This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: F1.4ZH2 Numerical Solution of PDEs Page 7 1.4 Taylor series and difference operators We can use Taylor series expansions in one of the variables to see how well Finite Difference (FD) approximations work. Consider the approximation ( u ( x + Δ x, t ) u ( x, t )) / Δ x to ∂u/∂x . F x u ( x j , t n ) = u ( x j + Δ x, t n ) u ( x j , t n ) = u + Δ xu x + 1 2 Δ x 2 u xx + 1 3! Δ x 3 u xxx + . . . ( x j ,t n ) u ( x j , t n ) = Δ xu x + 1 2 Δ x 2 u xx + . . . ( x j ,t n ) ≈ Δ x u x ( x j , t n ) when Δ x is sufficiently small So F x Δ x u ( x j , t n ) = u x ( x j , t n ) + O (Δ x ) where O (Δ x ) means a quantity which → 0 as Δ x → 0. Similarly D x u = u ( x j + Δ x, t n ) u ( x j Δ x, t n ) = u + Δ x u x + 1 2 Δ x 2 u xx + 1 3! Δ x 3 u xxx + 1 4! Δ x 4 u xxxx + O (Δ x 5 ) ( x j ,t n ) u Δ x u x + 1 2 Δ x 2 u xx 1 3! Δ x 3 u xxx + 1 4! Δ x 4 u xxxx + O (Δ x 5 ) ( x j ,t n ) ⇒ D x u 2Δ x = u x + Δ x 2 6 u xxx + O (Δ x 4 ) ( x j ,t n ) So D x 2Δ x u ( x j , t n ) = u x ( x j , t n ) + O (Δ x 2 ) . This is better because Δ x 2 → 0 faster than Δ x as Δ x → 0. In a similar way we can show that δ 2 x u ( x j , t n ) = Δ x 2 u xx ( x j , t n ) + O (Δ x 4 ) so δ 2 x Δ x 2 u ( x j , t n ) = u xx ( x j , t n ) + O (Δ x 2 ) when Δ x is small . By expanding first in one variable and then in the other(s), we can analyse approximations involving terms like u ( x j + Δ x, t n + Δ t ). Exercise: Show that F x F t Δ x Δ t u ( x j , t n ) = u xt ( x j , t n ) + O (Δ x, Δ t ) . and D x D t 4Δ x Δ t u ( x j , t n ) = u xt ( x j , t n ) + O (Δ x 2 , Δ t 2 ) . so that again Central differences are better. We now have enough background to start looking at PDEs, starting with parabolic problems. F1.4ZH2 Numerical Solution of PDEs Page 8 2 Parabolic PDE’s 2.1 Introduction The most practical approach is to study a concrete example. Hence we start by looking at the heat or diffusion equation ∂U ∂T = K ∂ 2 U ∂X 2 , (2.1) where U = U ( X, T ) is measured in some units (say ◦ C) and the physical time and space variables T > 0, X ∈ (0 , L ) are also measured in appropriate units. K is a constant which depends on the material, called the Thermal Conductivity . We specify Dirichlet boundary conditions (BCs) at x = 0 , L , i.e. U (0 , T ) and U ( L, T ) given for all T > 0, and initial conditions (ICs) U ( X, 0) given for X ∈ (0 , L ). X L U(X) given T U T =U XX U(0,T) given U(L,T) given Need to find the solution in the open box T > , X ∈ (0 , L ) The heat equation models many things, for example • the temperature U in a long thin insulated bar with fixed temperatures at each end X L U(0,T) given U(L,T) given • The diffusion of a chemical in a solid or stationary fluid (in this case U is the concentration of the chemical)....
View
Full
Document
This note was uploaded on 11/14/2011 for the course MATH 480 taught by Professor Sd during the Spring '11 term at Middle East Technical University.
 Spring '11
 sd
 Approximation, Taylor Series

Click to edit the document details