CFD of spray-steam condensation, spray-wall interaction and heat transfer

Abstract

This thesis investigates the possibilities for simulating spray cooling in pressurizer systems in pressurized water reactors (PWRs) using OpenFOAM. Spray cooling in pressurizers is part of the mechanism for regulating the pressure. The multiphase flow associated with the process is highly complex and difficult to study experimentally due to harsh operating conditions. An improved understanding of these processes is important for future investigations of thermal stress in the tank.

The work focuses on evaluating Lagrangian particle tracking (LPT) submodels for
atomization, breakup and droplet collision. Both simplified spray cases for investigating the submodels and validation cases based on experimental data provided by Vattenfall were evaluated. In addition, the hybrid VoF-LPT solver atomizationFOAM was tested.

The result showed a significant difference between the investigated LPT submodels. Spray width was shown to be strongly dependent on operating pressure due to the pressure dependence of breakup models. Validation against experimental data showed significant limitations of pure LPT approaches for dense sprays. Most model combinations were unable to reproduce experimentally observed droplet coalescence behavior. The Reitz KHRT breakup model
showed the best potential in reproducing coalescence behavior seen in experiments. Limitations remained in accurately reproducing liquid structures near the nozzle.

The hybrid VoF-LPT solver atomizationFOAM showed promising result by producing
spray shapes that agreed well with experimental observations, particularly regarding spray width and the representation of large scale fluid structures. However, the investigated setup was highly simplified and further validation under more representative conditions is required.

Popular Abstract

Have you ever been inside the pressurizer tank in a nuclear power plant? Probably not, since the pressure inside the tank is the same as 15 family cars on top of you and it is 345°C hot!

To meet the energy demand for the electrification required to battle global warming, all fossil free energy sources need to be employed. In Sweden, there are two operating pressurized water reactors at Ringhals. One of the core components is the pressurizer tank, which is responsible for maintaining the correct pressure inside the nuclear reactor. The pressurizer is half filled with water and the rest is filled with steam. To lower the pressure, spray cooling is used, where cold water is injected into the tank through a spray nozzle. As some of the spray reaches the tank wall, a thermal load is introduced into the wall, limiting the safe operational lifetime of the tank.

Much is unknown about the spray, how it broadens as it travels through the tank, how much it is heated as it travels through the hot steam and how it is broken up into individual drops. Therefore a lifetime was previously calculated, assuming worst case scenarios for the tank wall and spray. This is problematic as the actual lifetime is likely much longer than the calculated one, forcing the power plant to close prematurely.

To more correctly predict the lifetime, much more needs to be known about how the spray behaves. Therefore, this thesis evaluates different kinds of spray modeling approaches in computer simulations.

The results show that there is no clear cut approach to model the spray inside the pressurizer. Simpler approaches are just that, too simple, missing too much of the physics! However, the more complex approaches come at a high cost in computer power, requiring weeks or even months on super computers.

To achieve the end goal of increasing the estimated lifetime of the power plant, two main conclusions were drawn. First, new simple models that include more of the physics occurring are necessary. Second, mixed approaches of using both simple and complex modeling needs to be further studied and developed.

• author: Tufvesson, Hanna and Nimmermark, Simon
• supervisor: Christer Fureby; Eric Lillberg; Nicolas Edh; Kristian Angele
• organization: Department of Energy Sciences
• course: MVKM01 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Computational fluid dynamics, OpenFOAM, multiphase flow, Volume of Fluid, Lagrangian particle tracking, spray, atomization, breakup
• report number: LUTMDN/TMHP-26/5698-SE
• ISSN: 0282-1990
• language: English
• id: 9240626
• date added to LUP: 2026-06-24 09:24:38
• date last changed: 2026-06-24 09:24:38