Menu
▾
▴
Re: [Libmesh-users] training problem with an RB transient beam model
|
From: Kyunghoon L. <aer...@gm...> - 2012-12-04 01:45:52
|
Thanks for the comment. I got no errors when I run the model without any
outputs. Then I tried only one output at each time (I have two outputs:
average tip displacements in x & y), and the model gave the same error for
each case. My outputs are as follows (they are the same as those in ex5):
struct F0 : ForceAssembly
{
F0(RBConstruction& rb_sys_in, unsigned int boundary_id_in):
ForceAssembly(rb_sys_in, boundary_id_in) {}
// Apply a traction
virtual void boundary_assembly(FEMContext &c)
{
if(rb_sys.get_mesh().boundary_info->boundary_id(c.elem, c.side) ==
boundary_id)
{
const unsigned int u_var = 0;
const std::vector<Real> &JxW_side =
c.side_fe_var[u_var]->get_JxW();
const std::vector<std::vector<Real> >& phi_side =
c.side_fe_var[u_var]->get_phi();
// The number of local degrees of freedom in each variable
const unsigned int n_u_dofs = c.dof_indices_var[u_var].size();
// Now we will build the affine operator
unsigned int n_qpoints = c.side_qrule->n_points();
DenseSubVector<Number>& Fu = *c.elem_subresiduals[u_var];
for (unsigned int qp=0; qp != n_qpoints; qp++)
for (unsigned int i=0; i != n_u_dofs; i++)
{
Fu(i) += JxW_side[qp] * ( 1. * phi_side[i][qp] );
}
}
}
};
struct F1 : ForceAssembly
{
F1(RBConstruction& rb_sys_in, unsigned int boundary_id_in):
ForceAssembly(rb_sys_in, boundary_id_in) {}
// Apply a traction
virtual void boundary_assembly(FEMContext &c)
{
if(rb_sys.get_mesh().boundary_info->boundary_id(c.elem, c.side) ==
boundary_id)
{
const unsigned int u_var = 0;
const unsigned int v_var = 1;
const std::vector<Real> &JxW_side =
c.side_fe_var[u_var]->get_JxW();
const std::vector<std::vector<Real> >& phi_side =
c.side_fe_var[u_var]->get_phi();
// The number of local degrees of freedom in each variable
const unsigned int n_v_dofs = c.dof_indices_var[v_var].size();
// Now we will build the affine operator
unsigned int n_qpoints = c.side_qrule->n_points();
DenseSubVector<Number>& Fv = *c.elem_subresiduals[v_var];
for (unsigned int qp=0; qp != n_qpoints; qp++)
for (unsigned int i=0; i != n_v_dofs; i++)
{
Fv(i) += JxW_side[qp] * ( 1. * phi_side[i][qp] );
}
}
}
};
On Tue, Dec 4, 2012 at 9:31 AM, David Knezevic
<dkn...@se...>wrote:
> I doubt the solver is the problem here. There's no reason a direct solver
> should have trouble with the Poisson solves for the "output dual norms". If
> you remove the outputs for now, does everything work OK? If so, I'd have
> another look at how you defined your outputs. It is strange that you got a
> zero norm in your results below.
>
>
>
>
>
> On 12/03/2012 08:22 PM, Kyunghoon Lee wrote:
>
> I do have attached L2 as follows:
>
>
> set_L2_assembly(transient_elasticity_assembly_expansion.M0_assembly);
>
> and M0_assembly is from the below
>
> struct M0: ElemAssembly {
> // L2 matrix
> virtual void interior_assembly(FEMContext &c) {
>
> const unsigned int u_var = 0;
>
> const std::vector<Real> &JxW =
> c.element_fe_var[u_var]->get_JxW();
>
> const std::vector<std::vector<Real> >& phi =
> c.element_fe_var[u_var]->get_phi();
>
> // The number of local degrees of freedom in each variable
> const unsigned int n_u_dofs = c.dof_indices_var[u_var].size();
>
> // Now we will build the affine operator
> unsigned int n_qpoints = c.element_qrule->n_points();
>
> for (unsigned int qp=0; qp<n_qpoints; qp++) {
> for (unsigned int i=0; i<n_u_dofs; i++) {
> for (unsigned int j=0; j<n_u_dofs; j++) {
> c.elem_jacobian(i,j) += JxW[qp] *
> phi[j][qp]*phi[i][qp];
> }
> }
> } // end of qp loop
>
> } // end of interior_assembly
> };
>
> K. Lee.
>
> On Tue, Dec 4, 2012 at 9:00 AM, David Knezevic <dkn...@se...
> > wrote:
>
>> My guess is that you haven't attached an "L2" matrix to the system?
