Researchers from the High Flux Isotope Reactor (HFIR) at the Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL) have developed an advanced predictive software tool that forecasts the date and time of day when each HFIR reactor operating cycle will exhaust its uranium fuel supply – with an accuracy of plus or minus 3.6 hours.
Each year, HFIR attracts about 500 scientists from all over the world to conduct research on their samples of materials. Other users include NASA, national security agencies and the many ORNL industry partners that the lab supplies with isotopes.
These commitments must be factored into the six or seven HFIR operating cycles per year that are each about 24 days in length – with only a limited number of positions available around the reactor core for experiments. That is why time management is absolutely vital to allow reliably scheduling experiments right up to the end of each cycle.
“The advanced predictive tool supports DOE’s mission of providing user facilities for research and producing rare isotopes for its contracts with government and industry in support of national security, medical treatments, space missions and many other applications,” said Joel McDuffee, the team lead for HFIR’s Reactor Analysis and Modeling Group. “Due to a range of complex factors, cycle length had previously varied by a day or more in either direction, which limited how much of the operating cycle we could reasonably schedule for experiments. Now the more accurate predictions of when each HFIR reactor operating cycle will end can benefit hundreds of neutron scattering and isotope experiments each year, because this permits precise planning for the experiments down to the final hours of neutron production.”
Not only do HFIR researchers need to know cycle lengths to better plan their experiments, but also HFIR operators, engineers, technicians, craft people and managers can now better balance the experiment loadings (which typically reduce cycle length) against the schedules of invited neutron scientists.
The new tool combines two predictive frameworks: a statistical model based on historical reactor operation, and a simulation method for fuel depletion calculations. Researchers found strong agreement existed between the two methods, which demonstrated that cycle length variability can be reliably reduced to a manageable set of parameters.
The scientists found the biggest factor affecting cycle length was the number and types of experiments loaded into the beryllium ring surrounding the reactor core. Other factors included reactor operating power level, uranium fuel loading, and the age of the reactor’s moveable control plates.
The statistical method can be used for quick evaluations of potential experiment loadings, while the simulation methodology can be used for novel experiments without a previous cycle history and for confirming the cycle length estimate with a final experiment inventory.
Scientists can now more accurately optimize the mix of experiments to maximize each cycle length to increase the degree and effectiveness of the materials irradiation process.
From facilitating DOE and ORNL contracts with government and industry, to supporting academia, the military, and space missions, the HFIR cycle length predictive tool will aid scientists and HFIR staff for decades to come.
HFIR is a DOE Office of Science user facility.
UT-Battelle manages ORNL for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science.
