【摘要】 遗传算法作为近年来的热点在各个方面都得到了广泛的应用。但是遗传算法有其固有的缺陷,即易早熟,局部寻优能力差。为了改善这两方面的性能,该文从传统遗传算法(SGA)的原理出发,讨论了在初始种群中均匀布种的优越性。在此基础上引入了灾变操作,设计了对这种操作的控制方法。然后对传统的遗传操作(包括交叉操作和变异操作)进行了改进,提高了这两种操作的效率。最后应用C/C++实现了新算法,并对2个著名优化方法测试函数进行优化计算。计算结果证明新算法具有很强的摆脱局部极值的能力和比较快的收敛速度。更多还原

【Abstract】 Genetic algorithm is widely used in recent years as a common method.But GA has some defects like premature and slow convergence speed in local area. For improving the performance of GA,this paper introduces a catastrophe operation and control method based on the principle of simple genetic algorithm and advantages of even seeding in initial population.The traditional genetic operations,such as crossover and mutation,are improved in efficiency.At last,the new algorithm is programmed using C/C++ and tested by the calculation of two famous test functions of optimization method.The results confirm that the new algorithm not only has strong ability of getting out the restriction of local extremum point but also has fast convergent speed.更多还原