Problems while solving Implicit Runge Kutta method.

[sora]

Hi all,

I have to solve the continuity equation in 1D using Implicit Runge Kutta method. 

The formula I made is this: d(U_t+1)/dt = [-grad(\Gamma_e)+R_i]/exp(U_t)

      \Gamma_e ,R_i : Load the value calculated by another function.

      U_t : Solution calculated at the previous time.

When implicit_runge_kutta.evolve_one_time_step(..) and evaluate_diffusion_energy(..) is repeated, the value(U_t) I load changes.

U_t, the value that should be divided, becomes 0 and an error occurs.

I do not know why this value changes. I just want to get a fixed value.

template<int dim>

Vector<double> Electron<dim>::evaluate_diffusion_energy (const Vector<double> &y)

{

    Vector<double> tmp(dof_handler.n_dofs());

    tmp = 0.;

    system_matrix.vmult(tmp,y); //tmp=(A-B-C)*y

   const QGauss<dim> quadrature_formula(fe_degree + 1);

    FEValues<dim> fe_values(fe,

                          quadrature_formula,

                          update_values |update_gradients| update_quadrature_points |

                            update_JxW_values);

    const unsigned int dofs_per_cell = fe.dofs_per_cell;

    const unsigned int n_q_points    = quadrature_formula.size();

    std::vector<Tensor<1, dim>> grad_V(n_q_points);

    std::vector<Tensor<1, dim>> grad_energyflux(n_q_points);

    Vector<double> cell_source(dofs_per_cell);

    std::vector<types::global_dof_index> local_dof_indices(dofs_per_cell);

    for (const auto &cell : dof_handler.active_cell_iterators())

      {

        cell->get_dof_indices(local_dof_indices);

        cell_source = 0.;

        fe_values.reinit(cell);

        fe_values.get_function_gradients(Potential,grad_V);

        fe_values.get_function_gradients(EnergyFlux,grad_energyflux);

        for (unsigned int q_point = 0; q_point < n_q_points; ++q_point)

          {

            double E = coeffi_E(local_dof_indices,grad_V[q_point]);

            for (unsigned int i = 0; i < dofs_per_cell; ++i)

              cell_source(i) += fe_values.JxW(q_point)*fe_values.shape_value(i, q_point)* // phi_i(x)

                                (E-grad_energyflux[i][0])/exp(initial_EnergyDensity[local_dof_indices[i]]);

       }

        constraints_2.distribute_local_to_global(cell_source,

                                                 local_dof_indices,

                                                 tmp); //(A-B-C)*y+E

      }

    Vector<double> value(dof_handler.n_dofs());

    inverse_mass_matrix.vmult(value,tmp); //value=M^-1*tmp

    return value;

}

I look forward to hearing from you, thank you.

Best regards,

            sora

[Bruno Turcksin]

Sora,

On Thursday, August 8, 2019 at 3:42:54 AM UTC-4, sora wrote:

I have to solve the continuity equation in 1D using Implicit Runge Kutta method. 

The formula I made is this: d(U_t+1)/dt = [-grad(\Gamma_e)+R_i]/exp(U_t)

      \Gamma_e ,R_i : Load the value calculated by another function.

      U_t : Solution calculated at the previous time.

When implicit_runge_kutta.evolve_one_time_step(..) and evaluate_diffusion_energy(..) is repeated, the value(U_t) I load changes.

U_t, the value that should be divided, becomes 0 and an error occurs.

What becomes zero? U_t or exp(initial_EnergyDensity)? Where is initial_EnergyDensity updated?

Best,

Bruno