Constrained Optimization Problems in Mechanical Engineering Design Using a Real-Coded Steady-State Genetic Algorithm

Abstract

A parameter-less adaptive penalty scheme for steady-state genetic algorithms applied to constrained optimization problems was proposed previously by two of co-authors of this paper. For each constraint, a penalty parameter is adaptively computed along the run according to information extracted from the current population such as the existence of feasible individuals and the level of violation of each constraint. In this paper the performance of this scheme is extended using test problems from the mechanical engineering design, also, largely tested in the evolutionary computation literature. Using real coding, rank-based selection and operators available in the literature very competitive results are obtained and those are compared with other techniques.