Computational Geomechanics. Manuel Pastor

Computational Geomechanics - Manuel Pastor


Скачать книгу
(3.27), and (3.28), we shall illustrate the time‐stepping scheme on the simple example of (3.6) by adding a forcing term:

      From the initial conditions, we have the known values of Φn, ModifyingAbove bold upper Phi With ampersand c period dotab semicolon Subscript n. We assume that the above equation has to be satisfied at each discrete time, i.e. tn and tn+1. We can thus write:

      and

      From the first equation, the value of the acceleration at time tn can be found and this solution is required if the initial conditions are different from zero.

      (3.37)StartLayout 1st Row ModifyingAbove bold upper Phi With two-dots Subscript n plus 1 Baseline equals ModifyingAbove bold upper Phi With two-dots Subscript n Baseline plus normal upper Delta ModifyingAbove bold upper Phi With two-dots Subscript n Baseline 2nd Row ModifyingAbove bold upper Phi With ampersand c period dotab semicolon Subscript n plus 1 Baseline equals ModifyingAbove bold upper Phi With ampersand c period dotab semicolon Subscript n Baseline plus ModifyingAbove bold upper Phi With two-dots Subscript n Baseline normal upper Delta t plus beta 1 normal upper Delta ModifyingAbove bold upper Phi With two-dots Subscript n Baseline normal upper Delta t 3rd Row bold upper Phi Subscript n plus 1 Baseline equals bold upper Phi Subscript n Baseline plus ModifyingAbove bold upper Phi With ampersand c period dotab semicolon Subscript n Baseline normal upper Delta t plus one half ModifyingAbove bold upper Phi With two-dots Subscript n Baseline normal upper Delta t squared plus one half beta 2 normal upper Delta ModifyingAbove bold upper Phi With two-dots Subscript n Baseline normal upper Delta t squared EndLayout

      Alternatively, a higher order scheme can be chosen such as GN32 and we shall have:

      (3.38)equation

      (3.39)equation

      and

      (3.40)equation

      In the above equations, the only unknown is the incremental value of the highest derivative and this can be readily solved for.

      Returning to the set of ordinary differential equations we are considering here, i.e. (3.23), (3.27), and (3.28) and writing (3.23) and (3.28) at the time station tn+1, we have:

      assuming that (3.27) is satisfied.

      Using GN22 for the displacement parameters bold u overbar and GN11 for the pore pressure parameter bold p overbar w, we write:

      (3.43a)StartLayout 1st Row ModifyingAbove Above bold u overbar With two-dots Subscript n plus 1 Baseline equals ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline plus normal upper Delta ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline 2nd Row ModifyingAbove Above bold u overbar With ampersand c period dotab semicolon Subscript n plus 1 Baseline equals ModifyingAbove Above bold u overbar With ampersand c period dotab semicolon Subscript n Baseline plus ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline normal upper Delta t plus beta 1 normal upper Delta ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline normal upper Delta t 3rd Row bold u overbar Subscript n plus 1 Baseline equals bold u overbar Subscript n Baseline plus ModifyingAbove Above bold u overbar With ampersand c period dotab semicolon Subscript n Baseline normal upper Delta t plus one half ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline normal upper Delta t squared plus one half beta 2 normal upper Delta ModifyingAbove Above bold u overbar With two-dots Subscript n Baseline normal upper Delta t squared EndLayout

      and

      (3.43b)Скачать книгу