>>
>> If that's the case, refer to reduced_basis_ex3 about how to do that.
>>
>> David
>>
>>
>>
>> On 12/03/2012 07:58 PM, Kyunghoon Lee wrote:
>> > Hi David,
>> >
>> > After running
>> >
>> > gdb ./beam_2d-dbg
>> >
>> > and
>> >
>> > run -ksp_type preonly -pc_type lu -pc_factor_mat_solver_package mumps
>> >
>> > in gdb. I got the following message:
>> >
>> > ===================
>> > .....
>> > Compute output dual inner products
>> > [0]PETSC ERROR: --------------------- Error Message
>> > ------------------------------------
>> > [0]PETSC ERROR: Error in external library!
>> > [0]PETSC ERROR: Error reported by MUMPS in numerical factorization
>> phase:
>> > INFO(1)=-9, INFO(2)=1327859
>> > !
>> > [0]PETSC ERROR:
>> > ------------------------------------------------------------------------
>> > [0]PETSC ERROR: Petsc Release Version 3.3.0, Patch 4, Fri Oct 26
>> 10:46:51
>> > CDT 2012
>> > [0]PETSC ERROR: See docs/changes/index.html for recent updates.
>> > [0]PETSC ERROR: See docs/faq.html for hints about trouble shooting.
>> > [0]PETSC ERROR: See docs/index.html for manual pages.
>> > [0]PETSC ERROR:
>> > ------------------------------------------------------------------------
>> > [0]PETSC ERROR:
>> >
>> /home/aeronova/Development/RB_models/beam/beam_2d_iso_plane_strain_basic1_transient/offline_model/beam_2d-dbg
>> > on a arch-linu named KLee-Linux by aeronova Tue Dec 4 08:49:23 2012
>> > [0]PETSC ERROR: Libraries linked from
>> > /home/aeronova/Development/local/lib/petsc/x86_64_linux/3.3_p4_real/lib
>> > [0]PETSC ERROR: Configure run at Fri Nov 23 07:53:58 2012
>> > [0]PETSC ERROR: Configure options
>> >
>> --prefix=/home/aeronova/Development/local/lib/petsc/x86_64_linux/3.3_p4_real
>> >
>> --download-mpich=/home/aeronova//Development/local/share/petsc/mpich2-1.4.1p1.tar.gz
>> > --download-superlu=1 --download-blacs=1 --download-metis=1
>> > --download-parmetis=1 --download-scalapack=1 --download-mumps=1
>> > --download-umfpack=1 --with-clanguage=C++ --with-shared-libraries=1
>> > --with-debugging=1
>> > [0]PETSC ERROR:
>> > ------------------------------------------------------------------------
>> > [0]PETSC ERROR: MatFactorNumeric_MUMPS() line 670 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/mat/impls/aij/mpi/mumps/mumps.c
>> > [0]PETSC ERROR: MatLUFactorNumeric() line 2803 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/mat/interface/matrix.c
>> > [0]PETSC ERROR: PCSetUp_LU() line 160 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/ksp/pc/impls/factor/lu/lu.c
>> > [0]PETSC ERROR: PCSetUp() line 832 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/ksp/pc/interface/precon.c
>> > [0]PETSC ERROR: KSPSetUp() line 278 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/ksp/ksp/interface/itfunc.c
>> > [0]PETSC ERROR: KSPSolve() line 402 in
>> >
>> /home/aeronova/Development/local/share/petsc/3.3-p4/src/ksp/ksp/interface/itfunc.c
>> > [0]PETSC ERROR: solve() line 678 in
>> > "unknowndirectory/"src/solvers/petsc_linear_solver.C
>> > application called MPI_Abort(comm=0x84000000, 76) - process 0
>> > [unset]: aborting job:
>> > application called MPI_Abort(comm=0x84000000, 76) - process 0
>> > [Inferior 1 (process 20258) exited with code 0114]
>> > (gdb) bt
>> > No stack.
>> >
>> > ===================
>> >
>> > K. Lee.
>> >
>> > On Mon, Dec 3, 2012 at 11:16 PM, David Knezevic
>> > <dkn...@se...>wrote:
>> >
>> >> Hi K,
>> >>
>> >> These errors aren't particularly informative. Can you do a smaller
>> >> version of the problem so that you only need "-np 1", and run it in
>> >> debug mode, and let me know what happens then?
>> >>
>> >> David
>> >>
>> >>
>> >> On 12/03/2012 02:21 AM, Kyunghoon Lee wrote:
>> >>> Hello all,
>> >>>
>> >>> I made a simple cantilever beam model parametrized w.r.t tip loads in
>> x
>> >> & y
>> >>> directions with RB methods for steady solutions, then extended the
>> model
>> >>> for transient solutions. For the steady model, I run
>> >>>
>> >>> $PETSC_DIR/bin/mpirun -np 8 ./beam_2d-opt -ksp_type preonly -pc_type
>> lu
>> >>> -pc_factor_mat_solver_package mumps
>> >>>
>> >>> and the training was terminated after finding two basis functions.
