Page 87 - Computational Fluid Dynamics for Engineers
P. 87

72                                                 2.  Conservation  Equations



         familiar  and  less  perplexing.  In  a  real  external-flow  problem,  u e(x)  is  obtained
         from  the  solution  of the  inviscid  flow  equations.  It  the  external  flow  is  affected
         by the displacement  effect  of the shear  layer,  iteration  between the  external-flow
         and  shear-layer  calculations  is  necessary,  as  discussed  briefly  in  Chapter  7  and
         in  more  detail  in  [12]. The  alternative  is  to  solve  the  Navier-Stokes  equations
         throughout  the  flowfield.
            Internal  flows  (Fig.  2.5d)  consist  of shear  layer  or  layers  filling  part  or  all  at
         the  space  between  two  solid boundaries.  In  this  case the  pressure  distribution  is
         set  predominantly  by the  displacement  effect  of the  shear  layer.  It  is  convenient
         to distinguish  flows  in which the  shear  layers  fill the  cross section  and  flows  such
         as  that  near  the  entrance  to  a  duct  (the  left-hand  part  of  Fig.  2.5d)  where  a
         region  of  effectively  inviscid  flow  obeying  Bernoulli's  equation  remains.
            It  is  also  convenient  to  distinguish  the  "entrance  region",  in  which  the  ve-
         locities  change  with  x,  and  the  "fully  developed"  region  far  downstream  in  a
         constant  area  duct,  in  which  the  velocities  do  not,  but  of  course  the  pressure
         continues to decrease with  x.  As will be shown  in Chapter  7, the relevant  bound-
         ary  conditions  in each  case  are  similar  to  those  for  external  flows,  except  that  it
         is  also  necessary  to  satisfy  the  requirement  of  constant  mass  flow  between  the
        solid  surfaces.
            The  boundary  conditions  for  hyperbolic  equations  are  rather  involved  in
        comparison  to  those  for  elliptic  or  parabolic  equations.  For  this  reason,  their
        discussion  is  postponed  to  Chapters  4  and  10.  For  a  detailed  discussion,  the
        reader  is  referred  to  Hirsch  [11].



         References


         [1]  Anderson,  DA.,  Tannehill,  J.C.  and  Pletcher,  R.H.:  Computational  Fluid  Mechanics
            and  Heat  Transfer,  Hemisphere  Publishing  Co.,  1984.
         [2]  Arpaci,  V.S.  and  Larsen,  P.S.:  Convection  Heat  Transfer,  Prentice  Hall,  New  Jersey,
            1984.
         [3]  Cebeci, T.:  Convective  Heat  Transfer,  Horizons Pub., Long Beach,  Calif, and  Springer,
            Heidelberg,  2002.
         [4]  Anderson,  J.:  Aerodynamics,  McGraw  Hill,  N.Y.,  1988.
         [5]  Anderson,  J.:  Introduction  to  Computational  Fluid  Dynamics.  McGraw  Hill,  N.Y.,
            1995.
         [6]  Pulliam,  T.H.:  Efficient  Solution  Methods  for  the  Navier-Stokes  Equations,  Lecture
            notes  for  the  Von  Karman  Institute  for  Fluid  Dynamics  Lecture  Series:  Numerical
            Techniques  for Viscous Flow Computation  in Turbomachinery  Bladings, Brussels,  Bel-
            gium,  Jan  20-24,  1986.
         [7]  Cebeci,  T.:  Analysis  of  Turbulent  Flows.  Elsevier,  London,  2004.
         [8]  Blottner,  F.G.:  Significance  of the  Thin-Layer  Navier-Stokes  Approximation.  In  Nu-
            merical  and  Physical  Aspects  of  Aerodynamic  Flows  III,  p.  184  (ed.  T.  Cebeci),
            Springer,  N.Y.,  1996.
         [9]  Schlichting,  H.:  Boundary-Layer  Theory.  McGraw-Hill,  N.Y.,  1968.
   82   83   84   85   86   87   88   89   90   91   92