, large-scale grid simulations as part of their daily operations for the first time –essentially changing how grid capacities are assessed, unlocked, and managed.

The timing could not be more critical, envelio observed, noting that more than 2,500 gigawatts (GW) of generation, storage, and large industrial projects are currently waiting in grid connection queues worldwide – and that figure only counts large-scale requests in the megawatt range. Data centers critical for AI infrastructure, EV charging networks, heat pumps, and renewable generation assets are all experiencing the same bottleneck.

The company’s new GPU-based power flow solver aims to solve this problem. “Transparent and flexible grid management depends on complex time-series simulations, a process that traditionally took days or even weeks with legacy desktop grid calculation tools and fragile scripting around the solver core,” said Dr. Simon Koopmann, CEO of envelio, in a statement. “Our new solver reduces annual time-series simulations to less than 30 seconds, delivering performance that is up to 20,000 times faster than processes relying on conventional desktop solvers.”

Utilities can use the high-performance solver to simulate in real time how power grids will behave when new loads such as data centers, electric vehicle (EV) charging stations, heat pumps or generators such as photovoltaic (PV) systems, wind parks, or battery storage systems want to get connected, envelio said. The solver enables new use cases when combined with envelio’s proprietary intelligent grid platform’s (IGP) underlying digital twin of the power grid, the company said.

In contrast with black-box, AI-based estimation approaches, the new technology aims to solve the real physics of the grid on a digital grid model, envelio said. “This gives utilities results that their engineers can trust while making large-scale scenario simulations, faster connection assessments, more flexible grid operation and more robust grid investment decisions part of everyday practice,’’ the company maintained.

Grid operators are under pressure to evaluate an increasing number of connection requests, scenarios and flexibility options under tight timelines. The GPU-based solver is designed to transform grid planning by replacing periodic, worst-case-based studies with continuous, time-series-driven decision-making. “This allows grid operators to make better use of existing capacity and invest more effectively, which ultimately helps to keep the energy transition affordable,” Koopman said.

The GPU-based solver enhances the functionality of the existing IGP Grid Hub. It has been integrated into select time-series and hosting capacity workflows and is currently available, envelio said. Further expansion to additional use cases is planned throughout the year.