3.6. Constrained Problems

The major difference between a constrained and unconstrained problem is that for a constrained problem, an optimal solution must be sought in a feasible region; for an unconstrained problem, the feasible region contains the entire design domain. For a constrained problem, bringing an infeasible design into a feasible region is critical, in which gradients of active constraints are taken into consideration when determining the search direction for the next design. In this section, we first outline the nature of the constrained optimization problem and the concept of solution techniques. In Section 3.6.2, we then discuss a widely accepted strategy for dealing with the constraint functions, the so-called ε-active strategy. ...

Get Design Theory and Methods using CAD/CAE now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.