A Study on the Reliability of Modeling Compartment Fires With Respect to Thermocouple Response and Sprinkler Activation

Abstract

Real and numerical fire experiments involve temperature measurements with thermocouples, and thus some considerations on numerical modeling of this process are presented and a new approach to thermocouple modeling is introduced. Using ANSYS Fluent software a well-recognized analytical thermocouple model was implemented in each cell of the computational domain, which allows for determination of thermocouple responses as a continuous field. Similarly, sprinklers are key elements of fire-protection systems. Sprinklers activation is one of the breakthrough moments during the course of a compartment fire. Therefore, assumptions on sprinkler activation time are of crucial importance when designing a fire safety system. Just as for thermocouple modeling, virtual sprinklers based on a commonly admitted response time index (RTI) model were placed in all cells. The proposed approach provides data on sprinklers activation or thermocouple response for the whole domain instead of retrieving data point by point only for predefined locations. In this study, experimental data available in the literature were used for the validation of the proposed approach. In addition, the results were compared with those obtained with the commonly used Fire Dynamic Simulator (FDS) software. The outcomes might be of a significant importance for practitioners, who deal with fire experiments and fire protection. Furthermore, some issues on accurate modeling of fire gases flow are discussed extensively. It was found that commonly applied k-? and k-? turbulence models might fail in the case of modeling of fire plumes in confined spaces.