基于改进NSDE算法的含风电场的经济环境调度  

Improved NSDE Algorithm for Economic Environmental Dispatch Considering Wind Farms

作　　者：张磊 肖健梅[1] 王锡淮[1] ZHANG Lei;XIAO Jian-mei;WANG Xi-huai(Logistics Engineering College,Shanghai Maritime University,Shanghai 200135,China)

机构地区：[1]上海海事大学物流工程学院,上海200135

出　　处：《控制工程》2022年第11期1963-1973,共11页Control Engineering of China

基　　金：国家自然科学基金资助项目(61503241)。

摘　　要：随着以风能为代表的可再生能源接入电网系统,风电的随机性给原本的经济环境调度问题增加了不确定性和复杂性,采用传统方法难以获得经济可靠的解。针对风电的不可靠性,引入了高估和低估罚函数,建立考虑风电可用性的经济环境调度模型。对于传统差分进化算法求解此模型出现的收敛性和鲁棒性差的问题,提出一种改进的非支配排序差分进化算法来同时优化排放量和发电成本两个目标,对算法中的缩放因子和交叉概率进行自适应调整,增强全局搜索能力;改进精英保留策略结合基于熵的拥挤度评估方法,在有效提升收敛速度的同时维持了解集良好的分布性。将该方法在IEEE-30单元测试系统上进行了测试,实验表明,与其他算法相比,改进非支配排序差分进化算法(INSDE)算法在处理含风电场的经济环境调度(EED)问题上具有更强的寻优能力。As the renewable energy represented by wind energy is connected to the grid system, the randomness of wind power adds uncertainty and complexity to the original dispatch problem of economic environment, and it is difficult for traditional methods to obtain an economical and reliable solution. In view of the unreliability of wind power, overestimate and underestimate penalty functions are introduced, and an economic environmental dispatch model considering the availability of wind power is established in this paper.To solve the problems of poor convergence and poor robustness of traditional differential evolution algorithm,an improved non-dominated sorting differential evolution(INSDE) algorithm is proposed to optimize both emissions and power generation cost at the same time. The scaling factor and crossover probability of the algorithm are adjusted adaptively to enhance the global search ability. The improved elite retention strategy combined with the entropy-based congestion evaluation method can effectively improve the convergence speed and maintain the good distribution of the solution set. The method is tested on the IEEE 30-unit test system.The experimental results show that the INSDE algorithm has a stronger optimization ability in dealing with the economic environmental dispatch(EED) problem involving wind farms compared with other algorithms.

关 键 词：非支配排序 差分进化 风力发电 多目标优化 拥挤度 经济环境调度 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]