相似历史数据段高效查找方法研究  

Research on Effective Method of Similar Historical Data Segments Searching

作　　者：庞向坤 高嵩 张绪辉 颜庆 PANG Xiangkun;GAO Song;ZHANG Xuhui;YAN Qing(Shandong Electric Power Research Institute for State Grid Corporation of China,Ji'nan 250002)

机构地区：[1]国网山东省电力公司电力科学研究院,济南250002

出　　处：《计算机与数字工程》2022年第3期679-683,共5页Computer & Digital Engineering

摘　　要：随着新能源发电规模不断扩大,火力发电对电网稳定运行发挥的作用日益突出,因此提高火电机组安全运行水平,对电网多类型机组协调及稳定运行具有重要意义。以历史数据中相似特征数据段为基础展开的生产过程异常根源分析等技术受到广泛重视,其中提高相似特征数据段查找效率及准确性一直为研究重点。论文研究了相似趋势特征历史数据段查找问题,通过采用历史数据符号化表示转换技术,实现了数据段趋势信息和数幅值变化信息的融合,提高了相似特征数据段查找结果的效率和准确性。最后,论文通过数值仿真验证了所提方法的有效性。With the continuous increasing of renewable energies,thermal power generation units are to play increasingly important role in the stable running of power grids.Therefore,it is necessary to improve the operation level and safety of thermal power units for power grids to coordinate multiple types of power generation units.In order to determine abnormal root causes,the technology of similar historical data segments searching has been paid more and more attention.This paper studies the problem of finding the similar historical data segment by realizing the fusion of the trend information and the change information,based on the binary data representation and transformation technology of historical data.It is improved that the binary data representation is beneficial for the efficiency of similar data segments searching.Finally,the effectiveness of the proposed method is verified by numerical simulation.

关 键 词：相似数据段 数据转换 趋势 幅值 异常根源分析 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]