OCR Text |
Show TABLE 1 Summary of Transport Equations Solved Equation Continuity x-Momentum y-Momentum Thermal Ener~v Turbulent kinetic energy Energy Dissipation 1 u v T k e: lleff lleff lleffl (j T ~ o : + ~ (ueff :)+ ~ (ueff :) ap a (ll ff av ) a (ll ff a v _+_ e _+_ e _ ) ay ax ax a y ay o PG Pe: Pe: Pe:2 C1 - G - C2 k k and llt [(au av) 2 G -- - +- P ay ax + au 2 2 (-) + ax Constants In The Turbulent Mod,l C II 0.09 model uses finite volume methods to discretize the transport equations. A typical computational control volume, with its associated coordinate axes, is sketched in Fig. 1. Capital letters designate centers of adjacent control volumes, and lower case letters indicate the midpoints of each control surface. A grid generation scheme is used to divide the solution domain into a finite number of these volumes with the values of scalar quantities (e.g., pressure and temperature) associated with the center of each volume. The component velocities, however, are defined at the midpoint of the control surfaces. In other words, control volumes for component velocities are displaced in space relative to the control volume for the scalar quantities. SOLUTION PROCEDURE - To linearize the differential equations, a hybrid central/upwind differencing scheme is applied to the convection and diffusion terms. The finite difference equations are solved by integrating each variable over the control volume in which the variable is defined. Fluid properties outside the associated control volumes are determined by linear interpolation. Pressure and velocity are coupled 1.0 110 [ ADJACENT CONTROL VOLUI1E CENTER -~. 1.0 z I ~T ·oL -- Fig. 1 - Computational control volume showing coordinate directions (X,Y,Z), point a of definition for acalers (P), velocity components (n,s,e,w,t,b), and adjacent control volume centera (N,S,E,W,T,B). |