Engineering news
The model, developed by a team of scientists at the US Department of Energy’s (DOE) Argonne National Laboratory in Illinois, could maintain safety and efficiency in a type of nuclear reactor known as a sodium-cooled fast reactor (SFR).
An SFR uses liquid sodium to cool its core and efficiently create carbon-free electricity by splitting heavy atoms. While they are not yet used commercially in the US, many believe they could revolutionise power production and help reduce nuclear waste, the DOE said. Challenges include maintaining the purity of the high-temperature liquid sodium coolant, which is crucial for preventing corrosion and blockages in the system.
“Harnessing the power of machine learning to continuously monitor and detect anomalies advances the state of the art in instrumentation control,” said Alexander Heifetz, principal nuclear engineer at Argonne and co-author of an article on the project. “This will create a breakthrough in the efficiency and cost-effectiveness of nuclear energy systems.”
First, the team created a machine learning model to continuously monitor the cooling system at Argonne’s Mechanisms Engineering Test Loop (METL) facility. It uses 31 sensors to measure variables such as fluid temperatures, pressures and flow rates.
The METL facility is a unique experimental facility designed to safely and accurately test materials and components that could be used in SFR reactors. It also trains the engineers and technicians (and now machine learning models) who could help operate and maintain them.
“A comprehensive system enhanced with machine learning may facilitate more robust monitoring and prevent anomalies that could disrupt the functioning of an actual reactor,” the researchers said.
Next, the team demonstrated the model’s ability to detect operational anomalies “swiftly and accurately”. They put this to the test by simulating a loss-of-coolant type anomaly, marked by a sudden spike in temperature and flow rate. The model detected the anomaly within about three minutes. “This ability underscored its effectiveness as a safety mechanism,” the announcement said.
As it stands, the model flags any spike that exceeds a predetermined threshold. This method could lead to false alarms however, due to incidental spikes or sensor errors.
The team plans to refine the model to distinguish between genuine process anomalies and random measurement noise. This includes requiring the signal to remain above the threshold value for a certain period before it is considered an anomaly. They will also incorporate spatial and temporal correlations between sensors into the calculation of loss.
“Although we’re using the unique capabilities of METL to develop and test our algorithms in a liquid metal experimental research facility, there is potential to see applications in advanced reactors,” said Heifetz. “That can provide more carbon-free energy in the future.”
Alexandra Akins, a research aide at Argonne and co-author of the article, agreed. “Our research on anomaly detection using machine learning enhances the promise of nuclear energy.”
The article was published in Energies.
Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.