>> Now
>> >> for
>> >>> the transient model, if I use the same run options, it crashes while
>> >>> computing output dual norms as follows (the outputs are averaged tip
>> >>> displacements in x and y):
>> >>>
>> >>>
>> >>
>> -----------------------------------------------------------------------------------------------------------------
>> >>> $PETSC_DIR/bin/mpirun -np 8 ./beam_2d-opt -ksp_type preonly -pc_type
>> lu
>> >>> -pc_factor_mat_solver_package mumps
>> >>> *** Warning, This code is untested, experimental, or likely to see
>> future
>> >>> API changes: src/reduced_basis/rb_parametrized.C, line 41, compiled
>> Nov
>> >> 27
>> >>> 2012 at 17:00:18 ***
>> >>> Initializing training parameters with random training set...
>> >>> Parameter load_Fx: log scaling = 0
>> >>> Parameter load_Fy: log scaling = 0
>> >>>
>> >>>
>> >>> RBConstruction parameters:
>> >>> system name: RBTransientElasticity
>> >>> constrained_problem: 0
>> >>> Nmax: 10
>> >>> Basis training error tolerance: 1e-06
>> >>> Aq operators attached: 1
>> >>> Fq functions attached: 2
>> >>> n_outputs: 2
>> >>> output 0, Q_l = 1
>> >>> output 1, Q_l = 1
>> >>> Number of parameters: 2
>> >>> Parameter load_Fx: Min = -5, Max = 5, value = 0
>> >>> Parameter load_Fy: Min = -5, Max = 5, value = 5
>> >>> n_training_samples: 100
>> >>> single-matrix mode? 0
>> >>> reuse preconditioner? 1
>> >>> use a relative error bound in greedy? 0
>> >>> write out data during basis training? 0
>> >>> quiet mode? 1
>> >>>
>> >>>
>> >>> TransientRBConstruction parameters:
>> >>> Q_m: 1
>> >>> Number of time-steps: 100
>> >>> dt: 0.01
>> >>> euler_theta (time discretization parameter): 1
>> >>> delta_N (number of basis functions to add each POD-Greedy step): 1
>> >>> Using zero initial condition
>> >>>
>> >>> *** Warning, This code is deprecated, and likely to be removed in
>> future
>> >>> library versions! src/mesh/boundary_info.C, line 750, compiled Nov 27
>> >> 2012
>> >>> at 16:59:26 ***
>> >>> load_Fx: 0.000000e+00
>> >>> load_Fy: 1.000000e+00
>> >>>
>> >>> t_step[0] = 0.0e+00 truth_out[0] = 0.0000e+00 truth_out[1] =
>> >> 0.0000e+00
>> >>> t_step[1] = 1.0e-02 truth_out[0] = 6.4695e-06 truth_out[1] =
>> >> 2.9555e+01
>> >>> t_step[2] = 2.0e-02 truth_out[0] = -1.6811e-05 truth_out[1] =
>> >> 3.1663e+01
>> >>> t_step[3] = 3.0e-02 truth_out[0] = -2.3264e-05 truth_out[1] =
>> >> 3.3369e+01
>> >>> t_step[4] = 4.0e-02 truth_out[0] = -2.3126e-05 truth_out[1] =
>> >> 3.4889e+01
>> >>> t_step[5] = 5.0e-02 truth_out[0] = -2.0980e-05 truth_out[1] =
>> >> 3.6299e+01
>> >>> t_step[6] = 6.0e-02 truth_out[0] = -1.8536e-05 truth_out[1] =
>> >> 3.7639e+01
>> >>> t_step[7] = 7.0e-02 truth_out[0] = -1.6399e-05 truth_out[1] =
>> >> 3.8929e+01
>> >>> t_step[8] = 8.0e-02 truth_out[0] = -1.4733e-05 truth_out[1] =
>> >> 4.0183e+01
>> >>> t_step[9] = 9.0e-02 truth_out[0] = -1.3539e-05 truth_out[1] =
>> >> 4.1411e+01
