Machine Learning-Driven Uncertainty Quantification and Parameter Analysis in Fire Risk Assessment for Nuclear Power Plants
| dc.contributor.author | Sahin, Elvan | en |
| dc.contributor.committeechair | Pacheco Duarte, Juliana | en |
| dc.contributor.committeemember | Liu, Yang | en |
| dc.contributor.committeemember | Lattimer, Brian Y. | en |
| dc.contributor.committeemember | Wu, Zeyun | en |
| dc.contributor.department | Mechanical Engineering | en |
| dc.date.accessioned | 2025-01-28T09:00:20Z | en |
| dc.date.available | 2025-01-28T09:00:20Z | en |
| dc.date.issued | 2025-01-27 | en |
| dc.description.abstract | Fire poses a critical risk to the safe operation of nuclear power plants (NPPs), with electrical cabinet and liquid spill fires being among the most frequent and challenging scenarios to address. Traditional fire risk assessment models often lack precision due to complex physics and inherent uncertainties, especially in predicting the heat release rate (HRR) — a key parameter for assessing fire severity. This dissertation presents an innovative framework that integrates machine learning (ML) models, particularly neural networks and tree-based algorithms, with uncertainty quantification (UQ) techniques to enhance fire modeling and risk assessment in NPPs. The framework is applied to electrical enclosure cabinets and spill fires that represent about 50% of challenging fire scenarios in NPPs. By leveraging extensive experimental datasets, this study develops ML models that capture the influence of critical fire parameters on HRR, enabling more accurate predictions of fire behavior. Key features are evaluated to establish their influence on peak HRR. Advanced UQ tools, including Monte Carlo sampling and sensitivity analysis are applied to quantify uncertainties and identify parameters with the greatest impact on model output variability. The resulting ML-driven insights allow for a refined understanding of fire dynamics, guiding experimental planning and uncertainty reduction efforts. For electrical enclosure fires, the models highlight the importance of cable surface area, heat release rate per unit area of the cable, ignition source heat release rate, ventilation area, and cabinet volume in determining peak HRR. Sensitivity analysis revealed that HRRPUA is the most significant parameter. For spill fires, the models underscore the significance of substrate thermal conductivity and slope, ignition delay time, and fuel properties, showing that fuel amount and properties are key in fixed quantity spills, while fuel discharge rate and properties are most influential in continuous spills. | en |
| dc.description.abstractgeneral | Fires in nuclear power plants (NPPs) present serious risks, especially in areas with complex equipment, such as electrical cabinets or areas with potential fuel spills. Predicting how fires in these settings will behave is difficult, as many variables can impact how quickly and intensely a fire grows. Current models used to assess fire risk often struggle to capture these complexities, leading to conservative estimates that may not fully reflect real-world conditions. This dissertation introduces a new approach that uses machine learning (ML) to analyze experimental fire data relevant to NPP scenarios, identifying key factors that influence fire behavior. The models help predict a fire's peak heat release rate (HRR), a measure of fire intensity, by learning from past data and capturing important patterns that traditional models can miss. Additionally, advanced uncertainty analysis methods are used to assess which factors contribute the most to the variability in fire outcomes, helping researchers to focus on the most impactful areas for fire prevention and control. By applying this ML-based framework, the study aims to improve the accuracy of fire risk assessments, supporting safer NPP designs and more effective response strategies. The findings offer a pathway to more precise and reliable fire modeling, ultimately helping protect people, infrastructure, and the environment from fire-related hazards in nuclear facilities. | en |
| dc.description.degree | Doctor of Philosophy | en |
| dc.format.medium | ETD | en |
| dc.identifier.other | vt_gsexam:42490 | en |
| dc.identifier.uri | https://hdl.handle.net/10919/124406 | en |
| dc.language.iso | en | en |
| dc.publisher | Virginia Tech | en |
| dc.rights | In Copyright | en |
| dc.rights.uri | http://rightsstatements.org/vocab/InC/1.0/ | en |
| dc.subject | Fire | en |
| dc.subject | Machine Learning | en |
| dc.subject | Uncertainty Quantification | en |
| dc.subject | Sensitivity Analysis | en |
| dc.subject | Probabilistic Risk Assessment | en |
| dc.subject | Monte Carlo | en |
| dc.title | Machine Learning-Driven Uncertainty Quantification and Parameter Analysis in Fire Risk Assessment for Nuclear Power Plants | en |
| dc.type | Dissertation | en |
| thesis.degree.discipline | Nuclear Engineering | en |
| thesis.degree.grantor | Virginia Polytechnic Institute and State University | en |
| thesis.degree.level | doctoral | en |
| thesis.degree.name | Doctor of Philosophy | en |