Understanding the role of the Optimization module

You can use the Optimization module to perform the following tasks:

Create optimization tasks

An optimization task contains the definition of your optimization. You run an optimization in the Job module using an optimization process. An optimization process refers to an optimization task.

Create design responses

A design response is a single scalar value that is extracted from an optimization. A design response can be extracted directly from the output database, such as the volume of the model. Alternatively, the Optimization module can extract data from the output database and calculate the design response, such as the total strain energy of the model, a measure of its flexibility.

Create objective functions

An objective function defines the objective of the optimization and refers to the value of a design response or a combination of design responses. For example, the objective function of the optimization can be to minimize the total strain energy in the model (maximize its stiffness).

Create constraints

Constraints define the changes that the Optimization module can apply to the topology or the shape of the model during the optimization. For example, the volume of the optimized model can be constrained to be 50% of the original volume. If a constraint cannot be satisfied, the optimization is not feasible. A constraint also refers to the value of a design response, but it cannot refer to a combination of design responses.

Create geometric restrictions

A geometric restriction places restrictions on the changes that the Optimization module can make to the topology of the model. Geometrical restrictions include frozen regions from which material cannot be removed and manufacturing constraints, such as restrictions on cavities and undercuts, that would prevent the optimized model from being removed from a mold.

Create stop conditions

A stop condition is an indicator that the optimization has converged to a solution. For example, an optimization can be considered complete after a specified number of iterations or when the change in an optimization function between iterations is less than a specified value.