>> >>> t_step[10] = 1.0e-01 truth_out[0] = -1.2752e-05 truth_out[1] =
>> >>> 4.2618e+01
>> >>> t_step[11] = 1.1e-01 truth_out[0] = -1.2295e-05 truth_out[1] =
>> >>> 4.3811e+01
>> >>> t_step[12] = 1.2e-01 truth_out[0] = -1.2092e-05 truth_out[1] =
>> >>> 4.4991e+01
>> >>> t_step[13] = 1.3e-01 truth_out[0] = -1.2079e-05 truth_out[1] =
>> >>> 4.6161e+01
>> >>> t_step[14] = 1.4e-01 truth_out[0] = -1.2201e-05 truth_out[1] =
>> >>> 4.7324e+01
>> >>> t_step[15] = 1.5e-01 truth_out[0] = -1.2414e-05 truth_out[1] =
>> >>> 4.8481e+01
>> >>> t_step[16] = 1.6e-01 truth_out[0] = -1.2686e-05 truth_out[1] =
>> >>> 4.9632e+01
>> >>> t_step[17] = 1.7e-01 truth_out[0] = -1.2990e-05 truth_out[1] =
>> >>> 5.0780e+01
>> >>> t_step[18] = 1.8e-01 truth_out[0] = -1.3308e-05 truth_out[1] =
>> >>> 5.1923e+01
>> >>> t_step[19] = 1.9e-01 truth_out[0] = -1.3625e-05 truth_out[1] =
>> >>> 5.3064e+01
>> >>> t_step[20] = 2.0e-01 truth_out[0] = -1.3933e-05 truth_out[1] =
>> >>> 5.4202e+01
>> >>> t_step[21] = 2.1e-01 truth_out[0] = -1.4224e-05 truth_out[1] =
>> >>> 5.5338e+01
>> >>> t_step[22] = 2.2e-01 truth_out[0] = -1.4493e-05 truth_out[1] =
>> >>> 5.6471e+01
>> >>> t_step[23] = 2.3e-01 truth_out[0] = -1.4740e-05 truth_out[1] =
>> >>> 5.7602e+01
>> >>> t_step[24] = 2.4e-01 truth_out[0] = -1.4962e-05 truth_out[1] =
>> >>> 5.8732e+01
>> >>> t_step[25] = 2.5e-01 truth_out[0] = -1.5159e-05 truth_out[1] =
>> >>> 5.9860e+01
>> >>> t_step[26] = 2.6e-01 truth_out[0] = -1.5331e-05 truth_out[1] =
>> >>> 6.0986e+01
>> >>> t_step[27] = 2.7e-01 truth_out[0] = -1.5480e-05 truth_out[1] =
>> >>> 6.2111e+01
>> >>> t_step[28] = 2.8e-01 truth_out[0] = -1.5607e-05 truth_out[1] =
>> >>> 6.3234e+01
>> >>> t_step[29] = 2.9e-01 truth_out[0] = -1.5713e-05 truth_out[1] =
>> >>> 6.4356e+01
>> >>> t_step[30] = 3.0e-01 truth_out[0] = -1.5800e-05 truth_out[1] =
>> >>> 6.5477e+01
>> >>> t_step[31] = 3.1e-01 truth_out[0] = -1.5869e-05 truth_out[1] =
>> >>> 6.6596e+01
>> >>> t_step[32] = 3.2e-01 truth_out[0] = -1.5922e-05 truth_out[1] =
>> >>> 6.7713e+01
>> >>> t_step[33] = 3.3e-01 truth_out[0] = -1.5960e-05 truth_out[1] =
>> >>> 6.8830e+01
>> >>> t_step[34] = 3.4e-01 truth_out[0] = -1.5984e-05 truth_out[1] =
>> >>> 6.9945e+01
>> >>> t_step[35] = 3.5e-01 truth_out[0] = -1.5997e-05 truth_out[1] =
>> >>> 7.1059e+01
>> >>> t_step[36] = 3.6e-01 truth_out[0] = -1.5999e-05 truth_out[1] =
>> >>> 7.2172e+01
>> >>> t_step[37] = 3.7e-01 truth_out[0] = -1.5991e-05 truth_out[1] =
>> >>> 7.3284e+01
>> >>> t_step[38] = 3.8e-01 truth_out[0] = -1.5975e-05 truth_out[1] =
>> >>> 7.4394e+01
>> >>> t_step[39] = 3.9e-01 truth_out[0] = -1.5951e-05 truth_out[1] =
>> >>> 7.5504e+01
>> >>> t_step[40] = 4.0e-01 truth_out[0] = -1.5921e-05 truth_out[1] =
>> >>> 7.6612e+01
>> >>> t_step[41] = 4.1e-01 truth_out[0] = -1.5885e-05 truth_out[1] =
>> >>> 7.7719e+01
>> >>> t_step[42] = 4.2e-01 truth_out[0] = -1.5843e-05 truth_out[1] =
