Front Door of a Car – Reducing Weight

This section showcases the algorithm on a relatively large model of a car door.

The FE Model

The model is for the analysis of the sagging behavior of a car door. The Loads applied are, a moment load on the two hinges, which are constrained using boundary conditions and a concentrated load on the side opposite to the hinges (near the locking mechanism). The hinges and the locking area are excluded from the optimization domain, clustering has been used to create mechanically sensible results. The result was requested from the following discrete range:

Available Thicknesses
0.152 0.531
0.163 0.607
0.17 0.683
0.191 0.759
0.208 0.836
0.229 0.912
0.246 1.062
0.267 1.214
0.305 1.367
0.343 1.519
0.378 1.709
0.417 1.897
0.455 2.278
The FE model of the car door with boundary conditions and loads.


The Optimization Problem

we consider a basic weight minimization problem with a stiffness constraint. The par file is shown below.

		
  . . .
  
  DVCON_SIZING
    ID_NAME         = DVCON_SIZING_set_discr
    EL_GROUP        = ALL_ELEMENTS
    CHECK_TYPE      = DISCRETE
    DISCR_LIST_FILE = Sheet_sizing_truncated.csv
    DISCR_CYCLE     = 10
    DISCR_INTERVAL  = 4
    DISCR_FRACTION  = 0.2
    DISCR_CHANGE    = 10
  END_
  
  . . .

  OPT_PARAM
    STOP_CRITERION_ITER = 28
  END_
  
  . . .

Optimization Results

The figures below show the optimization results. A comparison between the optimization with continuous and discrete variables is also done.

History of objective and constraints from the sizing optimization for discrete variables.


History of objective and constraints from the sizing optimization for continuous variables.


Resulting thicknesses from the sizing optimization for discrete variables.




Resulting thicknesses from the sizing optimization for continuous variables.




The difference in the objective functions for the standard and discrete optimization is less than 2% and the constraints are fulfilled within an error of 0.001 in this example.