【摘要】 本文提出用遗传—模拟退火算法(GASA)混合优化策略来求解双层规划模型。混合优化策略结合了遗传算法的并行结构和模拟退火算法的概率突跳性,提高了找到全局最优解的可靠性和计算效率。数值模拟实验表明算法性能良好,GASA混合优化策略求得全局最优解时的进化代数比单一的遗传算法减少约35％,比模拟退火算法的迭代次数减少约50％。更多还原

【Abstract】 The author proposes GASA hybrid optimization strategy for the bilevel programming models. The proposed strategy is based on single genetic algorithm and simulated annealing algorithm and hence it combines the parallel searching structure of genetic algorithm with the probabilistic jumping property of simulated annealing algorithm, as a result, the GASA hybrid optimization strategy is much more efficient and more likely to find the global optimum, which is fully showed by the performance of the strategy using numerical example. In our numerical example, to find the global optimum, the evolutionary generations of GASA is saved about 35% than those of single GA, 50% than those of SA.更多还原