基于EMD-RVM的风电场机组分组功率预测

张晋华; 冯源; 黄远为; 阎洁

doi:10.16513/j.2096-2185.DE.2106026

PDF(3055 KB)

分布式能源 ›› 2021, Vol. 6 ›› Issue (2) : 22-31. DOI: 10.16513/j.2096-2185.DE.2106026

学术研究

基于EMD-RVM的风电场机组分组功率预测

张晋华 ¹^,² ,
冯源 ¹ ,
黄远为 ¹ ,
阎洁 ²^,³

作者信息 +

Power Prediction for Groups of Wind Farm Units Based on EMD-RVM

Jinhua ZHANG ¹^,² ,
Yuan FENG ¹ ,
Yuanwei HUANG ¹ ,
Jie YAN ²^,³

Author information +

文章历史 +

摘要

通过识别不同风电机组的相似性将风电场内各机组区分为不同的机组群，对每组机群分别建立功率预测模型，既提高了计算效率，又改善了大型风电场短期功率预测精度，较好解决了风电波动性和间歇性对电力系统安全运行的影响。提出了一种基于戴维森堡丁指数与聚类算法的风电场机组分组功率预测方法，以实际测量风速，测量功率以及两者的组合作为机组分组模型输入，分析其对聚类精度的影响程度。首先，采用经验模态分解方法将风电功率序列分解后，将与原信号具有高相关性的固有模态函数(intrinsic mode function，IMF)分量重构，作为K-means聚类算法的输入重新进行场内机组分组。然后，对每组机群分别构建经验模态分解－相关向量机(empirical mode decomposition-relevance vector machine，EMD-RVM)模型。最后，将预测功率分量叠加获得总功率的预测值。仿真实验结果表明，风速是影响聚类结果的主要因素，功率可作为不同聚类模型输入变量的重要补充；基于EMD-RVM的分组功率预测提升了预测精度，提高了预测效率。

Abstract

By identifying the similarity of different wind turbines in the wind farm unit of each unit is divided into different groups, each group of fleet power prediction model is set up, respectively, can not only improve the computing efficiency, improved at the same time a large wind farm short-term power prediction precision, better solved the volatility and intermittent wind effects on safe operation of power system. A wind farm grouping power prediction method based on Davidson-Boding index and clustering algorithm was proposed. The actual measured wind speed, measured power and the combination of the two were taken as the input of the grouping model, and the influence degree on the clustering accuracy was analyzed. After the wind power sequence is decomposed by the empirical mode decomposition method, the intrinsic mode function (IMF), which is highly correlated with the original signal, is reconstructed and used as the input of the k-means clustering algorithm to regroup the units in the field. Empirical Mode Decomposition and Relevance Vector Machine (EMD-RVM) models were constructed for each cluster, and the predicted power components were superimposed to obtain the predicted value of total power. The simulation results show that the wind speed is the main factor affecting the clustering results, and the power can be used as an important supplement to the input variables of different clustering models. Packet power prediction based on EMD-RVM improves the prediction accuracy and efficiency.

导出引用

张晋华, 冯源, 黄远为, 等. 基于EMD-RVM的风电场机组分组功率预测[J]. 分布式能源. 2021, 6(2): 22-31 https://doi.org/10.16513/j.2096-2185.DE.2106026

Jinhua ZHANG, Yuan FENG, Yuanwei HUANG, et al. Power Prediction for Groups of Wind Farm Units Based on EMD-RVM[J]. Distributed Energy Resources. 2021, 6(2): 22-31 https://doi.org/10.16513/j.2096-2185.DE.2106026

中图分类号： TK81

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	钱政，裴岩，曹利宵，等. 风电功率预测方法综述[J]. 高电压技术，2016, 42(4): 1047-1060. QIAN Zheng, PEI Yan, CAO Lixiao, et al. Review of wind power forecasting method[J]. High Voltage Engineering, 2016, 42(4): 1047-1060. 本文引用 [1]

[2]	刘永前，朴金姬，韩爽. 风电场输出功率预测中两种神经网络算法的研究[J]. 现代电力，2011, 28(2): 49-52. LIU Yongqian, PIAO Jinji, HAN Shuang. Study on two neural network algorithms to predict wind power[J]. Modern Electric Power, 2011, 28(2): 49-52. 本文引用 [1]

[3]	曹娜，于群. 风速波动情况下并网风电场内风电机组分组方法[J]. 电力系统自动化，2012, 36(2): 42-46. CAO Na, YU Qun. A grouping method for wind turbines in a grid connected wind farm during wind speed fluctuation[J]. Automation of Electric Power Systems, 2012, 36(2): 42-46. 本文引用 [1]

[4]	BIGDELI N, AFSHAR K, GAZAFROUDI A S, et al. A comparative study of optimal hybrid methods for wind power prediction in wind farm of Alberta, Canada[J]. Renewable and Sustainable Energy Reviews, 2013, 27: 20-29. 本文引用 [1]

