PSI - Issue 73

Juraj Králik et al. / Procedia Structural Integrity 73 (2025) 73–80 Juraj Králik/ Structural Integrity Procedia 00 (2025) 000–000

78 6

All roofs within the NPP have a slope of 5%, which represents a slope of 3°, which falls under flat roofs within the definition according to STN EN 1991-1-4 (2007). The roof area is divided into zones F (corner areas), G (zone between F zones), H (located behind zones F and G), I (represents the leeward zone within the flat roof). The individual lengths and widths of these zones are determined based on the parameter e , which depends on the geometry of the roof and is the smaller value of b or 2 h , see Fig. 5. The procedure for determining individual zones within the NPP was based on the principle that each NPP building was compared within the windward edge of a specific building and the height of the roof. It was similarly applied to the NPP walls. Recommended values of external pressure coefficients for individual zones by the Eurocode are presented in Tab. 1 in the case of the simple shape of the roof. All simulations were run with the same solver settings. It was a “Pressure-Based” model and a time-varying “transient” problem. The SST k- ω method was used based on recommendations from several authors (Kawoluk, 2023, Menter, 2020, Wilcox, 2008). The boundary conditions were defined as follows:  INLET At the beginning of the domain, the “INLET” area is defined. This area is defined as the area corresponding to the input wind parameters. The wind input itself was in the form of a logarithmic function according to STN EN 1991-1 4 (2007). This wind profile entered the solver using a UDF (User Defined Function). Furthermore, the quantities k (turbulent kinetic energy) and ω (specific dissipation rate) derived from the wind profile were defined, which were also part of the UDF.  OUTFLOW At the end of the domain, the wind exits the given space. This boundary condition did not require any settings, so it was a free movement of the wind.  SYMMETRY This boundary condition was defined on the sides of the domain and on its surface. This condition did not require any additional settings and represented the same boundary condition as at the entrance and exit.  WALL This boundary condition was defined at the lower level of the domain representing the terrain type, as well as the type of NPP surfaces. This area was set as "Stationary no slip wall", i.e. without roughness. The other solver settings under “Solution Methods” were as follows: • Pressure-Velocity Coupling Scheme – Simple • Spatial discretization Gradient – Least Squares Cell Based • Spatial discretization Pressure – Second Order • Spatial discretization Momentum – Second Order Upwind • Spatial discretization Turbulent Kinetic energy – Second Order Upwind • Spatial discretization Specific Dissipation Rate – Second Order Upwind • Transient Formulation – Second Order Implicit Monitors During the simulation, the residuals were monitored, namely continuity, X -velocity, Y -velocity, Z -velocity, parameters k and ω . The convergence of the solution was defined by setting the maximum value of the residual. The calculation was initiated using “Hybrid Initialization”. The calculation itself was selected as “transient” time varying analysis and the “Time Stepping Method” was defined as “Fixed”. The simulation started during the first 1.500-time steps with a time step of 0.03s. This represented the first 45 seconds of steady flow with the monitors set to 10 -4 . Then, data collection was started for a “transient” analysis, which ran for 4.000 steps with a time step of 0.03s. This meant 2 minutes of steady flow with the monitors set to 10 -5 . The wind logarithmic profile, which entered the simulation in the form of a UDF, was defined as follows:

v

⋅

κ

z z  + 

( )

ref

,

(3)

ln ⋅ 

v z

0

=

 

0 ref z z  +  

z

0

ln ⋅ 

κ

 

z



0

where the reference velocity is v ref = 35.64 ms ref = 10 m and terrain roughness z o = 0.05 m and Von Karman constant κ = 0.4. The dissipation parameter ω is defined in the UDF model as the fluid (or air) friction velocity, which was defined in the form: -1 for height z