发现不易受扰动影响的满意解——鲁棒性评估  

Robust evaluation to find the satisfied solutions with less disturbance

作　　者：刘健 韩哲 张志华 王毅钊 LIU Jian;HAN Zhe;ZHANG Zhihua;WANG Yizhao(State Grid Shaanxi Electric Power Research Institute,Xi’an 710100,China;State Grid Binzhou Power Supply Company,Binzhou 256600,China)

机构地区：[1]国网陕西省电力公司电力科学研究院,陕西西安710100 [2]国网山东省电力公司滨州供电公司,山东滨州256600

出　　处：《供用电》2022年第7期46-49,92,共5页Distribution & Utilization

基　　金：国家自然科学基金重点支持项目“城市智能配电网保护与自愈控制关键技术”资助(U1766208)。

摘　　要：工程实际的不确定性可以通过优化算法的满意解获得足够的抗扰动能力。为了发现不易受扰动影响的满意解,定义了一种能够反映抗扰动能力的鲁棒性指标,并论述了单目标优化问题和多目标优化问题的鲁棒性评估方法,提出了一种基于鲁棒空间等比例缩放的鲁棒性排序比较方法;以一个含有6台发电机组系统的环境和经济调度双目标优化问题为例,对鲁棒性评估过程进行了阐述。结果表明,采用所建议的鲁棒性评估方法能够从众多满意解中遴选出鲁棒性最好的解。The uncertainty of engineering practice requires that the satisfied solutions obtained by the optimization algorithm has sufficient anti-disturbance ability.In order to find the satisfied solutions with less disturbance, a robust index to indicate the performance of anti-disturbance is defined.The robust evaluation approaches for single-objective optimization and multi-objective optimization are described respectively, a robustness ranking comparison method based on equal scaling of robust space is proposed.A double objective optimization problem of environmental economic dispatching with six generators power system is used as an example to illustrate the application of the robust evaluation.The results show that the proposed robustness evaluation method can select the solution with the best robustness from many satisfactory solutions.

关 键 词：精英优化 鲁棒性评估 满意解 单目标优化 多目标优化 

分 类 号：TM72[电气工程—电力系统及自动化]