Heat Transfer and Safety Analysis Laboratory
Working for what's next in nuclear reactors

Risk Analysis

Identifying Needed Fire Input Data to Reduce Modeling Uncertainty (2020-2024)

Project image 2Probabilistic risk assessments (PRA) of fires in nuclear power plants commonly use models to predict the conditions and expected damage from different plausible fire scenarios in order to quantify risk. In these analyses, fire model input varies from general behavior (overall heat release rate, combustion product yields, heat of combustion, physical size, etc.) to more detailed properties (thermal properties, pyrolysis kinetics, ignition temperature, etc.). Much of the data used to generate input for models are based on historical experiments that were conducted prior to many of the advanced measurement techniques in fire. The proposed research effort aims to identify the fire parameters that have the largest impact on fire conditions, quantify those parameters contributing to uncertainties in the fire data through Monte Carlo simulation results, and use statistical analysis and machine learning models from simulation results to assess existing data and recommend appropriate new fire tests to reduce uncertainties that are important to risk. Based on this research, we plan to develop a framework capable of determining the significant contributors to uncertainty in physical events that are relevant to the risk assessment and, then, determine the need for new experiments that would be of the most value to reduce risk.

PI: Prof. Juliana P. Duarte

Collaborators: Prof. Brian Lattimer (Virginia Tech), Dr. Jun Wang and Prof. Michael Corradini (UW-Madison), and Dr. Kelly Senecal (Convergent Science)

Funding source: U.S. DOE – NEUP

Journal publications

  1. Sahin, E., Lattimer, B., Allaf, M.A., Duarte, J.P., “Uncertainty quantification of unconfined spill fire data by coupling Monte Carlo and artificial neural networks“, Journal of Nuclear Science and Technology, 2024.
  2. Sahin, E., Lattimer, B., and Duarte, J.P., “Assessing spill fire characteristics through machine learning analysis”, Annals of Nuclear Energy, 192, 109961, 2023.
  3. Salvi, U., Lattimer, B.Y., Alhadhrami, S., Wang, J., Sahin, E., Duarte, J.P., “Analysis of Historic Fire to Determine Most Frequent Challenging Events”, Progress in Nuclear Energy, v.146, 104146, 2022.

Peer-reviewed conference publications

  1. Sahin, E., Henkes, P., Lattimer, B., and Duarte, J.P., “Uncertainty Quantification and Sensitivity Analysis of a Machine Learning-Based Spill Fire Model for Nuclear Power Plants”, PSAM 2023 Topical Conference AI & Risk Analysis for Probabilistic Safety/Security Assessment & Management, Virtual Meeting, October 23–25, 2023.
  2. Sahin, E., Lattimer, B., and Duarte, J.P., “Using Machine Learning to Assess Spill Fire Data for Use in Fire PRA”, The 19th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH-20), Washington D.C., August 20–25, 2023.
  3. Islam, M., Lattimer, B., and Duarte, J.P., “Factors Affecting the Behavior of a Fixed Quantity Fuel Spill”, ASME’s International Mechanical Engineering Congress & Exposition (IMECE2023), New Orleans, LA, USA, Oct. 29 – Nov. 2, 2023.
  4. Sahin, E., Lattimer, B., and Duarte, J.P., “Using Machine Learning to Assess Spill Fire Data for Use in Fire PRA”, The 19th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH-19), Washington D.C., August 20–25, 2023.
  5. Sahin, E., Salvi, U., Wang, J., Lattimer, B.Y., Duarte, J.P., “Assessment of Fire Experimental Data for Use in Nuclear Power Plants Fire PRA”, ANS Annual Meeting, Anaheim, CA, June 12-16, 2022.
  6. Salvi, U., Lattimer, B.Y., Sahin, E., Duarte, J.P., “Statistical Analysis to Determine Significant Parameters that Affect the Heat Release Rate of Electrical Cabinets”, Advances in Thermal-Hydraulics (ATH 2022), Anaheim, CA, June 12-16, 2022.

Thesis and dissertations

Urvin Salvi, Computational Study of Parameters Affecting Electric Cabinet Fire Heat Release Rate, M.S. thesis. Committee members: Prof. Brian Lattimer (chair), Prof. Juliana P. Duarte, and Dr. Jun Wang, Mechanical Engineering Department, Virginia Tech. Defense date: 05/10/2022.

Investigation of Spent Nuclear Fuel (SNF) Dry Cask System for Long-Term Storage (2020 – 2021)

PI: Prof. Juliana P. Duarte

Collaborators: Prof. Rebecca Cai (MSE – Virginia Tech) and Prof. Sonja Schmid (STS – Virginia Tech)

Funding source: VT IIHCC DA

 

Accelerated Experiments to Investigate Chloride-Induced Stress Corrosion Cracking (2020 – 2021)

PI: Prof. Juliana P. Duarte

Collaborator: Prof. Kathy Lu (MSE)

Funding source: Institute for Critical Technology and Applied Science (ICTAS), Virginia Tech