Page 206 - Computational Fluid Dynamics for Engineers
P. 206

194                           6.  Inviscid  Flow  Equations  for  Incompressible  Flows


         6.5  A  Panel  Program    for  Airfoils

         In this section  a computer  program  is described  for  calculating the  inviscid  flow-
         field  over an  airfoil  with the Hess-Smith  panel method that  was discussed  in  the
         previous  section  (see  also  Appendix  B).  Before  reviewing  the  four  subroutines
         and  MAIN  of this  program,  it  is  useful  to  examine  the  solution  of Eqs.  (6.4.13)
         and  (6.4.14),  which  can  be  written  in  the  form  given  by  Eq.  (4.5.23), that  is,

                                                                           (4.5.23)

         Here  A  is  a  square  matrix  of  order  (N  +  1)
                        an     au   •  ..  aij  . ..  a 1N  Gl,JV+l
                        a<2i   ^22  • ..  a 2j  . ..  a 2N  «2,7V+1


                A  =    an     a>%2  •   a^        any     ttz,JV+l         (6.5.1)


                       ayvi   ayv2  ...  ajsfj  ...  a^jy  ajv,N+i
                      a
                                       a
                                                 a
                     | N+l,l  OJV+1,2  • • • N+lj  • • • N+l,N  Q>N+1,N+1
                              T                                    T
         and  x  =  (qi,...,qi,-..,qN,T)  and  b=  (6i,..., &*,...,  b N,  b N+i)  with  T  de-
         noting  the  transpose.  The  elements  of the  coefficient  matrix  A  follow  from  Eq.
         (6.4.14)
                                            z =  l,2,...,7V
                                   —  A n
                                n- a %3  —  J\ %3,                        (6.5.2a)
                                            j  =  l,2,...,7V
                                       N
                                        B
                             «i,N+i  =  Y, &   <  =  l,2,...,iV            (6.5.2b)
                                      J = I
         A%  are  given  by  Eq.  (6.4.9)  and  B%  by  Eq.  (6.4.11). The  relation  in  Eq.  (6.5.1)
         follows  from  the  definition  of  x  where  r  is essentially  £AT+I-
            To  find  ajsr+ij  (J  =  1,...,  N)  and  a7v+i,7V+i  in the  coefficient  matrix  A,  we
         use  the  Kutta  condition  and  apply  Eq.  (6.4.13)  to  Eq.  (6.4.8b)  and,  with  r  as
         a  constant,  we write  the  resulting  expression  as
                      N            N
                                 T
                                            v
                                     B
                                                 (
                     Y^ AjQj +  Yl h     + °° cos a  -  °i)
                     3 = 1        3=1
                              N             N
                                                     V
                                         T
                             I Y  At NjQj  + Y1  B Nj  + oo cos(a  -  0 N)
                             13=1          3=1
         or  as
                           N                    N
                          £ ( 4 ,  + 4,)^- + r ](^ + B* )
                                                J
                                                              Nj
                                                                            (6.5.3)
                          3=1                  3=1
                             =  -Foo cos(a  -Oi)-Voo  cos(a  -  6>^)
   201   202   203   204   205   206   207   208   209   210   211