| Summary: | 博士 === 國立臺灣大學 === 土木工程學研究所 === 87 === When oscillatory water wave propagating over a soft poroelastic bed, a boundary layer exists within the porous bed and near the homogeneous-water/porous-bed interface. Owing to the effect of boundary layer, the conventional evaluation of the second kind of longitudinal wave inside the soft poroelastic bed by one parameter is very inaccurate so that a systematic boundary layer correction approach for soft poroelastic bed is proposed to solve the nonlinear water wave problem in the present study. After analyzing the length scale and order of magnitude of physical variables, perturbation expansion for the boundary layer correction approach based on two small parameters is proposed and then solved. The solutions which are carried out to the first three terms for suppressing secular phenomena are valid for the first kind and the third kind of waves throughout the whole domain. The second kind of wave is solved systematically inside the boundary layer while it disappears outside the boundary layer. This study successfully derives the complete governing equations and boundary conditions which especially inside the boundary layer for the second kind of wave are built to overcome the difficulty that Chen, Huang & Song(1997) met. The linear water wave solution is compared with the result of Huang & Song (1993) in order to verify the validity of the present approach. Based on the analysis of linear solution, a simplified boundary value formulation, which is expected to be very useful in numerical computation, is also proposed by decomposing the physical variables into in-phase and out-of-phase parts. Shi (1998) adopted this concept to improve the execution of numerical computation as well as save much CPU time and then got satisfactory results. Then the nonlinear water wave solution is proposed to improve the result of Chen, Huang & Song(1997) who failed to estimate the second kind of wave accurately. Moreover, when considering current effect, the solutions obtained by Runge-Kutta method can be traced automatically to different flow regime of bed forms including dune, antidune and flat bed, which improve Chiang’s (1994) results much better without the discontinuous phenomena and artificial judgement. In addition, an approximate criterion of soil liquefaction at the surface of soft poroelastic bed is proposed and the liquefaction phenomena are also observed.
|
|---|
