不良数据检测与辨识算法的评估研究  

Research in evaluation for algorithms of bad-data detection and identification

作　　者：蒋德珑[1] 王克文[1] 

机构地区：[1]郑州大学电气工程学院,郑州450001

出　　处：《计算机工程与应用》2012年第22期239-243,共5页Computer Engineering and Applications

基　　金：国家自然科学基金项目(No.50177028)

摘　　要：针对算法评价的模糊性,提出一种基于模糊数学理论的算法综合评价模型,并用于状态估计中不良数据检测与辨识算法的综合评估。通过在不同系统配置条件下,对四种常用不良数据检测与辨识算法进行综合评估,比较出其优劣特性,验证了该评估系统的正确性,实现了算法评估中定性分析的定量化,为实际系统算法的选择提供了有效参考。To solve the fuzziness of algorithm evaluation, fuzzy mathematics theory is proposed to evaluate syntheti- cally algorithms of bad-data detection and identification. And then the best algorithm of state estimation will be found for specific power system. In different system configurations, the proposed evaluation system is used to evaluate four algorithms of bad-data detection and identification, and then the characters of the four algorithms can be found. Practical effect proves that the evaluation system is effective and can quantify the qualitative-analysis of algorithm evaluation. It can give effective reference to choose algorithm for actual power system.

关 键 词：电力系统 不良数据 检测与辨识 算法评价 模糊综合评价 

分 类 号：TM712[电气工程—电力系统及自动化]