[5]	叶林，赵永宁. 基于空间相关性的风电功率预测研究综述[J]. 电力系统自动化，2014, 38(14): 126-135. YE Lin, ZHAO Yongning. A review on wind power prediction based on spatial correlation approach[J]. Automation of Electric Power Systems, 2014, 38(14): 126-135. 本文引用 [1]

[6]	HAYES B P, ILIE I S, PORPODAS A, et al. Equivalent power curve model of a wind farm based on field measurement data[C]//2011 IEEE Trondheim PowerTech. IEEE, 2011. 本文引用 [1]

[7]	ALI M, ILIE I S, MILANOVIC＇ J V, et al. Probabilistic clustering of wind generators[C]//Power & Energy Society General Meeting. IEEE, 2010. 本文引用 [1]

[8]	MA Y, JIANG J N, RUNOLFSSON T, et al. Cluster analysis of wind turbines of large wind farm[C]//IEEE Power Systems Conference and Exposition, 2009(3): 1-7. 本文引用 [1]

[9]	张晋华. 风电场内机组优化调度研究[D]. 北京：华北电力大学，2014. ZHANG Jinhua. Research on unit optimal dispatch in wind farm[D]. Beijing: North China Electric Power University, 2014. 本文引用 [1]

[10]	阎洁. 风电场功率预测不确定性分析方法及其应用研究[D]. 北京：华北电力大学，2012. YAN Jie. Research on uncertainty analysis method for wind power prediction and its application[D]. Beijing: North China Electric Power University, 2012. 本文引用 [1]

[11]	王勃，刘纯，张俊，等. 基于Monte-Carlo方法的风电功率预测不确定性估计[J]. 高电压技术，2015, 41(10): 3385-3391. WANG Bo, LIU Chun, ZHANG Jun, et al. Uncertainty evaluation of wind power prediction based on Monte-Carlo method[J]. High Voltage Engineering, 2015, 41(10): 3385-3391. 本文引用 [1]

[12]	阎洁，刘永前，张浩，等. 基于风场景识别的动态风电功率概率预测方法[J]. 现代电力，2016, 33(2): 51-58. YAN Jie, LIU Yongqian, ZHANG Hao, et al. Dynamic wind power probabilistic forecasting based on wind scenario recognition[J]. Modern Electric Power, 2016, 33(2): 51-58. 本文引用 [1]

[13]	段海滨. 蚁群算法原理及其应用[M]. 北京：科学技术出版社，2005. 本文引用 [1]

[14]	蔡晓妍，戴冠中，杨黎斌. 谱聚类算法综述[J]. 计算机科学，2008, 35(7): 14-18. CAI Xiaoyan, DAI Guanzhong, YANG Libin. Survey on spectral clustering algorithms[J]. Computer Science, 2008, 35(7): 14-18. 本文引用 [1]

[15]	DAVIES D L, BOULDIN D W. A cluster separation measure[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1979, PAMI-1(2): 224-227. 本文引用 [1]

[16]	谭晓琳. 考虑集群光伏与风电出力相关性的场景生成及概率潮流研究[D]. 济南：山东大学，2018. TAN Xiaolin. Scenario generation and probabilistic load flow considering the correlation between clustered photovoltaic and wind power output[D]. Jinan: Shandong University, 2018. 本文引用 [1]

[17]	HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings Mathematical Physical & Engineering Sciences, 1998, 454: 903-995. 本文引用 [1]

[18]	张礼平，杨志勇，陈正洪. 典型相关系数及其在短期气候预测中的应用[J]. 大气科学，2000(3): 427-432. 本文引用 [1]

[19]	张强. 相关向量机理论在风电功率实时预测中的应用[D]. 吉林：东北电力大学，2017. ZHANG Qiang. Application of realtion vector machine theory in real-time prediction of wind power[D]. Jilin: Northeast Electric Power University, 2017. 本文引用 [1]

[20]	崔瑜琦. 基于最小二乘支持向量机风电机组的预测研究[D]. 北京：华北电力大学，2010. CUI Yuqi. Research on prediction of wind turbine generation based on LS-SVM[D]. Beijing: North China Electric Power University, 2010. 本文引用 [1]

[21]	许成志. 基于云理论的短期风电功率预测及不确定性分析方法研究[D]. 北京：华北电力大学，2018. XU Chengzhi. Research on short term wind power prediction and uncertainty analysis based on cloud theory[D]. Beijing: North China Electric Power University, 2018. 本文引用 [1]

[22]	高小力. 大型风电场分组建模方法及其在功率预测中的应用[D]. 北京：华北电力大学，2015. GAO Xiaoli. Clustering methods of wind turbines in large-scale wind farm and its application in wind power forecasts[D]. Beijing: North China Electric Power University, 2015. 本文引用 [1]

[23]	李相俊，许格健. 基于长短期记忆神经网络的风力发电功率预测方法[J]. 发电技术，2019, 40(5): 426-433. LI Xiangjun, XU Gejian. Wind Power Prediction Method Based on Long Short-term Memory Neural Network[J]. Power Generation Technology, 2019, 40(5): 426-433. 本文引用 [1]