Adjoint class

Adjoint solve

class HydrOpTop.Adjoints.Adjoint_Solve(algo=None)

Solve matrix equation Ax=B

Steady state sensitivity

class HydrOpTop.Adjoints.No_Adjoint(parametrized_mat_props, p_ids)

A dummy adjoint class that solve no adjoint…

class HydrOpTop.Adjoints.Sensitivity_Steady_Simple(solved_vars, parametrized_mat_props, solver, p_ids)

Compute the derivative of the cost function according to the material distribution parameter p considering a steady-state simulation with only one equality constraint (g(x,p) = 0) Arguments: Note: vector derivative should be numpy array, and matrix in (I,J,data) format as output by solver.get_sensitivity() method. I,J are 0 based indexing If cost_deriv_mat_prop is None, assume the cost function does not depend on the material property distribution.

compute_sensitivity(p, dc_dYi, dc_dXi, Xi_name)

Compute the total cost function derivative according to material density parameter p.

Parameters:

• p – the material parameter

• dc_dYi – derivative of the function wrt solved variable (Solver ordering)

• dc_dXi – derivative of the function wrt input variable (p ordering)

• Xi_name – name of the function input variables