一种改进的遗传算法

针对传统遗传算法在编码方案及交叉操作中存在的局限性，提出了一种新的遗传算法的改进 方法．该方法（１）以实数编码代替二进制编码，有效地解决了传统遗传算法中二进制编码串的长度与 计算精度、运算量之间的矛盾，（２）根据适应度值对父染色体进行重组操作，克服了传统遗传算法中 交叉操作所存在的盲目性．最后，以求解自然对数和神经网络的训练为例验证了所提出方法的有效 性．

Conventional genetic algorithm is modified by replacing the concepts named crossover and binary encoding with that of recombination and real number encoding respectively.The modified version of genetic algorithm can enhance the convergence rate dramatically compared with the conventional genetic algorithm. The effectiveness of the method is demonstrated by numerical simulations.