Abstract: A new Hybrid Genetic Algorithm(HGA), which combines the genetic algorithm with the traditional local search steps and uses new criterion of crossover and mutation, is proposed in this paper. The new HGA can avoid the slow convergence rate and premature convergence, which are two main drawbacks of conventional genetic algorithms. The algorithm can be well applied to a class of global optimization problems for certain continuous functions with box constraints and has powerful ability to find global optimums. Numerical experiments show that the new algorithm can yield encouraging results.

Key words: genetic algorithm, local search, global optimization

摘要： 为了克服传统遗传算法收敛速度缓慢且易于收敛到局部最优解的缺点，该文将遗传算法与传统的局部搜索方法相结合，采用新的交叉变异准则，提出一种新型的混合遗传算法。该算法可以很好地处理一类带上下界约束的全局优化问题，具有很强的全局寻优能力。数值实验表明，该算法的计算结果明显优于传统遗传算法。

关键词: 遗传算法, 局部搜索, 全局优化

CLC Number: 

• TP18