Abstract interface for Python-defined objective functions. More...

Inheritance diagram for ObjectiveFunctionData:

Public Member Functions
virtual MoFEMErrorCode	evalObjectiveFunction (MatrixDouble &coords, boost::shared_ptr< MatrixDouble > u_ptr, boost::shared_ptr< MatrixDouble > stress_ptr, boost::shared_ptr< MatrixDouble > strain_ptr, boost::shared_ptr< VectorDouble > o_ptr)=0
	Evaluate objective function f(coords, u, stress, strain)

virtual MoFEMErrorCode	evalObjectiveGradientStress (MatrixDouble &coords, boost::shared_ptr< MatrixDouble > u_ptr, boost::shared_ptr< MatrixDouble > stress_ptr, boost::shared_ptr< MatrixDouble > strain_ptr, boost::shared_ptr< MatrixDouble > o_ptr)=0
	Evaluate gradient of objective function w.r.t. stress: ∂f/∂σ

virtual MoFEMErrorCode	evalObjectiveGradientStrain (MatrixDouble &coords, boost::shared_ptr< MatrixDouble > u_ptr, boost::shared_ptr< MatrixDouble > stress_ptr, boost::shared_ptr< MatrixDouble > strain_ptr, boost::shared_ptr< MatrixDouble > o_ptr)=0
	Evaluate gradient of objective function w.r.t. strain: ∂f/∂ε

virtual MoFEMErrorCode	numberOfModes (int block_id, int &modes)=0
	Return number of topology optimization modes for given block.

virtual MoFEMErrorCode	blockModes (int block_id, MatrixDouble &coords, std::array< double, 3 > &centroid, std::array< double, 6 > &bbox, MatrixDouble &o_ptr)=0
	Define spatial modes for topology optimization.

virtual	~ObjectiveFunctionData ()=default

Detailed Description

Abstract interface for Python-defined objective functions.

This interface allows users to define custom objective functions in Python that can be called from C++ during the optimization process. The objective function and its gradients are evaluated at integration points using:

Coordinates of integration points
Displacement field values
Stress and strain tensors

Key methods:

evalObjectiveFunction: Compute f(x,u,σ,ε)
evalObjectiveGradientStress: Compute ∂f/∂σ
evalObjectiveGradientStrain: Compute ∂f/∂ε
numberOfModes: Return number of topology modes for optimization
blockModes: Define spatial topology modes for shape optimization

Examples: mofem/tutorials/vec-7/adjoint.cpp.

Definition at line 124 of file adjoint.cpp.

Constructor & Destructor Documentation

◆ ~ObjectiveFunctionData()

virtual ObjectiveFunctionData::~ObjectiveFunctionData ( )

virtualdefault

Examples: mofem/tutorials/vec-7/adjoint.cpp.

Member Function Documentation

◆ blockModes()

virtual MoFEMErrorCode ObjectiveFunctionData::blockModes	(	int	block_id,
		MatrixDouble &	coords,
		std::array< double, 3 > &	centroid,
		std::array< double, 6 > &	bbox,
		MatrixDouble &	o_ptr
	)

pure virtual

Define spatial modes for topology optimization.

Implemented in ObjectiveFunctionDataImpl.

Examples: mofem/tutorials/vec-7/adjoint.cpp.

◆ evalObjectiveFunction()

virtual MoFEMErrorCode ObjectiveFunctionData::evalObjectiveFunction	(	MatrixDouble &	coords,
		boost::shared_ptr< MatrixDouble >	u_ptr,
		boost::shared_ptr< MatrixDouble >	stress_ptr,
		boost::shared_ptr< MatrixDouble >	strain_ptr,
		boost::shared_ptr< VectorDouble >	o_ptr
	)

pure virtual

Evaluate objective function f(coords, u, stress, strain)

Implemented in ObjectiveFunctionDataImpl.

Examples: mofem/tutorials/vec-7/adjoint.cpp.

◆ evalObjectiveGradientStrain()

virtual MoFEMErrorCode ObjectiveFunctionData::evalObjectiveGradientStrain	(	MatrixDouble &	coords,
		boost::shared_ptr< MatrixDouble >	u_ptr,
		boost::shared_ptr< MatrixDouble >	stress_ptr,
		boost::shared_ptr< MatrixDouble >	strain_ptr,
		boost::shared_ptr< MatrixDouble >	o_ptr
	)

pure virtual

Evaluate gradient of objective function w.r.t. strain: ∂f/∂ε

Implemented in ObjectiveFunctionDataImpl.

Examples: mofem/tutorials/vec-7/adjoint.cpp.

◆ evalObjectiveGradientStress()

virtual MoFEMErrorCode ObjectiveFunctionData::evalObjectiveGradientStress	(	MatrixDouble &	coords,
		boost::shared_ptr< MatrixDouble >	u_ptr,
		boost::shared_ptr< MatrixDouble >	stress_ptr,
		boost::shared_ptr< MatrixDouble >	strain_ptr,
		boost::shared_ptr< MatrixDouble >	o_ptr
	)