Civil Engineering Reference
In-Depth Information
CALL mocouq(psi,dsbar,lode_theta,dq1,dq2,dq3)
CALL formm(stress,m1,m2,m3); flow=f*(m1*dq1+m2*dq2+m3*dq3)
erate=MATMUL(flow,stress); evp=erate*dt
evpt(:,i,iel)=evpt(:,i,iel)+evp; devp=MATMUL(dee,evp)
eload=MATMUL(devp,bee); bload=bload+eload*det*weights(i)*gc(1)
END IF
!------update the Gauss Point stresses and calculate pore pressures-------
IF(converged.OR.iters==limit)THEN
tensor(:,i,iel)=stress
etensor(:,i,iel)=etensor(:,i,iel)+eps+evpt(:,i,iel)
pore(i,iel)=(etensor(1,i,iel)+etensor(2,i,iel)+
&
etensor(4,i,iel))*bulk
END IF
END DO gauss_pts_2
!-----------------------compute the total bodyloads vector ---------------
bdylds(g)=bdylds(g)+bload; bdylds(0)=zero
END DO elements_3; IF(converged.OR.iters==limit)EXIT
END DO its; totd=totd+loads
WRITE(11,'(I5,3E12.4,I5)')iy,totd(no(1)),dsbar, pore(1,1),iters
IF(iters==limit)EXIT
END DO disp_incs
CALL dismsh(loads,nf,0.05_iwp,g_coord,g_num,13)
CALL vecmsh(loads,nf,0.05_iwp,0.1_iwp,g_coord,g_num,14)
STOP
END PROGRAM p69
New scalar reals:
bulk
apparent fluid bulk modulus (
K
e
)
cons
consolidating stress (
σ
3
)
New dynamic real arrays:
etensor
holds running total of all integrating point strain terms
pore
holds running total of all integrating point pore pressures
The example shown in Figure 6.42 represents a single axisymmetric 8-node element
subjected to vertical compressive displacement increments along its top face. The analysis
is of a “CU” triaxial test, in which the sample has been consolidated under a cell pressure of
σ
3
=
100 kN/m
2
, followed by undrained axial loading. In order to compute pore pressures
during undrained loading, it is necessary to update strains (
etensor
)aswellasstresses
after each increment. The pore pressure is also computed from equation (6.82).
This program assumes a homogeneous material described by a Mohr-Coulomb failure
criterion. In addition to the usual shear strength, dilation, and elastic parameters, the data
file must provide the apparent fluid bulk modulus
bulk
and the initial consolidation stress
cons
After the effective stress-strain matrix has been augmented by the fluid bulk modulus
according to (6.80), the global stiffness matrix is formed in the usual way. Prescribed axial
displacement increments are then applied to the top of the element using the “penalty”
method as used previously in Program 6.4.
Just before the displacement increment loop begins, the subroutine
deemat
is called to
form the effective stress-strain matrix. The data shown in Figure 6.42 is for an undrained
