Non-smooth/non-convex economic dispatch by a novel hybrid differential evolution algorithm
Non-smooth/non-convex economic dispatch by a novel hybrid differential evolution algorithm
复制标题
DOI：
10.1049/iet-gtd:20070183
复制
发表时间：
2007-08
影响因子：
2.5
通讯作者：
Sheng-Kuan Wang;J. Chiou;Chih-Wen Liu
中科院分区：
文献类型：
--
作者：
Sheng-Kuan Wang;J. Chiou;Chih-Wen Liu
本文提出了一种新的随机优化方法来确定考虑各种发电机约束的经济调度问题的可行最优解。考虑了发电机的许多实际约束条件，如斜坡速率限制、禁止操作区域和阀点效应。这些约束使得ED问题成为一个带有约束的非光滑/非凸最小化问题。提出的优化算法称为自调谐混合差分进化（self-tuning hybrid differential evolution，自调谐HDE）。自调优HDE利用了原HDE中1/5进化策略成功规则的概念，加速了对全局最优的搜索。应用3、13、40单元电力系统进行测试，比较了该算法与遗传算法、差分进化算法和HDE算法的性能。数值结果表明，本文提出的自调优HDE算法整体性能优于其他三种算法。
This paper presents a novel stochastic optimisation approach to determining the feasible optimal solution of the economic dispatch (ED) problem considering various generator constraints. Many practical constraints of generators, such as ramp rate limits, prohibited operating zones and the valve point effect, are considered. These constraints make the ED problem a non-smooth/non-convex minimisation problem with constraints. The proposed optimisation algorithm is called self-tuning hybrid differential evolution (self-tuning HDE). The self-tuning HDE utilises the concept of the 1/5 success rule of evolution strategies (ESs) in the original HDE to accelerate the search for the global optimum. Three test power systems, including 3-, 13- and 40-unit power systems, are applied to compare the performance of the proposed algorithm with genetic algorithms, the differential evolution algorithm and the HDE algorithm. Numerical results indicate that the entire performance of the proposed self-tuning HDE algorithm outperforms the other three algorithms.