# For simplicity we have suppressed the time dependence

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: x u ux Operator Symbol De nition Forward Di erence j j +1 ; j Backward Di erence r r j j ; j;1 Central Di erence j j +1=2 ; j ;1=2 Average ( j+1=2 + j;1=2) 2 j Shift j j +1 Derivative ( x)j j Table 2.1.1: De nition of nite di erence operators. u u u u u E u u u u u Eu D u Du u = u u Example 2.1.1. The centered di erence formula (2.1.7a) can be expressed in terms of the central di erence and averaging operators and . Observe that uj = ( j+1=2 ; j;1=2) = 1 ( j+1 ; j;1) 2 u u Thus, uj x u = j+1 ; j;1 2 u u x : u : 6 Finite Di erence Methods Example 2.1.2. An operator appearing to a positive integral power is iterated thus, 2u j = ( j+1=2 ; j;1=2) = j+1 ; 2 j + j;1 u u u u u : Thus, the centered second di erence approximation (2.1.9a) of the second derivative can be written as 2j ( xx)j 2 u u x : Example 2.1.3. Let us expand uj +1 in a Taylor's series about xj to obtain 1 x2 (u ) + : : : uj +1 = uj + x(ux )j + xx j 2 or, using the derivative operator D de ned in Table 2.1.1, 1 x2 D2 + : : : ]u : uj +1 = 1 + xD + j 2 We may use the Taylor's series expansion...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online