您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报 (工学版) ›› 2020, Vol. 50 ›› Issue (1): 82-94.doi: 10.6040/j.issn.1672-3961.0.2019.178

• 电气工程 • 上一篇    下一篇

风电爬坡事件的非精确条件概率预测

王勃1(),汪步惟1,杨明2,*(),赵元春3,朱文立2   

  1. 1. 中国电力科学研究院有限公司新能源与储能运行控制国家重点实验室, 北京 100192
    2. 山东大学电网智能化调度与控制教育部重点实验室, 山东 济南 250061
    3. 国网济南市供电公司, 山东 济南 250012
  • 收稿日期:2019-04-19 出版日期:2020-02-20 发布日期:2020-02-14
  • 通讯作者: 杨明 E-mail:wangbo@epri.sgcc.com.cn;myang@sdu.edu.cn
  • 作者简介:王勃(1983—),男,新疆伊犁人,硕士,高级工程师,主要研究方向为新能源发电功率预测技术. E-mail: wangbo@epri.sgcc.com.cn
  • 基金资助:
    国家电网公司科技项目(5215001600V4)

Imprecise conditional probability prediction of wind power ramp events

Bo WANG1(),Buwei WANG1,Ming YANG2,*(),Yuanchun ZHAO3,Wenli ZHU2   

  1. 1. State Key Laboratory of Operation and Control of Renewable Energy & Storage Systems, China Electric Power Research Institute, Beijing 100192, China
    2. Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education, Shandong University, Jinan 250061, Shandong, China
    3. tate Grid Jinan Power Company, Jinan 250012, Shandong, China
  • Received:2019-04-19 Online:2020-02-20 Published:2020-02-14
  • Contact: Ming YANG E-mail:wangbo@epri.sgcc.com.cn;myang@sdu.edu.cn
  • Supported by:
    国家电网公司科技项目(5215001600V4)

摘要:

风电爬坡事件(wind power ramp events, WPRE)易破坏电力系统的有功功率平衡,劣化频率稳定性及电能质量,威胁电网的安全稳定运行。由此,提出一种基于信度网络(credal network, CN)的WPRE非精确条件概率预测方法,对WPRE各状态发生概率的区间范围进行预测。运用贪婪搜索算法挖掘WPRE与多个气象变量之间的相依性关系,并搭建CN结构以抽象表达;在超参数设置方面对非精确狄利克雷模型(imprecise Dirichlet model, IDM)进行了拓展,使用拓展后的IDM对变量间的条件相依性关系进行不确定性量化,完成CN的参数估计;基于建立的CN模型,在获取气象预测信息的条件下,结合CN概率推断算法对多状态WPRE的分布进行非精确概率推断;采用宁夏某风电场的实测数据对本方法进行测试,验证了该方法在观测样本不充足的预测情景下优异的预测性能。

关键词: 非精确概率, 风电爬坡, 信度网络, 非精确狄利克雷模型

Abstract:

Wind power ramp events (WPRE) could easily destroy the active power balance of the power system, as well as did harm to the frequency stability and power quality, which threatened the safe and stable operation of power grid. A novel imprecise conditional probability prediction approach was proposed based on the credal network (CN), which could provide the interval range of the occurrence probability of each WPRE state. The approach excavated the dependence relationships between WPRE and meteorological variables using the greedy search algorithm, and constructed a CN structure to express the relationships in an abstract way. The proposed approach extended the imprecise Dirichlet model (IDM) on hyperparameter settings to quantify the uncertain conditional dependences among the variables, thus to realize the parameter estimation of the CN. Based on the constructed CN model, a CN probability inference algorithm was employed to estimate the imprecise probability distribution of the multi-state WPRE. The case study with wind-farm operating measurements in Ningxia Province demonstrated that the proposed approach had excellent performance even under the prediction scenarios with insufficient samples.