>> >>> 7.8825e+01
>> >>> t_step[43] = 4.3e-01 truth_out[0] = -1.5797e-05 truth_out[1] =
>> >>> 7.9930e+01
>> >>> t_step[44] = 4.4e-01 truth_out[0] = -1.5747e-05 truth_out[1] =
>> >>> 8.1034e+01
>> >>> t_step[45] = 4.5e-01 truth_out[0] = -1.5694e-05 truth_out[1] =
>> >>> 8.2137e+01
>> >>> t_step[46] = 4.6e-01 truth_out[0] = -1.5638e-05 truth_out[1] =
>> >>> 8.3239e+01
>> >>> t_step[47] = 4.7e-01 truth_out[0] = -1.5579e-05 truth_out[1] =
>> >>> 8.4339e+01
>> >>> t_step[48] = 4.8e-01 truth_out[0] = -1.5518e-05 truth_out[1] =
>> >>> 8.5439e+01
>> >>> t_step[49] = 4.9e-01 truth_out[0] = -1.5456e-05 truth_out[1] =
>> >>> 8.6538e+01
>> >>> t_step[50] = 5.0e-01 truth_out[0] = -1.5392e-05 truth_out[1] =
>> >>> 8.7635e+01
>> >>> t_step[51] = 5.1e-01 truth_out[0] = -1.5327e-05 truth_out[1] =
>> >>> 8.8732e+01
>> >>> t_step[52] = 5.2e-01 truth_out[0] = -1.5261e-05 truth_out[1] =
>> >>> 8.9828e+01
>> >>> t_step[53] = 5.3e-01 truth_out[0] = -1.5194e-05 truth_out[1] =
>> >>> 9.0922e+01
>> >>> t_step[54] = 5.4e-01 truth_out[0] = -1.5127e-05 truth_out[1] =
>> >>> 9.2016e+01
>> >>> t_step[55] = 5.5e-01 truth_out[0] = -1.5059e-05 truth_out[1] =
>> >>> 9.3109e+01
>> >>> t_step[56] = 5.6e-01 truth_out[0] = -1.4991e-05 truth_out[1] =
>> >>> 9.4200e+01
>> >>> t_step[57] = 5.7e-01 truth_out[0] = -1.4923e-05 truth_out[1] =
>> >>> 9.5291e+01
>> >>> t_step[58] = 5.8e-01 truth_out[0] = -1.4855e-05 truth_out[1] =
>> >>> 9.6381e+01
>> >>> t_step[59] = 5.9e-01 truth_out[0] = -1.4788e-05 truth_out[1] =
>> >>> 9.7470e+01
>> >>> t_step[60] = 6.0e-01 truth_out[0] = -1.4720e-05 truth_out[1] =
>> >>> 9.8558e+01
>> >>> t_step[61] = 6.1e-01 truth_out[0] = -1.4653e-05 truth_out[1] =
>> >>> 9.9646e+01
>> >>> t_step[62] = 6.2e-01 truth_out[0] = -1.4586e-05 truth_out[1] =
>> >>> 1.0073e+02
>> >>> t_step[63] = 6.3e-01 truth_out[0] = -1.4519e-05 truth_out[1] =
>> >>> 1.0182e+02
>> >>> t_step[64] = 6.4e-01 truth_out[0] = -1.4453e-05 truth_out[1] =
>> >>> 1.0290e+02
>> >>> t_step[65] = 6.5e-01 truth_out[0] = -1.4387e-05 truth_out[1] =
>> >>> 1.0399e+02
>> >>> t_step[66] = 6.6e-01 truth_out[0] = -1.4322e-05 truth_out[1] =
>> >>> 1.0507e+02
>> >>> t_step[67] = 6.7e-01 truth_out[0] = -1.4258e-05 truth_out[1] =
>> >>> 1.0615e+02
>> >>> t_step[68] = 6.8e-01 truth_out[0] = -1.4194e-05 truth_out[1] =
>> >>> 1.0723e+02
>> >>> t_step[69] = 6.9e-01 truth_out[0] = -1.4130e-05 truth_out[1] =
>> >>> 1.0831e+02
>> >>> t_step[70] = 7.0e-01 truth_out[0] = -1.4068e-05 truth_out[1] =
>> >>> 1.0939e+02
>> >>> t_step[71] = 7.1e-01 truth_out[0] = -1.4006e-05 truth_out[1] =
>> >>> 1.1047e+02
>> >>> t_step[72] = 7.2e-01 truth_out[0] = -1.3944e-05 truth_out[1] =
>> >>> 1.1155e+02
>> >>> t_step[73] = 7.3e-01 truth_out[0] = -1.3883e-05 truth_out[1] =
>> >>> 1.1262e+02
>> >>> t_step[74] = 7.4e-01 truth_out[0] = -1.3823e-05 truth_out[1] =
>> >>> 1.1370e+02
>> >>> t_step[75] = 7.5e-01 truth_out[0] = -1.3764e-05 truth_out[1] =
