Application of field modelling technique to simulate interaction of sprinkler and fire-induced smoke layer

The interaction between the sprinkler water spray and the fire induced convective air flow is studied using the field modelling technique. A system of equations describing conservation of momentum, enthalpy and mass is used to simulate the physical picture. Solution of the problem is divided into two parts: gas phase and liquid phase. In the gas phase, a two-equation k — e model is used to account for the turbulent effect with the solid wall boundary described by the traditional wall functions. Numerical finite difference method is employed to solve the system of coupled non-linear partial differential equations. The equations are firstly discretized by the Power Law scheme and then solved using the Pressure Implicit with Splitting of Operators (PISO) algorithm. For the liquid phase, the sprinkler water spray is described by a collection of water droplets with different values of initial velocity components and diameter calculated from the Rossin-Rammler distribution function. The motion of each droplet is described by the Newton's Second Law with air drag and convective heat transfer from the fire induced smoke layer. This set of ordinary differential equations is solved by the fourth order Runge-Kutta method for predicting the droplet trajectories. To simplify the physical picture and bearing in mind that evaporative heat loss measured experimentally is small, coupling of the momentum and heat transfer between the smoke layer and water droplets is described by the Particle-Source-In-Cell method. In this way, two-phase flow analysis is avoided by taking the sprinkler water spray as a system of 'hard-spheres’. Neither combustion nor water suppression effect on the burning object is included. However, a 'microscopic’ view on the resultant sprinklered fire air-flow pattern, temperature and droplet properties can be visualized. Macroscopic parameters such as the drag to buoyancy ratio and the amount of convective heat transfer are predicted.