Page 101 - Fundamentals of The Finite Element Method for Heat and Fluid Flow
P. 101

93
                        THE FINITE ELEMENT METHOD
                        (polynomials used) within an element remain unchanged under a linear transformation
                        from one Cartesian coordinate system to another. Polynomials that exhibit this invariance
                        property are said to possess ‘Geometric Isotropy’. Clearly, we cannot expect a realistic
                        approximation if our field variable representation changes with respect to a movement in
                        origin, or in the orientation of the coordinate system. Hence, the need to ensure geometric
                        isotropy in our polynomial interpolation functions is apparent. Fortunately, we have two
                        simple guidelines that allow us to construct polynomial series with geometric isotropy.
                        These are as follows:
                          (i) Polynomials of order ‘n’ that are complete, that is, those that contain all terms have
                             geometric isotropy. The triangle family satisfies this condition whether it be a linear,
                             quadratic or cubic form.
                         (ii) Polynomials of order ‘n’ that are incomplete yet contain the appropriate terms to
                             preserve ‘symmetry’ have geometric isotropy. We neglect only these terms that occur
                                                    3
                                                                 2
                                                            2
                                                       3
                             in symmetric pairs that is, (x ,y ), (x y, xy ), and so on.
                        Example: For an eight-node element, the following polynomial, P , satisfies geometric
                        isotropy, that is,
                                                                   2
                                        P (x, y) = α 1 + α 2 x + α 3 y + α 4 x + α 5 xy + α 6 y 2  (3.269)
                        with either
                                                          3
                                                       α 7 x + α 8 y 3                     (3.270)
                        or
                                                         2
                                                                 2
                                                      α 7 x y + α 8 y x                    (3.271)
                        added to it.

                        Example 3.5.1 Before concluding this chapter, it is important to consider a numerical prob-
                        lem for illustrating the theory presented. For this purpose, we consider again a fin problem
                        as shown in Figure 3.29. The linear variation for the temperature within each finite element
                        is assumed. We shall derive the element equations from the most general formulation given


                                             100°C    h = 120 W/m °C; T a = 25°C
                                                              2


                                                                  k = 200 W/m°C
                                                        2mm

                                                       3mm
                                                        x         Insulated
                                                  2 cm


                                        Figure 3.29  Heat transfer from a rectangular fin
   96   97   98   99   100   101   102   103   104   105   106