Page 605 - Handbook Of Integral Equations
P. 605

11.17. The Bubnov–Galerkin Method

                 11.17-1. Description of the Method
               Let
                                                     b

                                    ε[y(x)] ≡ y(x) – λ  K(x, t)y(t) dt – f(x) = 0.          (1)
                                                    a
               Similarly to the above reasoning, we seek an approximate solution of Eq. (1) in the form of a finite
               sum
                                                  n

                                    Y n (x)= f(x)+  A i ϕ i (x),  i =1, ... , n,            (2)
                                                 i=1
               where the ϕ i (x)(i =1, ... , n) are some given linearly independent functions (coordinate functions)
               and A 1 , ... , A n are indeterminate coefficients. On substituting the expression (2) into the left-hand
               side of Eq. (1), we obtain the residual
                                    n               b                   b

                          ε[Y n (x)] =  A j ϕ j (x) – λ  K(x, t)ϕ j (t) dt – λ  K(x, t)f(t) dt.  (3)
                                                   a                  a
                                   j=1
                   According to the Bubnov–Galerkin method, the coefficients A i (i =1, ... , n) are defined from
               the condition that the residual is orthogonal to all coordinate functions ϕ 1 (x), ... , ϕ n (x). This gives
               the system of equations
                                        b
                                        ε[Y n (x)]ϕ i (x) dx =0,  i =1, ... , n,
                                      a
               or, by virtue of (3),
                                      n

                                        (α ij – λβ ij )A j = λγ i ,  i =1, ... , n,         (4)
                                      j=1
               where
                      b                    b  b                         b  b

               α ij =  ϕ i (x)ϕ j (x) dx, β ij =  K(x, t)ϕ i (x)ϕ j (t) dt dx, γ i =  K(x, t)ϕ i (x)f(t) dt dx.
                     a                    a  a                          a  a
               If the determinant of system (4)
                                              D(λ) = det[α ij – λβ ij ]
               is nonzero, then this system uniquely determines the coefficients A 1 , ... , A n . In this case, formula (2)
               gives an approximate solution of the integral equation (1).


                 11.17-2. Characteristic Values
                                                                         ˜
                                                                  ˜
               The equation D(λ) = 0 gives approximate characteristic values λ 1 , ... , λ n of the integral equation.
               On finding nonzero solutions of the homogeneous linear system
                                      n
                                                    (k)
                                              ˜
                                        (α ij – λ k β ij )A ˜ j  =0,  i =1, ... , n,
                                      j=1
                                                                                         ˜
               we can construct approximate eigenfunctions Y ˜ (k) (x) corresponding to characteristic values λ k :
                                                     n
                                                      n
                                                          (k)
                                                         ˜
                                             Y ˜ (k) (x)=  A ϕ(x).
                                                          i
                                              n
                                                      i=1
                   It can be shown that the Bubnov–Galerkin method is equivalent to the replacement of the
               kernel K(x, t) by some degenerate kernel K (n) (x, t). Therefore, for the approximate solution Y n (x)
               we have an error estimate similar to that presented in Subsection 11.13-2.

                 © 1998 by CRC Press LLC








               © 1998 by CRC Press LLC
                                                                                                             Page 588
   600   601   602   603   604   605   606   607   608   609   610