Page 286 - Computational Fluid Dynamics for Engineers
P. 286

276                                                      9.  Grid  Generation




                                                                     ^ - J / L ^ i
                                              \  P.P.  /
                                                 V
                                                 /  N
                                               /    X               /      !  \
                                             i'      N  v
                                                                 :           ^—
         Fig.  9.11.  Algebraic  interpolation;  (a)  Individual  projector  Pi;  (b)  Individual  projector
         Pj\  (c)  Product  projector  PiPj\  (d)  Boolean  projector  Pi  0  Pj.



         The  structure  of operators  given  above  allows multidirectional  interpolations  to
         be  constructed  systematically  from  unidirectional  interpolations.  This  interpo-
         lation  that  matches  the  function  on  the  entire  boundary  is  called  "Transfinite
         Interpolation".
            In  two  dimensions,  the  Transfinite  Interpolation  can  be  implemented  in  the
         following  sequence:
         1.  Interpolation  of  r  in  the  i-direction:  F\  =  Pi?
         2.  Evaluation  of  the  discrepancy  between  r  and  this  result  on  the  j-lines  that
            will  be  used  in  the  j-interpolation:  (? —  F\)
         3.  Interpolation  of the  discrepancy  in the  j-direction:  F2  — Pj?—  F\)
         4.  Addition  of  the  results  of  this  ^'-interpolation  to  the  results  of  the  i-
            interpolation:  ?(i,j)  =  F\  +  F2
         Pi  and  Pj  can  be  any  one  of  the  unidirectional  operators  studied  earlier,  i.e.
         Lagrange  interpolations,  Hermite  interpolations,  splines,  or  non-polynomial  in-
         terpolations  such  as  the  hyperbolic  tangent  function.
            The  methodology  described  above  can  be  used  to  generate  grids  from  any
         four  arbitrary  bounding  curves  [4]. The  program  can  handle  any  grid  topology
         (C,  O  or  H,  see  Section  9.6)  since  the  boundaries  can  be  any  arbitrary  curve.
         Figure  9.12  shows  different  grids  produced  around  an  ellipse  using  a  C-grid
         topology.  Figure  9.12a  shows  a  grid  generated  using  unidirectional  interpola-
         tion  with  linear  Lagrange  polynomials.  The  input  to  the  program  is  a  sequence
         of  inner,  outer,  left  and  right  boundaries.  The  inner  (j  =  1)  boundary  is  the
         contour  of the  ellipse  plus the  branch  cut  of the  C-grid.  The  outer  {j  =  JMAX)
         boundary  is the  far  field  contour  around  the  grid  excluding  the  right  hand  ver-
         tical  boundary.  The  left  (i  =  1)  and  right  (i  — IMAX)  boundaries  are the  lower
         and  upper  halves  of the  downstream  vertical  boundary.  Figure  9.12b  shows  the
         grid  obtained  by  unidirectional  interpolation  using  Hermite  polynomials.  Her-
         mite  interpolations  allow  the  slopes  of  the  grid  lines  at  the  boundaries  to  be
         specified.  In  this  case,  they  are  set  to  correspond  to  near-orthogonality  at  the
         boundaries. Using Hyperbolic  Tangent  spacing to concentrate the grid  lines  near
         the  inner  boundary  results  in  grids  shown  in  Fig.  9.12c  for  Lagrange  interpola-
         tion  and  in  Fig.  9.12d  for  Hermite  interpolation.
   281   282   283   284   285   286   287   288   289   290   291