This preview shows page 1. Sign up to view the full content.
Unformatted text preview: = 1 2 : : : r: (7.1.8) Again, we'll assume that a means of nding u( ) , i = 1 2 : : : r, to satisfy the initial
condition (7.1.1b) can be found. With this, we observe that the work of solving
this system is about half that of the previous procedure (7.1.2c) when the number
of terminal conditions r is approximately m=2.
i 3. Write the general solution of the linear BVP (7.1.1) as y(x) = c1u(1) (x) + c2u(2) (x) + : : : + cr u(r) (x) + v(x)
or y(x) = U(x)c + v(x) (7.1.9a) where U(x) = u(1) (x) u(2) (x) : : : u(r) (x)]
6 c = c1 c2 : : : cr ]T : (7.1.9b) 4. By construction, this representation satis es the ODE (7.1.1a) and the initial conditions (7.1.1b). It remains to satisfy the terminal conditions (7.1.1c) and this can
be done by determining c so that Ry(b) = R U(b)c + v(b)] = r
or RU(b)c = r ; Rv(b): (7.1.10) As noted, the main savings of this procedure relative to (7.1.2-7.1.3) is the reduced
number of ODEs that have to be solved. This is o set by (possible) di culties associated
View Full Document
- Spring '14