Key words: imprecise probability, wind ramp, credal network, imprecise Dirichlet model

中图分类号: 

  • TM614

图1

贪婪搜索算法搭建信度网络结构的流程"

图2

简单三节点网络结构示意图"

图3

爬坡事件非精确概率分布预测的关键步骤"

表1

爬坡状态的划分"

需满足的条件 爬坡状态
Pt_max-Pt_min>5.2 MW且t_max≥t_min 上爬坡
Pt_max-Pt_min>5.2 MW且t_max < t_min 下爬坡
|Pt_max-Pt_min|≤5.2 MW 不爬坡

表2

训练集与验证集的划分"

数据集名称 时间跨度 样本容量
训练集 2015年1月1日0:00—2015年12月31日23:45 34 516
验证集 2016年1月1日0:00—2017年12月31日23:45 66 626

表3

信度网络中变量状态的划分"

节点变量 风速最大波动量
V/(m·s-1)
风速
S/(m·s-1)
风向D/(°) 温度T/℃ 湿度h/% 爬坡事件H
(最大功率波动量)/MW
状态1 [-16, -1.77) [0, 3.09) [0, 144) [-22, 5.17) [9, 36.2) [-5.2, 5.2]
状态2 [-1.77, 1.48) [3.09, 5.41) [144, 252) [5.17, 16.77) [36.2, 59.93) (5.2, 36]
状态3 [1.48, 16.7) [5.41, 24) [252, 360) [16.77, 36.5) [59.93, 98.49) [-36, -5.2)

图4

贪婪搜索算法搭建的最佳信度网络结构"

表4

使用CWC指标评价模型预测性能"

λ/% 模型 PICP/% PINAW CWC 宽度>10%的
区间占比/%
宽度>20%的
区间占比/%
宽度>30%的
区间占比/%
宽度>50%的区间占比/%
90 CN 86.97 0.195 4 0.553 8 89.30 42.39 10.70 0.00
CLT 65.71 0.232 2 30.150 0 71.74 51.30 37.04 4.32
75 CN 73.39 0.144 4 0.343 8 73.66 17.70 3.29 0.00
CLT 57.06 0.162 4 6.029 1 62.83 38.68 18.79 0.00
60 CN 57.20 0.106 9 0.294 1 48.15 5.76 0.41 0.00
CLT 47.60 0.118 8 1.537 6 51.99 25.10 0.00 0.00

图5

7种气象条件下预测的爬坡事件非精确概率分布"

表5

使用WCWS指标评价模型的预测性能"

wt1 类型 模型 PICP/% PINAW WCWS
0.3 冒险型 CN 67.08 0.127 0 0.112 3
CLT 56.38 0.157 9 0.058 6
0.5 中立型 CN 76.82 0.157 8 0.305 2
CLT 56.38 0.157 9 0.203 0
0.7 稳健型 CN 87.65 0.198 4 0.554 1
CLT 56.38 0.157 9 0.347 3

表6

重新划分变量D和T的状态"

节点变量 风向D/(°) 温度T/℃
状态1 [0, 216) [-22, 16.77)
状态2 [216, 276) [16.77, 22.57)
状态3 [276, 360) [22.57, 36.5)

图6

重新定义变量状态后由GS算法构建的CN结构"

表7

CN模型于极端气象场景下的预测结果"

预测的非精确概率 真实概率
Pim(H1|C1)=[0.82, 0.97] P*(H1|C1)=1.00
Pim(H2|C1)=[0.01, 0.09] P*(H2|C1)=0.00
Pim(H3|C1)=[0.01, 0.10] P*(H3|C1)=0.00
Pim(H1|C2)=[0.18, 0.45] P*(H1|C2)=0.25
Pim(H2|C2)=[0.07, 0.27] P*(H2|C2)=0.25
Pim(H3|C2)=[0.38, 0.72] P*(H3|C2)=0.50
Pim(H1|C3)=[0.42, 0.72] P*(H1|C3)=0.67
Pim(H2|C3)=[0.23, 0.52] P*(H2|C3)=0.33
Pim(H3|C3)=[0.02, 0.11] P*(H3|C3)=0.00
Pim(H1|C4)=[0.41, 0.61] P*(H1|C4)=0.60
Pim(H2|C4)=[0.33, 0.53] P*(H2|C4)=0.40
Pim(H3|C4)=[0.02, 0.10] P*(H3|C4)=0.00

