基于时间序列数据挖掘的常规机组爬坡性能估计  

The Climbing Performance Estimation of Conventional Power Units Based on Time Series Data Mining

作　　者：刘恩仁 高嵩 白卓 杨子江 李乐蒙 李军 LIU Enren;GAO Song;BAI Zhuo;YANG Zijiang;LI Lemeng;LI Jun(State Grid Shandong Electric Power Research Institute,Jinan 250003,China;State Grid Tianjin Electric Power Company Chengnan Power Supply Branch,Tianjin 300201,China;College of Electrical Engineering and Automation,Shandong University of Science and Technology,Qingdao 266590,China;State Grid Shandong Electric Power Construction Company,Jinan 250000,China)

机构地区：[1]国网山东省电力公司电力科学研究院,山东济南250003 [2]国网天津市电力公司城南供电分公司,天津300201 [3]山东科技大学电气与自动化工程学院,山东青岛266590 [4]国网山东省电力公司建设公司,山东济南250000

出　　处：《山东电力技术》2024年第12期53-60,共8页Shandong Electric Power

基　　金：国家自然科学基金项目(62273214);国网山东省电力公司电力科学研究院自主研发项目(520626230084)。

摘　　要：随着新能源发电并网规模以及电网负荷峰谷差的不断扩大,电网电力平衡难度日益增加。火电等常规机组的爬坡性能是电网调节机组出力、维护电力平衡的重要参数。提出一种基于时间序列数据挖掘的常规机组爬坡性能估计方法。首先,利用均值变化检测方法确定机组自动发电控制(automatic generation control,AGC)指令中的同向大范围连续调节数据段,并得到对应时间段的机组实发功率数据段;其次,在对机组实发功率数据段进行分段线性表示的基础上,采用加权平均的方式得到机组在该时段的爬坡性能值;最后,通过对由多组爬坡性能值和AGC指令变化值构成的二维样本集合进行密度峰值挖掘及线性回归分析,得到机组的爬坡性能估计值及置信区间。所提方法能有效避免机组启停等非正常运行状态造成的爬坡性能估计误差,对电网利用常规机组的爬坡性能维护电力平衡具有重要意义。With the continuous expansion of renewable energy generation and the peak-valley differences of power grids,it is difficult to keep power balance in power grids.The climbing performance of conventional generation units is an important parameter,and it is used to regulate unit output and keep power balance.A climbing performance estimation method of conventional generation units is proposed based on time series data mining technique.Firstly,a step-change detection method is adopted to determine the co-directional,large scale changed,continuous adjustment data segment in the automatic generation control(AGC)instructions of the units,and the actual power data segment is obtained in the corresponding time period;Secondly,based on the piece-wise linear representation of the actual power data segments,the climbing performance value is calculated through a weighted average method during the period;Finally,by performing density peak mining and linear regression analysis on the two-dimensional sample set which is composed by multiple sets of climbing performance values and AGC instruction values,the estimated climbing performance and confidence intervals can be obtained.The proposed method can avoid the estimation error of climbing performance caused by the startup and shutdown of the units,as well as other abnormal conditions,and it is of great significance for power grids to keep power balance by utilizing the climbing performance of conventional units.

关 键 词：常规机组 爬坡性能 密度峰值挖掘 线性回归分析 

分 类 号：TM614[电气工程—电力系统及自动化]