>> >>> 1.1477e+02
>> >>> t_step[76] = 7.6e-01 truth_out[0] = -1.3705e-05 truth_out[1] =
>> >>> 1.1585e+02
>> >>> t_step[77] = 7.7e-01 truth_out[0] = -1.3647e-05 truth_out[1] =
>> >>> 1.1692e+02
>> >>> t_step[78] = 7.8e-01 truth_out[0] = -1.3589e-05 truth_out[1] =
>> >>> 1.1799e+02
>> >>> t_step[79] = 7.9e-01 truth_out[0] = -1.3532e-05 truth_out[1] =
>> >>> 1.1907e+02
>> >>> t_step[80] = 8.0e-01 truth_out[0] = -1.3476e-05 truth_out[1] =
>> >>> 1.2014e+02
>> >>> t_step[81] = 8.1e-01 truth_out[0] = -1.3421e-05 truth_out[1] =
>> >>> 1.2121e+02
>> >>> t_step[82] = 8.2e-01 truth_out[0] = -1.3366e-05 truth_out[1] =
>> >>> 1.2228e+02
>> >>> t_step[83] = 8.3e-01 truth_out[0] = -1.3312e-05 truth_out[1] =
>> >>> 1.2335e+02
>> >>> t_step[84] = 8.4e-01 truth_out[0] = -1.3258e-05 truth_out[1] =
>> >>> 1.2441e+02
>> >>> t_step[85] = 8.5e-01 truth_out[0] = -1.3205e-05 truth_out[1] =
>> >>> 1.2548e+02
>> >>> t_step[86] = 8.6e-01 truth_out[0] = -1.3153e-05 truth_out[1] =
>> >>> 1.2655e+02
>> >>> t_step[87] = 8.7e-01 truth_out[0] = -1.3101e-05 truth_out[1] =
>> >>> 1.2761e+02
>> >>> t_step[88] = 8.8e-01 truth_out[0] = -1.3050e-05 truth_out[1] =
>> >>> 1.2868e+02
>> >>> t_step[89] = 8.9e-01 truth_out[0] = -1.2999e-05 truth_out[1] =
>> >>> 1.2974e+02
>> >>> t_step[90] = 9.0e-01 truth_out[0] = -1.2949e-05 truth_out[1] =
>> >>> 1.3081e+02
>> >>> t_step[91] = 9.1e-01 truth_out[0] = -1.2900e-05 truth_out[1] =
>> >>> 1.3187e+02
>> >>> t_step[92] = 9.2e-01 truth_out[0] = -1.2851e-05 truth_out[1] =
>> >>> 1.3293e+02
>> >>> t_step[93] = 9.3e-01 truth_out[0] = -1.2803e-05 truth_out[1] =
>> >>> 1.3399e+02
>> >>> t_step[94] = 9.4e-01 truth_out[0] = -1.2755e-05 truth_out[1] =
>> >>> 1.3505e+02
>> >>> t_step[95] = 9.5e-01 truth_out[0] = -1.2708e-05 truth_out[1] =
>> >>> 1.3611e+02
>> >>> t_step[96] = 9.6e-01 truth_out[0] = -1.2661e-05 truth_out[1] =
>> >>> 1.3717e+02
>> >>> t_step[97] = 9.7e-01 truth_out[0] = -1.2615e-05 truth_out[1] =
>> >>> 1.3823e+02
>> >>> t_step[98] = 9.8e-01 truth_out[0] = -1.2569e-05 truth_out[1] =
>> >>> 1.3929e+02
>> >>> t_step[99] = 9.9e-01 truth_out[0] = -1.2524e-05 truth_out[1] =
>> >>> 1.4034e+02
>> >>> Compute output dual inner products
>> >>> [0]PETSC ERROR: [1]PETSC ERROR: --------------------- Error Message
>> >>> ------------------------------------
>> >>> [1]PETSC ERROR: Error in external library!
>> >>> [1]PETSC ERROR: Error reported by MUMPS in numerical factorization
>> phase:
>> >>> INFO(1)=-1, INFO(2)=4
>> >>> !
>> >>> [1]PETSC ERROR:
>> >>>
>> ------------------------------------------------------------------------
>> >>> [1]PETSC ERROR: Petsc Release Version 3.3.0, Patch 4, Fri Oct 26
>> 10:46:51
>> >>> CDT 2012
>> >>> [1]PETSC ERROR: See docs/changes/index.html for recent updates.
>> >>> [1]PETSC ERROR: See docs/faq.html for hints about trouble shooting.
>> >>> [1]PETSC ERROR: See docs/index.html for manual pages.
