Page 12 - Essentials of applied mathematics for scientists and engineers
P. 12

book   Mobk070    March 22, 2007  11:7








                     2  ESSENTIALS OF APPLIED MATHEMATICS FOR SCIENTISTS AND ENGINEERS
                            A partial differential equation expresses a dependent variable, say u, as a function of more
                       than one independent variable, say x, y,and z. Partial derivatives are normally written as ∂u/∂x.
                       This is the first-order derivative of the dependent variable u with respect to the independent
                       variable x. Sometimes we will use the notation u x or when the derivative is an ordinary derivative
                                                                   2
                                                                         2
                       we use u . Higher order derivatives are written as ∂ u/∂x or u xx . The order of the differential

                       equation now depends on the orders of the derivatives of the dependent variables in terms of
                       each of the independent variables. For example, it may be of order m for the x variable and of
                       order n for the y variable. A boundary value problem consists of a partial differential equation
                       defined on a domain in the space of the independent variables, for example the x, y, z space,
                       along with conditions on the boundary. Once again, if the partial differential equation and the
                       boundary conditions contain only terms of first degree in u and its derivatives the problem is
                       linear. Otherwise it is nonlinear.
                            A differential equation or a boundary condition is homogeneous if it contains only terms
                       involving the dependent variable.

                       Examples
                       Consider the ordinary differential equation


                                               a(x)u + b(x)u = c (x),  0 < x < A.                    (1.1)
                       Two boundary conditions are required because the order of the equation is 2. Suppose

                                                  u(0) = 0    and     u(A) = 1.                      (1.2)

                       The problem is linear. If c (x) is not zero the differential equation is nonhomogeneous. The first
                       boundary condition is homogeneous, but the second boundary condition is nonhomogeneous.
                            Next consider the ordinary differential equation


                                                a(u)u + b(x)u = c     0 < x < A                      (1.3)
                       Again two boundary conditions are required. Regardless of the forms of the boundary condi-
                       tions, the problem is nonlinear because the first term in the differential equations is not of first
                       degree in u and u since the leading coefficient is a function of u. It is homogeneous only if

                       c = 0.
                            Now consider the following three partial differential equations:

                                                        u x + u xx + u xy = 1                        (1.4)
                                                        u xx + u yy + u zz = 0                       (1.5)

                                                        uu x + u yy = 1                              (1.6)
   7   8   9   10   11   12   13   14   15   16   17