放弃全局最优去寻找满意解--精英优化法  

An elitist searching approach to find the satisfied solutions

作　　者：刘健 赵磐 张志华 王毅钊 LIU Jian;ZHAO Pan;ZHANG Zhihua;WANG Yizhao(State Grid Shaanxi Electric Power Research Institute,Xi’an 710100,China;China Coal Xi'an Design Engineering Co.,Ltd.,Xi’an 710054,China)

机构地区：[1]国网陕西省电力有限公司电力科学研究院,陕西西安710100 [2]中煤西安设计工程有限公司,陕西西安710054

出　　处：《供用电》2022年第6期1-5,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：为了以简单的寻优方法获得满足工程应用需要的满意解,提出一种基于有限次随机抽样的精英优化法。推导了能够以期望的可信度获得性能排序处于前列的满意解所需的最少抽样次数,论述了采用精英优化法解决多目标优化问题的关键技术,提出了多目标条件下候选解的性能排序方法,指出精英优化法是一种通用方法,且便于程序实现。以IEEE 30节点系统为例论述了精英优化法在电力系统无功优化问题中的应用,并与采用其他优化方法获得的结果进行了比较,结果表明精英优化法是可行的。In order to obtain the satisfied solution to meet the needs of engineering applications in a simplified way,an elitist searching approach based on finite-times random sampling is put forward.The minimum sampling times to obtain the satisfied solution of performance ranking at the forefront with the desired credibility is deduced.The way to solve the multi-objective optimization problems by the elitist searching approach is described.The ranking approach for candidate solutions in the condition of multi-objectives is discussed.It is pointed out that the elitist searching approach is an universal tool and easy for programming.The IEEE-30 system is used as an example to illustrate the application of the elitist searching approach in reactive power optimization,the results of which are compared with the other methods showing that the proposed approach is feasible.

关 键 词：优化方法 随机抽样 满意解 多目标优化 无功优化 

分 类 号：TM72[电气工程—电力系统及自动化]