Abstract
Five risk analysis as currently used in nuclear plants is outlined, and the relevant uncertainties are considered. The use of the computer code COMPBRN in the fire growth analysis is discussed along with response surface methodology to quantify uncertainties in the code's application. Generalized response surfaces are developed for temperature versus time for a cable tray, as well as surfaces for the temperature and depth of a hot gas layer for a room of arbitrary geometry within a typical nuclear power power plant compartment. These surfaces are then used to simulate the cable tray damage time in a compartment fire experiment. Insights and problems in the application of response surface techniques in this context are presented.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have