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

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
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

摘要(Abstract)：

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(KeyWords)：常规机组；爬坡性能；密度峰值挖掘；线性回归分析
conventional power units；climbing performance；density peak mining；linear regression analysis
基金项目(Foundation):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）
National Natural Science Foundation of China（62273214）；Independent Research and Development Project of State Grid Shandong Electric Power Research Institute（520626230084）

作者(Author): 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

DOI: 10.20097/j.cnki.issn1007-9904.2024.12.007
收稿日期(Received): 2023-09-26; 修回日期(Revised): 2024-03-20

参考文献(References)：

[1]中国电力企业联合会.中国电力行业年度发展报告2023[EB/OL].(2023-07-07)[2023-09-20].https：//www.cec.org.cn/detail/index.html?3-322625.
[2]VEERS P,DYKES K,LANTZ E,et al.Grand challenges in the science of wind energy[J].Science,2019,366(6464):443-452.
[3]国家能源局.电力辅助服务管理办法[EB/OL].(2021-12-21)[2023-09-20].http：//zfxxgk.nea.gov.cn/2021-12/21/c_1310391161.htm.
[4] HASAN F,KARGARIAN A. Topology-aware learning assisted branch and ramp constraints screening for dynamic economic dispatch[J].IEEE Transactions on Power Systems,2022,37(5):3495-3505.
[5]何成明，杨金刚，王洪涛，等.应对风电功率爬坡事件备用需求分析和预防控制[J].电力系统保护与控制，2017,45(7):51-57.

LEI C,BU S Q,WANG Q G,et al.Look-ahead rolling economic dispatch approach for wind-thermal-bundled power system considering dynamic ramping and flexible load transfer strategy[J].IEEE Transactions on Power Systems,2024,39(1):186-202.
[6]HE Chengming,YANG Jingang,WANG Hongtao,et al. Reserve demand analysis and preventive control strategy to deal with wind power ramp[J].Power System Protection and Control,2017,45(7):51-57.
[7]李明节，陈国平，董存，等.新能源电力系统电力电量平衡问题研究[J].电网技术，2019,43(11):3979-3986.

LI Mingjie,CHEN Guoping,DONG Cun,et al.Research on power balance of high proportion renewable energy system[J]. Power System Technology,2019,43(11):3979-3986.
[8]王浩元，别朝红.考虑不确定性物理边界的灵活爬坡备用分布鲁棒经济调度[J/OL].电力自动化设备：1-11[2023-09-20].https：//doi.org/10.16081/j.epae.202308032.

WANG Haoyuan,BIE Zhaohong.Distributionally robust economic dispatch of flexible ramping reserve considering physical boundaries of uncertainty[J/OL].Electric Power Automation Equipment:1-11[2023-09-20]. https：//doi. org/10.16081/j.epae.202308032.
[9]杨龙杰，周念成，胡博，等.计及火电阶梯式爬坡率的耦合系统优化调度方法[J].中国电机工程学报，2022,42(1):153-164.

YANG Longjie,ZHOU Niancheng,HU Bo,et al. Optimal scheduling method for coupled system based on ladder-type ramp rate of thermal power units[J].Proceedings of the CSEE,2022,42(1):153-164.
[10]朱西平，罗健，李姿霖，等.考虑灵活爬坡产品的能源枢纽低碳经济调度[J].电力自动化设备，2023,43(1):9-15.

ZHU Xiping,LUO Jian,LI Zilin,et al. Low-carbon economic dispatching of energy hub considering flexible ramping product[J].Electric Power Automation Equipment,2023,43(1):9-15.
[11]马洪艳，贠靖洋，严正.基于分布鲁棒优化的灵活爬坡备用调度方法[J].中国电机工程学报，2020,40(19):6121-6132.

MA Hongyan,YUN Jingyang,YAN Zheng.Distributionally robust optimization based dispatch methodology of flexible ramping products[J].Proceedings of the CSEE,2020,40(19):6121-6132.
[12]亢丽君，王蓓蓓，薛必克，等.计及爬坡场景覆盖的高比例新能源电网平衡策略研究[J].电工技术学报，2022,37(13):3275-3288.

