HPC software gives renewable energy a boost
One hundred and forty wind turbines in the North Sea – and a few GPUs in the cloud – pumped wind under the wings of David Standingford and Jamil Appa’s dream.
As colleagues at a UK aerospace company, they shared a vision of building a company to apply their expertise in high performance computing across many industries.
They formed Zenotech and coded what they learned about computational fluid dynamics into a program called zCFD. They also created a tool, EPIC, that made it easier to run it and other HPC tasks on the latest hardware in public clouds.
But gaining visibility beyond their home in Bristol, UK, was a challenge, given the lack of large open datasets to show what their tools could do.
Collect data from a wind farm
Their problem is similar to that of the wind sector.
As government subsidies for wind farms dwindled, investors demanded a deeper analysis of a project’s likely return on investment than traditional tools allowed. A UK government project has partnered Zenotech with renewable energy consultancy firms and SSE, a major UK utility to share data on its North Sea wind farm, one of the largest in the world.
Using zCFD, Zenotech simulated the likely energy output of the farm’s 140 turbines. The program represented dozens of wind speeds and directions. This included key but previously untracked phenomena in front of and behind a turbine, such as the combined effects that turbines had on each other.
“These so-called ‘wake’ and ‘blocking’ effects around a wind farm can even impact air currents,” said Standingford, director and co-founder of Zenotech.
The program can also track small but significant effects of terrain on wind, such as when trees in a nearby forest lose their leaves.
SSE confirmed that the final simulation was within 2% of the data measured by the utility, giving zCFD a stellar benchmark.
Accelerated in the Cloud
Best of all, Zenotech has shown that cloud GPUs deliver fast and cost-effective results.
For example, the program ran 43x faster on NVIDIA A100 Tensor Core GPUs than on CPUs at a quarter of CPU costs, the company reported in a recent GTC session (viewable on demand). NVIDIA NCCL libraries that accelerate communications between GPU systems further improved results by up to 15%.
As a result, work that took more than five hours on CPUs ran in less than 50 minutes on GPUs. The ability to analyze the nuanced effects of wind in detail and complete a report in a day “got people’s attention,” Standingford said.
“The tools and computing power needed to perform high-fidelity simulations of wind projects are now affordable and accessible to the entire industry,” he concluded in a report on the project.
Reduce carbon, mitigate climate change
Wind power is one of the largest and most cost-effective contributors to reducing carbon emissions, notes Farah Hariri, technical lead of the team that helps NVIDIA customers manage their transitions to net-zero emissions. .
“By modeling both wake interactions and blocking effects, zCFD helps wind farms extract the maximum energy for minimum installation cost,” she said.
This kind of quick but detailed analysis reduces risk for investors, making wind farms more economically attractive than traditional energy sources, said Richard Whiting, partner at Everose, one of the consultants who worked on the project. with Zenotech.
Looking ahead, Whiting estimates that more than 2,100 gigawatts of wind farms could be in operation worldwide by 2030, 3 times more than in 2020. This is a growing opportunity on many levels.
“Going forward, we expect projects to use larger arrays of larger turbines, so the modeling challenge will only increase,” he added.
Less climate change, more business
Helping renewable energy projects get off the ground has also put the wind in Zenotech’s sails.
Since the HSE analysis, the company has contributed to the design of wind farms or turbines in at least 10 other projects in Europe and Asia. And half of Zenotech’s business now takes place outside the UK
As the company grows, it is also revisiting its aerospace roots, paving the way for new types of drone and air taxi businesses.
A parallel opportunity
For example, Cardiff Airport is supporting live trials where emerging companies are using zCFD on cloud GPUs to predict wind shifts in urban environments so they can map safe and efficient routes.
“It’s a forward-thinking way to use today’s managed airspace to plan for future services like air taxis and automated airport inspections,” Standingford said.
“We are seeing a lot of innovation in small aircraft platforms, and we are working with the best platform manufacturers and key sites in the UK”
It’s one more way for the company to keep its finger on the wind.