Page 146 - Applied Numerical Methods Using MATLAB
P. 146

INTERPOLATION BY CUBIC SPLINE  135
            to write

               S k,1 − S k−1,1 = h k−1 (S k,2 − S k−1,2 ) + 2h k−1 S k−1,2 = h k−1 (S k,2 + S k−1,2 )
                                                                         (3.5.3)
            In order to use this for eliminating S k,1 from Eq. (3.5.2), we subtract (3.5.2b)
            from (3.5.2a) to write

             h k (S k+1,2 + 2S k,2 ) − h k−1 (S k,2 + 2S k−1,2 ) + 3(S k,1 − S k−1,1 ) = 3(dy k − dy k−1 )


            and then substitute Eq. (3.5.3) into this to write

                  h k (S k+1,2 + 2S k,2 ) − h k−1 (S k,2 + 2S k−1,2 ) + 3h k−1 (S k,2 + S k−1,2 )
                     = 3(dy k − dy k−1 )

                  h k−1 S k−1,2 + 2(h k−1 + h k )S k,2 + h k S k+1,2 = 3(dy k − dy k−1 )  (3.5.4)
                             for k = 1: N − 1

            Since these are N − 1 equations with respect to N + 1 unknowns {S k,2 ,k = 0:
            N}, we need two more equations from the boundary conditions to be given as
            listed in Table 3.4.
              How do we convert the boundary condition into equations? In the case where
            the first-order derivatives on the two boundary points are given as (i) in Table 3.4,
            we write Eq. (3.5.2a) for k = 0as

                h 0 (S 1,2 + 2S 0,2 ) + 3S 0,1 = 3 dy 0 ,  2h 0 S 0,2 + h 0 S 1,2 = 3(dy 0 − S 0,1 )
                                                                        (3.5.5a)
            We also write Eq. (3.5.2b) for k = N as


                           h N−1 (S N,2 + 2S N−1,2 ) + 3S N−1,1 = 3 dy N−1

            and substitute (3.5.3)(k = N) into this to write


                 h N−1 (S N,2 + 2S N−1,2 ) + 3S N,1 − 3h N−1 (S N,2 + S N−1,2 ) = 3 dy N−1
                          h N−1 S N−1,2 + 2h N−1 S N,2 = 3(S N,1 − dy N−1 )  (3.5.5b)

            Equations (3.5.5a) and (3.5.5b) are two additional equations that we need to solve
            Eq. (3.5.4) and that’s it. In the case where the second-order derivatives on the
            two boundary points are given as (ii) in Table 3.4, S 0,2 and S N,2 are directly
            known from the boundary conditions as



                             S 0,2 = s (x 0 )/2,  S N,2 = s (x N )/2     (3.5.6)
                                    0                 N
   141   142   143   144   145   146   147   148   149   150   151