摘要
根据梯级水电站优化调度特点,建立遗传算法(GA)求解多阶段最优化问题的数学模型。针对标准遗传算法(SGA)局部寻优能力较差、易早熟等不足之处,从编码方法、遗传算子和混合算法方面对其进行改进,提出了采用超立方体浮点数编码自适应遗传算法(AGA)和超立方体浮点数编码遗传模拟退火算法(SA-GA)。通过16种不同策略的GA在雅砻江梯级优化调度中的应用,其结果表明了改进策略在解决水库群优化问题方面的有效性和优越性。最后将GA与动态规划(DP)算法的性能进行比较分析,充分体现了GA的优点。
According to the characteristics of optimal dispatch of cascade reservoirs for hydropower stations the mathematical model for solving the multi-stage optimization problem is established.To overcome the disadvantage of low ability of local searching and easy-premature of standard genetic algorithm,the coding approach,new genetic operator and hybrid algorithm are introduced to improve the algorithm.On this basis the hypercube floating-point coding adaptive genetic algorithm(AGA)and hypercube floating-point co...
出处
《水利学报》
EI
CSCD
北大核心
2008年第5期550-556,共7页
Journal of Hydraulic Engineering
基金
973重大项目(2003CB415203)
关键词
遗传算法
优化调度
梯级水电站
improved genetic algorithm
optimal dispatch
coding approach
operator
hybrid algorithm
cascade hydropower station