提出了一种求解约束函数优化问题的新算法。设计了3种新的多父体杂交算子。这3种算子都使用了统计信息来决定搜索方向,使算法具有较高的收敛速度,同时又具有互补的特性,使得种群在演化过程中能保持较好的多样性,不容易陷入局部最优。对常见测试函数的数值实验证实了新方法的有效性、通用性和稳健性,其性能优于现有的一些演化算法。
A new approach (MMEA) is presented to handle constrained function optimization problems using evolutionary algorithms. It designs three novel multi-parent crossover operators which can speedup the constringency dramatically because of their strong direction. Meanwhile, the complementarity among these crossover operators can maintain population diversity , which makes MMEA more likely to find the global optimum than other evolutionary algorithms. The new approach is compared against other evolutionary optimization techniques in several benchmark functions. The results show that the new approach is a general, effective and robust method. Its performance outperforms some other techniques.