Validation of the Neeswood PBSD Procedure: Testing of a Full-Scale Six-Story Building at Japan's E-Defense

Author(s):  
J. W. van de Lindt ◽  
S. Pei ◽  
Steve Pryor ◽  
Hidemaru Shimizu ◽  
Izumi Nakamura
Keyword(s):  
Fire Safety ◽  
2020 ◽  
Vol 36 ◽  
pp. 121-130
Author(s):  
T. Shnal ◽  
S. Pozdieiev ◽  
R. Yakovchuk ◽  
O. Nekora ◽  
S. Sidnei

The aim of the work was mathematical modeling of fire development in a three-storey residential building during full-scale fire tests; research of accuracy and reliability of parameters of temperature modes of fire in separate rooms of the building. To achieve this goal, it is advisable to use computational gas-hydrodynamics, which allow to determine the limits of application of this approach to predict the behavior of building structures in a fire. The Pyrosim computer system, which serves as a user shell for the Fire Dynamics Simulator program, was used to calculate the temperature in fire room models. This FDS system uses numerical algorithms to solve the complete system of Navier-Stokes differential equations to determine temperature and other hazards in a fire. To visualize the results of calculations, the software module of the PyroSim Smokeview system was used, which allows to build appropriate graphical representations of temperature distributions.A numerical experiment was performed to model full-scale tests of rooms with fire in a three-story building using computer gas-hydrodynamics methods. The nature of the fire and the time dependences of its main parameters were revealed, which in turn allowed to analyze the adequacy of the simulation results and investigate their adequacy and accuracy.The obtained results of research on the accuracy of modeling of full-scale tests of rooms with fire in a three-story building showed that the error determined when comparing experimental and calculated data was not significant. The relative error did not exceed 28%, and the standard deviation did not exceed 51 ° C. The values of the statistical criteria of Cochren, Student and Fisher for the simulation results due to the difference between the calculated and experimental data have values of no more than 0.98, 0.84 and 1.008, respectively, and do not exceed the tabular values. This means that the simulation results are adequate, which allows you to use this approach to predict the behavior of building structures in a fire that is close to real.


2009 ◽  
Vol 74 (641) ◽  
pp. 1353-1361 ◽  
Author(s):  
Yuichi MATSUOKA ◽  
Keiichiro SUITA ◽  
Satoshi YAMADA ◽  
Yuko SHIMADA ◽  
Motoki AKAZAWA ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document