>> >>> [1]PETSC ERROR:
>> >>>
>> ------------------------------------------------------------------------
>> >>> [1]PETSC ERROR: ./beam_2d-opt on a arch-linu named KLee-Linux by
>> aeronova
>> >>> Mon Dec 3 14:18:44 2012
>> >>> .....
>> >>>
>> >>
>> -----------------------------------------------------------------------------------------------------------------
>> >>> If I change the run option as
>> >>>
>> >>> $PETSC_DIR/bin/mpirun -np 8 ./beam_2d-opt -ksp_type preonly
>> >>>
>> >>> then it looks working, but the output values seem wrong, and somehow
>> the
>> >>> output dual norm for the second output is zero:
>> >>>
>> >>>
>> >>
>> -----------------------------------------------------------------------------------------------------------------
>> >>> *** Warning, This code is untested, experimental, or likely to see
>> future
>> >>> API changes: src/reduced_basis/rb_parametrized.C, line 41, compiled
>> Nov
>> >> 27
>> >>> 2012 at 17:00:18 ***
>> >>> Initializing training parameters with random training set...
>> >>> Parameter load_Fx: log scaling = 0
>> >>> Parameter load_Fy: log scaling = 0
>> >>>
>> >>>
>> >>> RBConstruction parameters:
>> >>> system name: RBTransientElasticity
>> >>> constrained_problem: 0
>> >>> Nmax: 10
>> >>> Basis training error tolerance: 1e-06
>> >>> Aq operators attached: 1
>> >>> Fq functions attached: 2
>> >>> n_outputs: 2
>> >>> output 0, Q_l = 1
>> >>> output 1, Q_l = 1
>> >>> Number of parameters: 2
>> >>> Parameter load_Fx: Min = -5, Max = 5, value = 0
>> >>> Parameter load_Fy: Min = -5, Max = 5, value = 5
>> >>> n_training_samples: 100
>> >>> single-matrix mode? 0
>> >>> reuse preconditioner? 1
>> >>> use a relative error bound in greedy? 0
>> >>> write out data during basis training? 0
>> >>> quiet mode? 1
>> >>>
>> >>>
>> >>> TransientRBConstruction parameters:
>> >>> Q_m: 1
>> >>> Number of time-steps: 100
>> >>> dt: 0.01
>> >>> euler_theta (time discretization parameter): 1
>> >>> delta_N (number of basis functions to add each POD-Greedy step): 1
>> >>> Using zero initial condition
>> >>>
>> >>> *** Warning, This code is deprecated, and likely to be removed in
>> future
>> >>> library versions! src/mesh/boundary_info.C, line 750, compiled Nov 27
>> >> 2012
>> >>> at 16:59:26 ***
>> >>> load_Fx: 0.000000e+00
>> >>> load_Fy: 1.000000e+00
>> >>>
>> >>> t_step[0] = 0.0e+00 truth_out[0] = 0.0000e+00 truth_out[1] =
>> >> 0.0000e+00
>> >>> t_step[1] = 1.0e-02 truth_out[0] = -1.6962e-04 truth_out[1] =
>> >> 1.5122e-01
>> >>> t_step[2] = 2.0e-02 truth_out[0] = -2.3491e-04 truth_out[1] =
>> >> 1.5378e-01
>> >>> t_step[3] = 3.0e-02 truth_out[0] = -2.9471e-04 truth_out[1] =
>> >> 1.5459e-01
>> >>> t_step[4] = 4.0e-02 truth_out[0] = -3.3419e-04 truth_out[1] =
>> >> 1.5489e-01
>> >>> t_step[5] = 5.0e-02 truth_out[0] = -3.5753e-04 truth_out[1] =
>> >> 1.5500e-01
>> >>> t_step[6] = 6.0e-02 truth_out[0] = -3.7049e-04 truth_out[1] =
>> >> 1.5505e-01
>> >>> t_step[7] = 7.0e-02 truth_out[0] = -3.7738e-04 truth_out[1] =
>> >> 1.5507e-01
>> >>> t_step[8] = 8.0e-02 truth_out[0] = -3.8093e-04 truth_out[1] =
>> >> 1.5508e-01
>> >>> t_step[9] = 9.0e-02 truth_out[0] = -3.8271e-04 truth_out[1] =
>> >> 1.5508e-01
>> >>> t_step[10] = 1.0e-01 truth_out[0] = -3.8359e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[11] = 1.1e-01 truth_out[0] = -3.8401e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[12] = 1.2e-01 truth_out[0] = -3.8422e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[13] = 1.3e-01 truth_out[0] = -3.8431e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[14] = 1.4e-01 truth_out[0] = -3.8436e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[15] = 1.5e-01 truth_out[0] = -3.8438e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[16] = 1.6e-01 truth_out[0] = -3.8439e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[17] = 1.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[18] = 1.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[19] = 1.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[20] = 2.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[21] = 2.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[22] = 2.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[23] = 2.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[24] = 2.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[25] = 2.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[26] = 2.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[27] = 2.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[28] = 2.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[29] = 2.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[30] = 3.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[31] = 3.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[32] = 3.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[33] = 3.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[34] = 3.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[35] = 3.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[36] = 3.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[37] = 3.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[38] = 3.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[39] = 3.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[40] = 4.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[41] = 4.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[42] = 4.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[43] = 4.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[44] = 4.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[45] = 4.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[46] = 4.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[47] = 4.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[48] = 4.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[49] = 4.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[50] = 5.