【摘要】 首先简要介绍了传统的遗传算法,然后分析了遗传算法的优点和不足。针对遗传算法容易产生早熟现象和局部寻优能力差的特点,通过引入改进的灾变操作和模拟退火算法跟遗传算法相结合,而增强了算法的全局收敛性,并且提高了算法的收敛速度。最后使用一个典型的遗传算法性能测试函数验证了改进算法对函数最优化的有效性,其性能明显优于传统的遗传算法和模拟退火算法。更多还原

【Abstract】 This paper first introduces the traditional genetic algorithm briefly, then analysis the advantages and disadvantages of GA. Aimed at the premature convergence problem and the badly local-optimization existed in GA, an improved genetic-annealing algorithm is proposed by incorporating disaster-modification and the simulated annealing into genetic algorithm. The algorithm enhances the globe convergence, and improves the convergence velocity. Simulation results based on one benchmarks demonstrate the effectiveness of the proposed algorithms applied to optimization of functions. The performances of the improved genetic-annealing algorithm are quite better than those of classical GA and SA.更多还原