基于改进差分进化算法的水火电短期优化调度  

Short-term Dispatch Study of Hydrothermal Power Systems Based on Enhanced Differential Evolution

作　　者：黄迪[1] 徐刚[2,3] 向红艳[1] 李英海[2] 

机构地区：[1]湖北清江水电开发有限责任公司,湖北宜昌443000 [2]三峡大学水利与环境学院,湖北宜昌443002 [3]水资源安全保障湖北省协同创新中心,湖北武汉430072

出　　处：《水力发电》2017年第5期76-80,共5页Water Power

基　　金：国家自然科学基金资助项目(51409152)

摘　　要：在处理具有复杂约束条件的水火电系统短期优化调度问题时,针对差分进化算法(DE)变异率控制难点和"早熟"问题,引入自适应变异算子,以距离方差校验,借正态分布生成种群,提出改进差分进化算法(EDE)。为使可行解在约束范围内收敛,运用双适应度函数对求解过程评价。以国际上通用4个水电站和3个火电站组成的系统仿真分析,EDE使水火电总燃料费用最小,为水火电力系统求解提供了有效手段。In view of the complexity and difficulty to decide the mutation rate in high-dimensional problems such as short-term dispatch of hydrothermal power systems and premature convergence problem in Differential Evolution （DE）, an Enhanced Differential Evolution （EDE） algorithm is proposed. In EDE, a self-adaption mutation operator is adopted. Thereafter, the distance variance is used to check the premature convergence, and new population is generated by normal distribution. A deputy fitness function is constructed in order to make the feasible solution be within the constraints of convergence. From the analysis of simulated system consisting of four hydropower stations and three thermal power stations, it can be seen that the EDE gives a minimum cost of system and provides an effective method for the optimization of hydrothermal power systems.

关 键 词：改进差分进化算法 梯级水电站 联合优化调度 水火电系统 

分 类 号：TV697.1[水利工程—水利水电工程]