求解全局优化问题的智能遗传算法

Intelligent Genetic Algorithm for Global Optimization Problem

提出了一种智能遗传算法,该算法融合了5种交叉算子、8种变异算子和5种灾变算子,能根据当前优化结果智能地选择交叉算子、变异算子和灾变算子,在不影响搜索过程随机性的前提下收敛于全局最优解。不同于传统遗传算法,本算法增加了对各种算子优化性能的统计,在优化过程中尽可能使用那些优化性能高的算子,从而提高了智能遗传算法的优化性能。为了验证本算法的性能,采用12种传统遗传算法和本算法同时对20个测试函数进行了求解。最终的数据实例表明,方法是可行的、正确的和有效的。

Intelligent Genetic Algorithm （IGA） was proposed for solving global optimization problem. Five cross operators, eight mutation operators and five rebound operators were joined into IGA. This algorithm selects the appropriate cross operator, mutation operator and rebound operator according to the current optimization results, and converges the global optimization solution without the influence of random search process. Other than traditional GA, IGA increased statistical function to the optimization performance of these operators and applied the appropriate operator with best performance to advance optimization process. In order to validate the performance of this algorithm, 20 testing functions were solved by IGA and other 12 traditional genetic algorithms. Numerous examples suggest that the algorithm is feasible, correct and valid.