Page 35 - IJOCTA-15-2
P. 35

M. M. Parsa, K. Sayevand, H. Jafari, I. Masti / IJOCTA, Vol.15, No.2, pp.225-244 (2025)
                                                                  We obtain
                                                               "                                           #
             dg(x i )  g(x i+1 ) − g(x i−1 )                 1  β U + U n−1  n−1           U + U l−1
                                                                                              l
                                                                    n
                                               2
                                                                             X
                                                                                                          U
                    =                   + O(h ),                b  i   i  −     b β  − b β   i   i  − b β n−1 i 0
             dx               2h                             λ  0    2           n−l−1  n−l    2
            
            
                                                                             l=1
                                                                                      n−1    n−1   n−1  !
                                                                            n
                                                                A   U  n  − 2U − U  n  U  − 2U  − U
             2                                               =      i+1   i    i−1  +  i+1  i     i−1
            
             d g(x i )  g(x i+1 ) − 2g(x i ) − g(x i−1 )        2        h 2               h 2
                      =                           + O(h ).
                                                        2
            
                dx 2                h 2                             U  n  − U  n  U n−1  − U n−1  !
                                                       (28)   +  G   i+1  i−1  +  i+1  i−1
                                                                 2     2h          2h
                                                                    n   n−1    n   n−1
                                                                   U + U        f + f
                                                              − H    i   i    +  i   i  .
                By inserting the i-th space surface in Eq. (25),       2           2
            we have                                                                                      (33)
                                                                  Consider the following definitions:
                                                                                     n    n−1
                            n−1                                           n−  1  U + U
              1  β           X     β       β                              U   2  :=  i    i   ,
                b u(x i , t n ) −  b    − b     u(x i , t l )               i            2
                                                                         
              λ  0                 n−l−1   n−l                                 1     n    n−1            (34)
                             l=1                                          n−       f + f i
                                                                                     i
                                                                          f   2  :=         .
                                  2                                        i
                                  ∂ u(x i , t n )  ∂u(x i , t n )                       2
                 β
              − b n−1 u(x i , t 0 ) = A      + G
                                     ∂x 2           ∂x            Therefore
              − Hu(x i , t n ) + f(x i , t n ),
                                                       (29)     "                                           #
                                                              1   β  n−  1  n−1   β     β     l−  1  β   0
                                                                           X
                                                                                                         U
                                                                 b U i  2  −    b     − b     U i  2  − b n−1 i
                                                                  0
                                                              λ                  n−l−1   n−l
                where U i n  = u(x i , t n ) and f i n  = f(x i , t n ) for    n−  1  l=1  n−  1  n−  1  
            i = 0, 1, 2, · · · , M x , n = 0, 1, 2, · · · , N t .    U i+1 2  − 2U i  2  − U i−1 2
                                                               = A                        
                                                                               h 2
                Based on the definitions and by placing Eq.
            (28) in Eq. (29), we have                                n− 1    n− 1  
                                                                     U   2  − U  2        n−  1   n−  1
                                                              + G    i+1     i−1    − HU   2  + f  2  .
                                                                          2h              i      i
              "                                        #
             1   β  n   n−1   β      β     l   β    0                                                  (35)
                        X
                b U −       b      − b     U − b     U
             λ   0  i        n−l−1    n−l    i   n−1 i
                        l=1                                       The BICs will be discretized as follows:
                    n       n     n          n      n
                   U i+1  − 2U − U i−1        U i+1  − U i−1
                             i
             = A                        + G
                           h 2                     2h          
                                                                   n
                                                               U = λ n := λ(t n ),
                                                               
                        n
                   n
             − HU + f .                                         0
                        i
                   i
                                                       (30)                          n = 0, 1, 2, · · · , N t , (36)
                                                               
                                                                  n
                                                                 U    = ζ n := ζ(t n ),
                                                               
                Again, we derive the Crank-Nicolson scheme.        M x
            We have
                                                                   0
               "                                       #          U = υ i := υ(x i ), i = 0, 1, 2, · · · , M x .  (37)
             1   β  n   n−1   β      β    l    β    0            i
                        X
                b U −       b      − b     U − b    U
                    i
                                            i
             λ   0           n−l−1    n−l        n−1 i
                        l=1                                       To better understand the method, its compu-
                    n       n     n         n      n
                   U    − 2U − U              U    − U        tational molecule is presented in Figure 1
             = A     i+1     i    i−1   + G    i+1    i−1
                           h 2                    2h
                        n
                   n
             − HU + f ,
                        i
                   i
                                                       (31)
                and also
              1    β  n−1  n−1   β     β    l−1
                          X
                b U     −      b      − b    U
              λ  0  i           n−l−1    n−l   i
                          l=1
                              n−1      n−1     n−1
                               U    − 2U     − U
                 β    0         i+1      i       i−1   (32)
                    U
              − b n−1 i  = A
                                        h 2
                     n−1    n−1
                    U    − U
              + G    i+1     i−1   − HU i n−1  + f i n−1 .    Figure 1. The computational molecule of the pro-
                         2h                                   posed method
                                                           230
   30   31   32   33   34   35   36   37   38   39   40