Creating constraints

A constraint constrains the Optimization module from making changes to the topology of the model during the optimization. If a constraint cannot be satisfied, the optimization is not feasible. To configure a constraint, you select a design response or a combination of design responses and specify the value of the constraint; for example, the volume of the optimized model can be constrained to be 50% of the original volume.

When you execute an optimization process, Abaqus generates history output from the constraints you defined in the Optimization module. For volume design responses the history output is always reported as a fraction of the initial value. For all other design responses the history output is reported as an absolute value.

Related Topics
About constraints

  1. From the main menu bar, select ConstraintCreate.

    The Create Constraint dialog box appears.

    Tip: You can initiate the Create procedure in two other ways:

    • Click Create in the Constraint Manager. (You can display the Constraint Manager by selecting ConstraintManager from the main menu bar.)

    • Click the tool in the Optimization module toolbox.

  2. From the Create Constraint dialog box that appears, enter the name of the constraint.

    The Edit Constraint dialog box appears.

  3. From the menu, select the design response that will be used to define this constraint. Alternatively, you can select to create a new design response and add it to the list of design responses.

    The constraint editor displays the type of the design response, which is the variable you are constraining. If you are performing a condition-based optimization, you must select a volume constraint.

  4. Define the constraint. The editor provides different options, depending on the type of optimization task you selected.

    • If you are defining a constraint for a condition-based topology or shape optimization, the volume constraint can be either of the following:

      • Equal to a fixed value.

      • Equal to a fraction of the value before the optimization starts.

    • If you are defining a constraint for a general topology optimization, the constraint can be one of the following:

      • Less than, greater than, or equal to a fixed value.

      • Less than, greater than, or equal to a fraction of the value before the optimization starts.

      Note:

      During a condition-based topology or shape optimization, the volume is reduced incrementally until the volume constraint is satisfied during the last step. In contrast, a general topology optimization satisfies the volume constraint in the first design cycle. As a consequence, a general topology optimization starts with a much weaker structure, and the element density is scaled to satisfy the remaining constraints and objectives.

  5. Click OK to save your data and to exit the editor.