Scientists in the United States have come up with a physics-based algorithm that could one day make it possible for nuclear microreactors to adjust the power output based on need.<\/p>\n
The study was funded by the US Department of Energy\u00a0Office of Nuclear Energy and was led by the University of Michigan.<\/p>\n
Nuclear microreactors would be able to generate up to 20 megawatts of thermal energy for heat or electricity. Their small size could make them an efficient energy source in remote locations, military installations, or even ships.<\/p>\n
Further, they can even be integrated into electrical grids, where they would be able to provide a stable energy flow. However, there would be a need for them to adjust the power output to match the grid\u2019s shifting demand.<\/p>\n
While in large reactors, there are human operatives who make the adjustments, it could get difficult or infeasible to deploy personnel in remote locations for a nuclear microreactor.<\/p>\n
Autonomous output adjustment for nuclear microreactors<\/p>\n
The University of Michigan-led team focused the study on\u00a0High-Temperature Gas-Cooled Reactors\u00a0(HTGR), which can be scaled from microreactor to large modules. The study was based on the Holos-Quad (Gen 2+) model, a HTGR-type microreactor design.<\/p>\n
The research team leveraged model predictive control (MPC), a method that predicts future behavior to optimize control over a defined period of time under certain constraints, according to a press release by the university.<\/p>\n
They developed an<\/a> MPC controller that optimized the rotation of control drums that surround the microreactor\u2019s central core, that decrease power when facing inwards and increase power when facing outwards.<\/p>\n The team also integrated PROTEUS \u2013 a simulation toolset for high-fidelity reactor physics analysis \u2013 to ensure the model was accurately representing the nuclear microreactor\u2019s operation.<\/p>\n Testing the algorithm<\/p>\n According to the release, when asked to ramp the power up or down at 20 percent per minute, the control algorithm stayed within 0.234 percent of the target.<\/p>\n Moreover, it was able to do this without the help of artificial intelligence (AI), \u201cmeaning everything about the automated control for load follow operation is grounded in physics and mathematics and readily explainable\u2014an essential feature for passing regulatory review.\u201d<\/p>\n \u201cMany startup and legacy companies in the U.S. are pushing towards near-term and broad deployment of nuclear microreactors, and our work establishes a clear avenue to achieve that in an economically viable way,\u201d said\u00a0Brendan Kochunas, an associate professor of nuclear engineering and radiological sciences at U-M and corresponding author of the study.\u00a0<\/p>\n He went on to add that the method can help vendors design reactors with autonomous control systems that are safer and more secure.<\/p>\n There are several nuclear microreactor projects being pursued by different companies in the US alone. The US Air Force had recently awarded<\/a> a Direct to Phase II (D2P2) innovation research contract to Nano Nuclear Energy to explore the feasibility of deploying its Kronos micro modular nuclear reactor at Joint Base Anacostia-Bolling (JBAB) in Washington, D.C.<\/p>\n