图7

极端气象条件下预测的爬坡事件非精确概率分布"

1 李俊峰,蔡丰波,乔黎明. 2012中国风电发展报告[R].北京:中国环境科学出版社, 2012.
LI Junfeng, CAI Fengbo, QIAO Liming. Report of China wind power outlook in 2012[R]. Beijing: China Environmental Science Press, 2012.
2 娄素华, 杨天蒙, 吴耀武, 等. 含高渗透率风电的电力系统复合储能协调优化运行[J]. 电力系统自动化, 2016, 40 (7): 30- 35.
LOU Suhua , YANG Tianmeng , WU Yaowu , et al. Coordinated optimal operation of hybrid energy storage in power system accommodated high penetration of wind power[J]. Automation of Electric Power Systems, 2016, 40 (7): 30- 35.
3 欧阳庭辉, 查晓明, 秦亮, 等. 基于小波特征提取和筛选的爬坡事件分类[J]. 中国电机工程学报, 2016, 36 (9): 2373- 2380.
OUYANG Tinghui , ZHA Xiaoming , QIN Liang , et al. Classification of wind power ramps based on screened wavelet energy characteristics[J]. Proceedings of the CSEE, 2016, 36 (9): 2373- 2380.
4 QI Yongzhi , LIU Yutian . Wind power ramping control using competitive game[J]. IEEE Transactions on Sustainable Energy, 2016, 7 (4): 1516- 1524.
doi: 10.1109/TSTE.2016.2558584
5 崔明建, 孙元章, 柯德平, 等. 考虑电网侧频率偏差的风电功率爬坡事件预测方法[J]. 电力系统自动化, 2014, 38 (5): 8- 13.
CUI Mingjian , SUN Yuanzhang , KE Deping , et al. Prediction method for wind power ramp events considering frequency deviation of power grid side[J]. Automation of Electric Power Systems, 2014, 38 (5): 8- 13.
6 CUI Mingjian , VENKAT K , BRI-MATHIAS H , et al. A copula-based conditional probabilistic forecast model for wind power ramps[J]. IEEE Transactions on Smart Grid, 2018, 1- 1.
7 张东英, 代悦, 张旭, 等. 风电爬坡事件研究综述及展望[J]. 电网技术, 2018, 42 (6): 1783- 1792.
ZHANG Dongying , DAI Yue , ZHANG Xu , et al. Review and prospect of research on wind power ramp events[J]. Power System Technology, 2018, 42 (6): 1783- 1792.
8 LIU Yongqian, SUN Ying, HAN Shuang. A WT-ARMA based method for wind power ramp events forecasting[C]//5th IET International Conference on Renewable Power Generation. London, UK: IET, 2016: 21-23.
9 NAYAK A K, SHARMA K C, BHAKAR R. ARIMA based statistical approach to predict wind power ramps[C]//2015 IEEE Power and Energy Society General Meeting. Denver, USA: IEEE, 2015: 1-4.
10 CUI M , ZHANG J , FLORITA A R , et al. An optimized swinging door algorithm for identifying wind ramping events[J]. IEEE Transactions on Sustainable Energy, 2016, 7 (1): 150- 162.
doi: 10.1109/TSTE.2015.2477244
11 ZHANG J , FLORITA A , HODGE B M , et al. Ramp forecasting performance from improved short-term wind power forecasting[J]. Energy, 2017, 122, 528- 541.
doi: 10.1016/j.energy.2017.01.104
12 熊一, 查晓明, 秦亮, 等. 风电功率爬坡气象场景分类模型及阈值整定研究[J]. 电工技术学报, 2016, 31 (19): 155- 162.
doi: 10.3969/j.issn.1000-6753.2016.19.016
