Page 350 - DMTH504_DIFFERENTIAL_AND_INTEGRAL_EQUATION
P. 350

Unit 21: Classifications of Second Order Partial Differential Equations




          It will be seen that the first three terms of equation (1) allow us to classify the equation into one  Notes
          of three distinct types: Elliptic, for example Laplace s equation, Parabolic, for example the diffusion
          equation or Hyperbolic, for example the wave equation as follows:
                   2    2
                   V    V
                            0                        (Laplace equations for two variables x, y)
                   x   2  y   2
                    2 V  V
                  K  2                                                (Diffusion equation)
                    x    t 
                   2       2
                   V   1   V
                    2   2  2                                             (Wave equation)
                   x   C  t 
          Each of these types of equation has distinctive properties. We would like to know about those
          properties of equation (1) that are unchanged by any change of co-ordinates since these must be
          of fundamental significance and not just a result of our choice of co-ordinate system. We can
          write this change of co-ordinates as
                 (x, y)   { (x, y),  (x, y)}
          with

                     
                   
                   ( , )
                         0                                                         ...(2)
                     y
                   x
                   ( , )
          If equation represents a model physical system, a change of co-ordinates should not affect its
          qualitative behaviour. Writing  (x, y)    ( ,  ) and using subscripts to denote partial derivatives,
                                        
          we find that
                                                    2
                                    2
                          ,       2               
                  x   x    x    xx  x   x x  x  x   xx    xx 
          and similarly for the other derivatives. Substituting these into equation (1) gives us
                                       
                                      
                                               
                                                 
                                                                 
                                                          
                  Α    2B    +C   b 1 ( , )   b 2 ( , )    +  ( , )  g ( , )  ...(3)
                                                                  
                                                      b
                                                         
                                                       3
          where
                  A  a 2 x  2b   c 2 y ,
                            x x
                  B  a   ( b       ) c 
                      x x
                                         y y
                                  y x
                             x y
                                                                                  ....(4)
                  C  a 2  2b   c 2  ,
                       x    x y   y
          We do not need to consider other co-efficient functions b , ( ,  ), b ( ,  ), b ( ,  ).
                                                        1      2     3
          We can express (4) in a concise matrix form as
                   A B     x   x  a b   x   y
                                                                                   ...(5)
                   B C     y   y  b c   x    y
          which shows that
                                            2
                      A B        a b   ( , )
                                          
                                        
                  det        det                                                   ...(6)
                                          y
                      B C        b c   ( , )
                                        x
                    
                 ( , )
                  
          In (6)        = Jacobian of transformation.
                 ( , )
                  x
                    y
                                      2
          This shows that the sign of a c   b  is independent of the choice of co-ordinate system which
          allows us to classify the equation.
                                  2
          An Elliptic equation has ac < b , for example Laplace equation
                                           LOVELY PROFESSIONAL UNIVERSITY                                   343
   345   346   347   348   349   350   351   352   353   354   355