FireDOC Search

Author
Mitler, H. E.
Title
Comparison of Several Compartment Fire Models: An Interim Report.
Coporate
National Bureau of Standards, Gaithersburg, MD
Sponsor
Nuclear Regulatory Commission, Washington, DC
Report
NBSIR 85-3233, October 1985, 34 p.
Distribution
Available from National Technical Information Service
Keywords
comparison | compartment fires | fire models | predictive models | mathematical models
Abstract
A substantial number of mathematical models for compartment fires have been developed in the past decade. This report analyzes and compares in depth three such models. This is done with particular emphasis on the needs of the Nuclear Regulatory Commission and Sandia National Lab, for their Risk Methods Integration and Evaluation Program. The models examined are (1) The Harvard family of models, Mark 5,5.2, 5.3, and 6; (2) COMPBRN; and (3) FAST. The capabilities of a model can be understood in terms of the answers to a series of questions; these are formulated, placed into categories, and then answered. The answers are given in brief form in a table (section 4); the bulk of the report consists of elaborations and discussions of these points. A number of conclusions emerge from this study: first, there is at present no one model which will do everything which is wanted by an arbitrary user - some have certain capabilities; others, different capabilities. Therefore each user must look for a model which does what he (she) wants, and does it well. Second, from the point of view of the NRC, Harvard Mark 5.3 may be the best choice, because it is the only one which has all of the following characteristics: it is time-dependent; it considers forced ventilation; it has good numerics; and it takes into account heating and ignition of target objects. The last section of this report briefly outlines which developments would be desirable, for each model.