XIONG Yi , ZHA Xiaoming , QIN Liang , et al. Study on wind power ramping weather scenario classification model and threshold setting[J]. Transactions of China Electrotechnical Society, 2016, 31 (19): 155- 162.
doi: 10.3969/j.issn.1000-6753.2016.19.016
13 GALLGO-CASTILLO C , CUERVA-TEJERO A , LOPEZ-GARCIA O . A review on the recent history of wind power ramp forecasting[J]. Renewable and Sustainable Energy Reviews, 2015, 52, 1148- 1157.
doi: 10.1016/j.rser.2015.07.154
14 KUSIAK A . Prediction of wind farm power ramp rates: a data-mining approach[J]. Journal of Solar Energy Engineering, 2009, 131 (3): 376- 385.
15 崔明建, 孙元章, 柯德平. 基于原子稀疏分解和BP神经网络的风电功率爬坡事件预测[J]. 电力系统自动化, 2014, 26 (12): 6- 11.
doi: 10.7500/AEPS20130418003
CUI Mingjian , SUN Yuanzhang , KE Deping . Wind power ramp events forecasting based on atomic sparse decomposition and BP neural networks[J]. Automation of Electric Power Systems, 2014, 26 (12): 6- 11.
doi: 10.7500/AEPS20130418003
16 ZAREIPOUR H, HUANG D, ROSEHART W. Wind power ramp events classification and forecasting: a data mining approach[C]//2011 IEEE Power and Energy Society General Meeting. Detroit, USA: IEEE, 2011: 1-3.
17 POTTER C W, GRIMIT E, NIJSSEN B.Potential benefits of a dedicated probabilistic rapid ramp event forecast tool[C]//2009 IEEE/PES Power Systems Conference and Exposition. Seattle, USA: IEEE, 2009: 1-3.
18 BOSSAVY A, GIRARD R, KARINIOTAKIS G.Forecasting uncertainty related to ramps of wind power production[C]//European Wind Energy Conference and Exhibition 2010. Warsaw, Poland: European Wind Energy Association, 2010: 2-9.
19 LI Y Q, PETR M, EDWARD L, et al. Temporal uncertainty of wind ramp predictions using probabilistic forecasting technique[C]//2016 IEEE Second International Conference on Big Data Computing Service and Applications (BigDataService). Oxford, UK: IEEE, 2016: 166-173.
20 BILLINTON R , KARKI R , VERMA Ajit Kumar . Reliability and risk evaluation of wind integrated power systems[M]. New Delhi, India: Springer, 2013: 29- 44.
21 BOSSAVY A , GIRARD R , KARINIOTAKIS G . Forecasting ramps of wind power production with numerical weather prediction ensembles[J]. Wind Energy, 2013, 16 (1): 51- 63.
doi: 10.1002/we.526
22 CUI M , ZHANG J , WANG Q , et al. A data-driven methodology for probabilistic wind power ramp forecasting[J]. IEEE Transactions on Smart Grid, 2019, 10 (2): 1326- 1338.
doi: 10.1109/TSG.2017.2763827
23 COZMAN F G . Credal networks[J]. Artificial Intelligence, 2000, 120 (2): 199- 233.
doi: 10.1016/S0004-3702(00)00029-1
