Page 493 - Fluid-Structure Interactions Slender Structure and Axial Flow (Volume 1)
P. 493

Appendix A


                First-principles Derivation of the

                    Equation of Motion of a Pipe

                                 Conveying Fluid







             Consider the system of Figure A.l(a), free to oscillate in the horizontal  (X, ZJ-plane, so
             that gravity is inoperative. Externally imposed tension and pressurization effects are not
             present and, for simplicity, dissipative effects are neglected. Elements of the fluid and the
             pipe of length 6x are shown in Figure A. 1 (c,d), with the forces and moments at the ends
             apportioned slightly differently from Figure 3.6.
               The  acceleration  of  the  fluid  element  (still  making  the  plug-flow  approximation)  is
             derived by the standard dynamics approach, following Ginsberg (1973). An inertial refer-
             ence  frame  (X, Y, 2) with  Y  into  the  plane  of  the  paper  and  unit  vectors  I,  J,  K,
             and  an  (x, y, :)  frame embedded  in  the  pipe  element  with  unit  axes i, j, k  are  utilized
             [Figure A. 1 (b)], together with the expression

                             af = a0 + I, x r + 20 x v,,1  + o x  (o x r) + arel.   (A. 1)
             which  may  be  found in  any  book  on dynamics  [e.g. Meriam  (1980)l; o is the angular
             velocity of the pipe (and of the (x, y, 2) frame) with respect to the inertial frame, and the
             subscript  ‘rel’ denotes quantities relative to the  (x, y, z} frame. The various components
             of  (A. 1 ) may be expressed and then approximated according to the assumptions made in
             Section 3.3.1 as follows:
                  a’u    azw     a’  U’                                           aw
             a0  = -I+   -K    2:  -K,     v,~ = Ui = U cos @I+ U  sin @Kz UI+ U -K,
                  at’    at2     at’                                              as
                                                                                  (A.3)
                   a@
                                                             dU
                                               dU
                                                      U2
                  _-  Jz __               arel = -i  + - 2 -I+      dU  aw  K   , a2w  K,
                                                         k
                    as      :t  ($) J,          dt    vn     dt     - - + U- ~   8s’
                                                                    dt  as
             assuming a positive (counterclockwise) rotation, a@/&  it is also noted that r, the distance
             from the origin of {x, y, z) to other points within the element is of second order smallness,
             so that the second and fourth terms of  (A. 1) are negligible. Hence, (A. 1) may be written as


             correct to O(E) - which is the same as equation (3.28).


                                                                                    463
   488   489   490   491   492   493   494   495   496   497   498