Nuclear fusion game-changer: New method can cut reactor design time by decade
Researchers at Brigham Young University have demonstrated a way to significantly reduce nuclear reactor development time, cutting the process by a decade or more. Currently, designing and licensing processes for modern nuclear reactors take around 20 years and cost approximately $1 billion. Additionally, construction of a modern nuclear reactor takes an extra five years and billions more after the licensing process. This lengthy timeline adds up to roughly 25 years to develop a nuclear reactor in the US.
The research team, led by Professor Matthew Memmott, aims to reduce the duration and cost involved in bringing new nuclear plants online with the help of artificial intelligence (AI). They have demonstrated how AI can be used to accelerate design and licensing processes for modern nuclear reactors.
“To shorten it, make it safer, cheaper, and faster to get the nuclear power, rather than take 20 years to get the license,” Professor Memmott stated.
The demand for energy and sustainability has grown rapidly, driving scientists worldwide to work on renewable energy sources. Nuclear fusion is considered a key to generating limitless, clean energy on Earth. Therefore, government and private organizations are trying to build nuclear reactors that can successfully execute the fusion process.
“Our demand for electricity is going to skyrocket in years to come, and we need to figure out how to produce additional power quickly,” Professor Memmott said. “The only baseload power we can make in the Gigawatt quantities needed that is completely emissions-free is nuclear power.”
Nuclear reactor design involves complex processes on multiple scales, from neutrons on the quantum scale to macro-scale coolant flow and heat transfer. The research team replaced a portion of traditional thermal hydraulic and neutronics simulations with a trained machine learning model.
The new method predicts temperature profiles based on variable geometric reactor parameters, which are then optimized to create an optimal nuclear reactor design at a fraction of the computational cost of traditional methods.
Testing and validation showed that the researchers’ refined model can geometrically optimize design elements significantly faster than traditional methods. For instance, they tested their idea by using their AI algorithm to replicate a shield designed by a local nuclear company. The algorithm produced a near-perfect match in just two days.
“This is amazing because within a couple of days we were able to do the same work that it took a team of engineers six months to do,” expressed Professor Memmott.
This breakthrough could pave the way for faster and more cost-effective expansion of nuclear power.
Source: https://interestingengineering.com/innovation/nuclear-fusion-method-cuts-reactor-design-time