This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: MATH 220: FIRST ORDER SCALAR SEMILINEAR EQUATIONS First order scalar semilinear equations have the form (1) a ( x, y ) u x + b ( x, y ) u y = c ( x, y, u ); here we assume that a, b, c are at least C 1 , given real valued functions. Let V be the vector field on R 2 given by V ( x, y ) = ( a ( x, y ) , b ( x, y )) , so a ( x, y ) u x + b ( x, y ) u y is the directional derivative of u along V . Let γ = γ ( s ) be an integral curve of V , i.e. γ ( s ) = ( x ( s ) , y ( s )) has tangent vector V = V ( x ( s ) , y ( s )) for each s . Explicitly, this says that (2) x ′ ( s ) = a ( x ( s ) , y ( s )) , y ′ ( s ) = b ( x ( s ) , y ( s )) . Now let v ( s ) = u ( γ ( s )) = u ( x ( s ) , y ( s )). Thus, by the chain rule, v ′ ( s ) = x ′ ( s ) u x ( x ( s ) , y ( s )) + y ′ ( s ) u y ( x ( s ) , y ( s )) = a ( x ( s ) , y ( s )) u x ( x ( s ) , y ( s )) + b ( x ( s ) , y ( s )) u y ( x ( s ) , y ( s )) = c ( x ( s ) , y ( s ) , u ( x ( s ) , y ( s ))) , where in the last step we used the PDE. Thus, v ′ ( s ) = c ( x ( s ) , y ( s ) , v ( s )) , i.e. v satisfies an ODE along each integral curve of V . To solve the PDE, we parameterize the integral curves by an additional param eter r , i.e. the integral curves are γ r = γ r ( s ) = ( x r ( s ) , y r ( s )), where r is in an interval (or the whole real line), and each γ r is an integral curve for V , i.e. (3) x ′ r ( s ) = a ( x r ( s ) , y r ( s )) , y ′ r ( s ) = b ( x r ( s ) , y r ( s )) , so v r ( s ) = u ( γ r ( s )) solves (4) v ′ r ( s ) = c ( x r ( s ) , y r ( s ) , v r ( s )) . Note that here the subscript r denotes a parameter, not a derivative! We may equally well write x r ( s ) = x ( r, s ), y r ( s ) = y ( r, s ), and we will do so; we adopted the subscript notation to emphasize that along each integral curve r is fixed, i.e. is a constant. Which parameterization should we use? We are normally also given initial con ditions along a curve Γ = Γ( r ) with Γ( r ) = (Γ 1 ( r ) , Γ 2 ( r )), namely u (Γ( r )) = φ ( r ) , where φ is a given function. For example, we are given an initial condition on the x axis: u ( x, 0) = φ ( x ), in which case we may choose r = x , Γ( r ) = ( r, 0). Then we want the integral curve with parameter r to go through Γ( r ) at ‘time’ 0, i.e. we want γ r (0) = Γ( r ) , and we want v r (0) = u ( γ r (0)) = φ ( r ) . 1 2 Combining these, we have two groups of ODEs: a system for the integral curves of V , also called (projected) characteristic curves , with initial conditions given by Γ, and a scalar ODE along the integral curves with initial condition given by φ : x ′ r ( s ) = a ( x r ( s ) , y r ( s )) , x r (0) = Γ 1 ( r ) , y ′ r ( s ) = b ( x r ( s ) , y r ( s )) , y r (0) = Γ 2 ( r ) , (5) and (6) v ′ r ( s ) = c ( x r ( s ) , y r ( s ) , v r ( s )) , v r (0) = φ ( r ) ....
View
Full
Document
This note was uploaded on 03/16/2010 for the course CME 303 taught by Professor Vasy during the Fall '10 term at Stanford.
 Fall '10
 Vasy

Click to edit the document details