Next-gen nuclear

Analyzing real-world data in virtual twins helps Kairos Power accelerate clean energy development

Dan Headrick
31 March 2021

3 min read

Microsoft cofounder-turned-sustainability-advocate Bill Gates recently called nuclear power the world’s best hope for generating enough net-zero-carbon power to meet global energy demand by the UN target date of 2050. Compass talked with Lou Martinez Sancho, executive vice president of Strategy & Innovation for Alameda, California-based Kairos Power, about the company’s work to commercialize a new generation of nuclear technology – fluoride salt-cooled high-temperature reactors – using rapid prototyping and iterations alongside virtual technology to accelerate the demonstration of real hardware.

Compass: KP-FHR advanced nuclear reactor technology entails completely novel design-build models. Please talk about the scale and scope of innovation required for that kind of undertaking.

Lou Martinez Sancho: We have observed that the conventional nuclear development cycle is very long and slow, and very capital intensive. The conventional cycle is ‘plan, design, build, test,’ which is common for other industries. But what is very difficult in nuclear is that you create kind of an endless cycle between ‘plan-design, plan-design,’ and at the end we have a lot of paper-designed reactors. In the industry we call them ‘paper reactors,’ because we have never built them.

So, at Kairos we asked, ‘How can we move forward from that plan-and-design cycle into really building?’ When we begin construction and manufacturing, that is where the real problems become visible, allowing us to solve problems while building solutions.

We learn from testing on pieces of hardware that we have designed and built, and introduce those learnings in the next iteration. This is a huge disruption in the nuclear industry. We are mirroring SpaceX, as we have looked closely at how they improved and accelerated their development phase.

Kairos Power’s technology includes never-before attempted features that demand rapid feedback testing and experimental evidence, proving they work as expected. How do you use virtual modeling in that process?

LMS: Fail fast so you learn fast. That’s the philosophy of rapid learning.

We are in a game-changing situation on the innovation side, and those [virtual] tools have improved tremendously in recent years. We find value in virtual models when coupled with physical hardware. If you want to avoid that cycle of ‘plan and design,’ our approach of leveraging multiple design-build-test cycles with nuclear and non-nuclear systems allow us to link real data into the virtual models. Linking real-world demonstrations with virtual models allows us to develop physical systems that are parallel with their digital counterparts.

In my experience, if there is not a strong connection between the building you are doing and its virtual or digital counterpart, you don’t get the benefit of either.

Top image: The Engineering Test Unit Demonstration Experiment (ETUDE) test stand in Kairos Power’s Alameda Rapid Analysis, Prototyping and Iterative Design Laboratory (R-Lab) validates integral hydrodynamics performance and pebble fuel recirculation methods. (Image © Kairos Power)
The Engineering Test Unit Demonstration Experiment (ETUDE) test stand in Kairos Power’s Alameda Rapid Analysis, Prototyping and Iterative Design Laboratory (R-Lab) validates integral hydrodynamics performance and pebble fuel recirculation methods. (Image © Kairos Power)

What about shifting market factors? How does that enter into the rapid iteration process?

LMS: The data virtual twins receive from our physical systems allow us to shorten the design duration, the qualifications, the reviews, the manufacturing and operating costs, and they support the licensing. And, of course, it helps us to understand the market requirements.

"For me, virtual twins and simulations are the perfect tools for leveraging the most out of rapid iterations and rapid prototyping.”

Right now, we are seeing a huge growth in renewable energy as the movement towards a more distributed and decarbonized energy system continues gaining traction. Therefore, small modular reactors – and the KP-FHR – will fit neatly with intermittent renewable energy systems. The fact that we are utilizing a rapid iterative approach allows us to keep introducing requirements into the design. For me, virtual twins and simulations are the perfect tools for leveraging the most out of rapid iterations and rapid prototyping.

LMS: Data is an asset, and it is for any company. But for us even more so because a key part of Kairos Power’s future success is our ability to learn, retain and transfer knowledge efficiently and effectively. The only way to do that is by building a learning organization, and that’s why data is critical for us. SpaceX generates terabytes of data a day. That’s why they were able to develop their programs in such a disruptive way.

Right now, we are constructing very big pieces of hardware. What we are able to learn from that construction phase, we can introduce into the next iteration.

That’s where we are currently. Everything is linked to the real hardware Kairos Power is building and manufacturing. It is very important to build things, and those [virtual] tools help us to be more agile and accelerate quickly. The real strength of virtual twins come from data provided by a physical build.

Lou Martinez Sancho, Executive Vice President, Strategy & Innovation for Kairos Power

For more than 20 years, Lou Martinez Sancho has led teams in various industries including biotech, energy, automotive, architecture, engineering and construction (AEC). Her goal at Kairos Power: To enable the world’s transition to clean energy with nuclear-generated power that is clean, affordable and safe.

Read more about how the energy and materials industry is aiming to reduce its carbon footprint

Related resources