KANG Lijun,WANG Beibei,XUE Bike,et al. Research on the balance strategy for power grid with high proportion renewable energy considering the ramping scenario coverage[J].Transactions of China Electrotechnical Society,2022,37(13):3275-3288.
[13]高嵩，路宽，张琳，等.基于火电—风电机组调节速率的电网AGC指令分配方法[J].山东电力技术，2022,49(2):41-46.

GAO Song,LU Kuan,ZHANG Lin,et al.AGC command allocation method based on ramp rates of thermal power generating units and wind turbines[J].Shandong Electric Power,2022,49(2):41-46.
[14]李军，刘恩仁，王敏，等.基于时间序列分段的风电机组调节速率估计[J].科学技术与工程，2021,21(27):11609-11614.

LI Jun,LIU Enren,WANG Min,et al.Ramp rate estimation of wind power generation units based on time series segmentation[J].Science Technology and Engineering,2021,21(27):11609-11614.
[15]王康，张青蕾，王泽，等.高比例风电系统的爬坡备用需求评估[J].电网与清洁能源，2022,38(8):94-101.

WANG Kang,ZHANG Qinglei,WANG Ze,et al. Evaluation of ramping reserve requirement for high-proportion wind power systems[J].Power System and Clean Energy,2022,38(8):94-101.
[16]卢楠滟.风电功率爬坡事件识别技术研究[D].北京：华北电力大学，2022.
[17] LOBATO E,EGIDO I,ROUCO L.Monitoring frequency control in the Turkish power system[J]. Electric Power Systems Research,2012,84(1):144-151.
[18] BONDY D E M,THAVLOV A,TOUGAARD J B M,et al.Performance requirements modeling and assessment for active power ancillary services[C]//2017 IEEE Manchester PowerTech.IEEE,2017:1-6..
[19] ZHANG X S,XU Z,YU T,et al. Optimal mileage based AGC dispatch of a GenCo[J]. IEEE Transactions on Power Systems,2020,35(4):2516-2526.
[20] VIJAYSHANKAR S,STANFEL P,KING J,et al. Deep reinforcement learning for automatic generation control of wind farms[C]//2021 American Control Conference(ACC). IEEE,2021:1796-1802.
[21]赵俊杰，冯树臣，孙同敏，等.基于ICS和弹性运行控制技术在燃煤智慧电厂AGC优化应用[J].能源科技，2020,18(3):1-6.

ZHAO Junjie,FENG Shuchen,SUN Tongmin,et al. AGC optimization application of coal-fired intelligent power plants based on ICS and elastic operation control technology[J].Energy Science and Technology,,2020,18(3):1-6.
[22]沈乾坤，吴恒运，杨涛，等.超临界350 MW循环流化床机组模拟量控制系统及AGC优化[J].热力发电，2020,49(5):126-131.

SHEN Qiankun,WU Hengyun,YANG Tao,et al. Optimization of analog control system and AGC for a supercritical 350 MW circulating fluidized bed unit[J].Thermal Power Generation,2020,49(5):126-131.
[23]赵源筱，耿光超，江全元，等.考虑功率变化速率的储能辅助单机调频控制策略[J].电力自动化设备，2020,40(1):141-147.

ZHAO Yuanxiao,GENG Guangchao,JIANG Quanyuan,et al.Frequency control strategy of single-generator supporting by energy storage considering power change rate[J].Electric Power Automation Equipment,2020,40(1):141-147.
[24]尚洪奎，赵博，朱志军，等.考虑电网运行备用的火电机组负荷快速调节研究与实施[J].工程技术研究，2020,2(9):31-32.

SHANG Hongkui,ZHAO Bo,ZHU Zhijun,et al. Research and implementation of fast load regulation of thermal power units considering standby operation of power grid[J]. Engineering and Technological Research,2020,2(9):31-32.
[25] XU J W,WANG J D,IZADI I,et al.Performance assessment and design for univariate alarm systems based on FAR,MAR,and AAD[J]. IEEE Transactions on Automation Science and Engineering,2012,9(2):296-307.
[26]CHATTERJEE S,HADI A S. Regression Analysis by Example[M].5th ed.New York:Wiley,2012.
[27]RODRIGUEZ A,LAIO A. Clustering by fast search and find of density peaks[J].Science,2014,344(6191):1492-1496.