Materials, restraints, and loads should be
consistent for all referenced studies used in an optimization study.Before creating an optimization study, you should create at least one stud that you will use to define the objective function and constraints. To create an optimization study, right-click the document icon (top icon) in the COSMOS AnalysisManager tree and select Study. Give a name for the study and select the Optimization analysis type. Note that you do not need to specify the mesh type. The optimization study uses the meshes of the studies used to define the objective and constraints. You can specify the maximum allowable number of optimization loops in the properties of the optimization study.
An optimization study is defined by an objective function, design variables, and constraints. For example, you can vary dimensions in your model to seek the least possible material while stresses cannot exceed a stress limit. In this case, reducing volume is the objective function, the varying dimensions are the design variables, and the stress limit is the behavior constraint.
Materials, restraints, and loads should be
consistent for all referenced studies used in an optimization study.
For each optimization study, the program creates three icons named Objective, Design Variables, and Constraints, and a Report folder in the COSMOS AnalysisManager tree.
Objective. The objective (or objective function) defines the goal of the optimization process. You can only specify one objective in an optimization study. The objective can be one of the following:
minimizing volume or mass
For parts, minimizing the
volume is identical to minimizing the weight since a part is made of one
material. For assemblies, minimizing volume and weight can be different
if parts are made of materials with different densities.
maximizing buckling load factor
maximizing or minimizing the resonant frequency
To define the objective function, right-click the Objective icon and click Add.
Design Variables. Design variables are the changeable dimensions of the model. Any dimension can be defined as a design variable. For each design variable, you need to specify the lower and upper bounds as well as a tolerance.
When specifying the bounds of a design variable, make sure that they do not contradict with any relations specified in the model.
You can define up to 20 design variables in an optimization study.
To define design variables, right-click the Design Variables icon, select Add., select a dimension from the graphics area, and specify lower and upper bounds. If you do not see dimensions of the model to select, double-click the model in the graphics area.
Constraints. Constraints, also called behavior constraints, define the conditions that the optimized design must satisfy.
A constraint is associated with an initial study and can be defined as a component of one of the following types:
For static studies. Nodal or element stresses, strains, or displacements
For buckling studies. First through tenth buckling load factors
For frequency studies. First through tenth resonant frequencies
For thermal studies. Temperatures, temperature gradients, or heat fluxes
To define constraints, right-click the Constraints icon and select Add. A constraint is defined by a lower bound, an upper bound, and a tolerance. You can define up to 60 constraints in an optimization study.