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[51] = 5.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[52] = 5.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[53] = 5.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[54] = 5.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[55] = 5.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[56] = 5.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[57] = 5.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[58] = 5.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[59] = 5.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[60] = 6.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[61] = 6.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[62] = 6.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[63] = 6.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[64] = 6.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[65] = 6.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[66] = 6.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[67] = 6.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[68] = 6.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[69] = 6.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[70] = 7.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[71] = 7.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[72] = 7.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[73] = 7.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[74] = 7.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[75] = 7.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[76] = 7.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[77] = 7.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[78] = 7.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[79] = 7.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[80] = 8.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[81] = 8.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[82] = 8.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[83] = 8.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[84] = 8.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[85] = 8.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[86] = 8.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[87] = 8.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[88] = 8.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[89] = 8.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[90] = 9.0e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[91] = 9.1e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[92] = 9.2e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[93] = 9.3e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[94] = 9.4e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[95] = 9.5e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[96] = 9.6e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[97] = 9.7e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[98] = 9.8e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> t_step[99] = 9.9e-01 truth_out[0] = -3.8440e-04 truth_out[1] =
>> >>> 1.5508e-01
>> >>> Compute output dual inner products
>> >>> output_dual_innerprods[0][0] = 8.3935e+01
>> >>> output_dual_innerprods[1][0] = 0.0000e+00
>> >>>
>> >>> ---- Performing Greedy basis enrichment ----
>> >>>
>> >>> ---- Basis dimension: 0 ----
>> >>> Performing RB solves on training set
>> >>> Maximum (absolute) error bound is 1.8090e+00
>> >>>
>> >>> Performing truth solve at parameter:
>> >>> load_Fx: 4.971986e+00
>> >>> load_Fy: 4.072318e+00
>> >>> .....
>> >>>
>> >>
>> -----------------------------------------------------------------------------------------------------------------
>> >>> I'm afraid the problem has something to do with using proper solver
>> >>> options, and I'd appreciate any comments/suggestions.
>> >>>
>> >>> K. Lee.
>> >>>
>> >>
>> ------------------------------------------------------------------------------
>> >>> Keep yourself connected to Go Parallel:
>> >>> BUILD Helping you discover the best ways to construct your parallel
>> >> projects.
>> >>> http://goparallel.sourceforge.net
>> >>> _______________________________________________
>> >>> Libmesh-users mailing list
>> >>> Lib...@li...
>> >>> https://lists.sourceforge.net/lists/listinfo/libmesh-users
>> >>
>> >>
>> >>
>> ------------------------------------------------------------------------------
>> >> Keep yourself connected to Go Parallel:
>> >> BUILD Helping you discover the best ways to construct your parallel
>> >> projects.
>> >> http://goparallel.sourceforge.net
>> >> _______________________________________________
>> >> Libmesh-users mailing list
>> >> Lib...@li...
>> >> https://lists.sourceforge.net/lists/listinfo/libmesh-users
>> >>
>> >
>> ------------------------------------------------------------------------------
>> > LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
>> > Remotely access PCs and mobile devices and provide instant support
>> > Improve your efficiency, and focus on delivering more value-add services
>> > Discover what IT Professionals Know. Rescue delivers
>> > http://p.sf.net/sfu/logmein_12329d2d
>> > _______________________________________________
>> > Libmesh-users mailing list
>> > Lib...@li...
>> > https://lists.sourceforge.net/lists/listinfo/libmesh-users
>>
>>
>>
>> ------------------------------------------------------------------------------
>> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
>> Remotely access PCs and mobile devices and provide instant support
>> Improve your efficiency, and focus on delivering more value-add services
>> Discover what IT Professionals Know. Rescue delivers
>> http://p.sf.net/sfu/logmein_12329d2d
>> _______________________________________________
>> Libmesh-users mailing list
>> Lib...@li...
>> https://lists.sourceforge.net/lists/listinfo/libmesh-users
>>
>
>
>
|
