摘要
针对传统遗传算法的不足 ,提出了一种采用最优个体保护、适应函数调整、双点交叉相结合的方法进行繁殖操作 ,同时又能自适应地改变交换和变异率的改进的遗传算法 ,最后在水质模型参数估值应用中将此改进法与传统遗传算法的计算结果进行了对比 ,结果证明前者具有更好的收敛性能。
The mechanism of the traditional genetic algorithm is introduced. A modified adaptive GA(MAGA) is presented according to the weakness of the traditional GA(MAGA) by using optimal individual reservation, adaptive function adjustment, two point crossover and automatic adaptive regulation in the probability of crossover and mutation. Comparison of MAGA and SGA in application to the parameter estimation of water environment proves that the former one is more effective.
出处
《水电能源科学》
2002年第1期38-40,共3页
Water Resources and Power
基金
国家自然科学基金资助项目 (4 92 0 1 0 1 5 )
高等学校优秀年轻教师教学科研奖励计划资助
关键词
遗传算法
水环境
参数估值
优化
水质评价
genetic algorithm
water environment
estimation of parameters
optimization
