A Novel Constraint-Handling Method Based on Evolutionary Algorithm
Ximing Liang, Wen Long, Haoyu Qn, Shanchun Li · 2009
A new method is proposed to deal with constrained optimization problem to overcome certain disadvantages of the current methods. The proposed algorithm adopts a new constraint-handling method to deal with constraints and does not introduce penalty parameters. In the evolutionary process, our algorithm searches the solution space of the problem through a mixture crossover of feasible and infeasible solutions, and performs a selection operation on feasible and infeasible populations respectively. In addition, the boundary mutation and non-uniform mutation operators are used to enrich the exploration and exploitation abilities of the proposed approach. It uses a diversity conservation mechanism to allow infeasible solution to remain in the population. Numerical results show that the proposed method is effective.