24 FRIEDMAN N , GEIGER D , GOLDSZMIDT M . Bayesian network classifiers[J]. Machine Learning, 1997, 29 (2/3): 131- 163.
doi: 10.1023/A:1007465528199
25 BOUCHAALA L , MASMOUDI A , GARGOURI F , et al. Improving algorithms for structure learning in Bayesian networks using a new implicit score[J]. Expert Systems with Applications, 2010, 37 (7): 5470- 5475.
doi: 10.1016/j.eswa.2010.02.065
26 SAMMUT C , WEBB G . Encyclopedia of machine learning[M]. New York, USA: Springer, 2010: 483- 489.
27 CHICKERING D M , HECKERMAN D , MEEK C . Large-sample learning of Bayesian networks is NP-hard[J]. Journal of Machine Learning Research, 2004, 5, 1287- 1330.
28 WALLEY P . Inferences from multinomial data: learning about a bag of marbles[J]. Journal of the Royal Statistical Society, 1996, 58, 3- 57.
29 YANG M , WANG J H , DIAO H R . Interval estimation for conditional failure rates of transmission lines with limited samples[J]. IEEE Transactions on Smart Grid, 2018, 9 (4): 2752- 2763.
doi: 10.1109/TSG.2016.2618623
30 KOTSIANTIS S , KANELLOPOULOS D . Discretization techniques: a recent survey[J]. GESTS International Transactions on Computer Science and Engineering, 2006, 32 (1): 47- 58.
31 HECKENBERGEROVA J, MUSILEK P, JANATA M. Sensitivity analysis of PCA method for wind ramp event detection[C]//IEEE 16th International Conference on Environment and Electrical Engineering (EEEIC). Florence, Italy: IEEE, 2016: 1-3.
32 FERREIRA C , GAMA J , MATIAS L . A survey on wind power ramp forecasting[J]. Energy and Power Engineering, 2010, 5, 368- 372.
[1] 荆业飞1,张承慧1*,徐蓓蓓2,李珂1,褚晓广1. 基于阻抗匹配的小型风电系统功率输出优化方法[J]. 山东大学学报(工学版), 2013, 43(5): 39-43.
[2] 荆业飞1,徐蓓蓓2,张承慧1*,李珂1,褚晓广1. 基于模式搜索的风能最大功率跟踪控制[J]. 山东大学学报(工学版), 2013, 43(5): 44-48.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 施来顺,万忠义 . 新型甜菜碱型沥青乳化剂的合成与性能测试[J]. 山东大学学报(工学版), 2008, 38(4): 112 -115 .
[2] 孔祥臻,刘延俊,王勇,赵秀华 . 气动比例阀的死区补偿与仿真[J]. 山东大学学报(工学版), 2006, 36(1): 99 -102 .
[3] 陈瑞,李红伟,田靖. 磁极数对径向磁轴承承载力的影响[J]. 山东大学学报(工学版), 2018, 48(2): 81 -85 .
[4] 李可,刘常春,李同磊 . 一种改进的最大互信息医学图像配准算法[J]. 山东大学学报(工学版), 2006, 36(2): 107 -110 .
[5] 季涛,高旭,孙同景,薛永端,徐丙垠 . 铁路10 kV自闭/贯通线路故障行波特征分析[J]. 山东大学学报(工学版), 2006, 36(2): 111 -116 .
[6] 刘文亮,朱维红,陈涤,张泓泉. 基于雷达图像的运动目标形态检测及跟踪技术[J]. 山东大学学报(工学版), 2010, 40(3): 31 -36 .
[7] 孙国华,吴耀华,黎伟. 消费税控制策略对供应链系统绩效的影响[J]. 山东大学学报(工学版), 2009, 39(1): 63 -68 .
[8] 孙殿柱,朱昌志,李延瑞 . 散乱点云边界特征快速提取算法[J]. 山东大学学报(工学版), 2009, 39(1): 84 -86 .
[9] 王,张艳宁,申家振,刘俊成 . 基于信息测度和支持向量机的图像边缘检测[J]. 山东大学学报(工学版), 2006, 36(3): 95 -99 .
[10] 李芳佳, 高尚策, 唐政, 石井雅博, 山下和也. 基于元胞自动化模型的三维雪花晶体近似模式的产生(英文)[J]. 山东大学学报(工学版), 2009, 39(1): 102 -105 .