Distributed Wind Power Forecasting Method Based on Frequency Domain Decomposition and Precision-weighted Ensemble

doi:10.12204/j.issn.1000-7229.2023.05.009

Abstract

Abstract:

Research on high-precision forecasting for distributed small wind power with strong random fluctuations is of great significance for enhancing the stability of small wind turbines and providing reliable support for distribution. Therefore, a distributed wind power forecasting method based on frequency domain decomposition and a precision-weighted ensemble was proposed. First, the original wind power signal was decomposed into different frequency bands through complete ensemble empirical mode decomposition with adaptive noise to capture the local fluctuation characteristics of wind power. Then, a precision-weighted Stacking model with two-layer heterogeneous learners was constructed to fully utilize the performance advantages of different learners and increase the generalization ability to deal with strong fluctuations. Finally, the model was verified on actual dataset from four wind turbines. It was observed that the proposed method is superior to several advanced prediction methods currently used in large wind farms, wind farm clusters, and single wind turbines, which proves the validity and generalization of the proposed method for distributed wind power forecasting.

Key words: distributed wind power, power forecasting, complete ensemble empirical mode decomposition with adaptive noise, precision weighting, ensemble learning

CLC Number:

TM614

WANG Shaomin, WANG Shouxiang, ZHAO Qianyu, DONG Yichao. Distributed Wind Power Forecasting Method Based on Frequency Domain Decomposition and Precision-weighted Ensemble[J]. ELECTRIC POWER CONSTRUCTION, 2023, 44(5): 84-93.

Figures/Tables 10

Fig.1 Framework diagram of CEEMDAN-PWS model

Table 1 The pseudocode of PWSj building process

Fig.2 Structure diagram of AM-LSTM model

Table 2 Comparison of NRMSE for heterogeneous learners

Fig.3 Error distribution of different learners for four wind turbines

Table 3 Composition of comparison models

Table 4 Error evaluation indexes comparison of different models

Fig.4 Multi-time scale error distribution of different models

Table A1 Comparison of TIC and R2 for heterogeneous learners

Fig.A1 Multi-time scale error distribution of different models for WT2, WT3 and WT4

References 32

[1]	吴问足, 乔颖, 鲁宗相, 等. 风电功率概率预测方法及展望[J]. 电力系统自动化, 2017, 41(18): 167-175.
	WU Wenzu, QIAO Ying, LU Zongxiang, et al. Methods and prospects for probabilistic forecasting of wind power[J]. Automation of Electric Power Systems, 2017, 41(18): 167-175.
[2]	崔杨, 穆钢, 刘玉, 等. 风电功率波动的时空分布特性[J]. 电网技术, 2011, 35(2): 110-114.
	CUI Yang, MU Gang, LIU Yu, et al. Spatiotemporal distribution characteristic of wind power fluctuation[J]. Power System Technology, 2011, 35(2): 110-114.
[3]	康重庆, 姚良忠. 高比例可再生能源电力系统的关键科学问题与理论研究框架[J]. 电力系统自动化, 2017, 41(9): 2-11.
	KANG Chongqing, YAO Liangzhong. Key scientific issues and theoretical research framework for power systems with high proportion of renewable energy[J]. Automation of Electric Power Systems, 2017, 41(9): 2-11.
[4]	卿湘运, 杨富文, 王行愚. 采用贝叶斯-克里金-卡尔曼模型的多风电场风速短期预测[J]. 中国电机工程学报, 2012, 32(35): 107-114.
	QING Xiangyun, YANG Fuwen, WANG Xingyu. Short-term wind speed forecasting for multiple wind farms using Bayesian krigedkalman mode[J]. Proceedings of the CSEE, 2012, 32(35): 107-114.
[5]	BOTTERUD A, ZHOU Z, WANG J H, et al. Demand dispatch and probabilistic wind power forecasting in unit commitment and economic dispatch: a case study of Illinois[J]. IEEE Transactions on Sustainable Energy, 2013, 4(1): 250-261. doi: 10.1109/TSTE.2012.2215631 URL
[6]	FOLEY A M, LEAHY P G, MARVUGLIA A, et al. Current methods and advances in forecasting of wind power generation[J]. Renewable Energy, 2012, 37(1): 1-8. doi: 10.1016/j.renene.2011.05.033 URL
[7]	LIU Guangbiao, ZHOU Jianzhong, JIA Benjun, et al. Advance short-term wind energy quality assessment based on instantaneous standard deviation and variogram of wind speed by a hybrid method[J]. Applied Energy, 2019, 238: 643-667. doi: 10.1016/j.apenergy.2019.01.105
[8]	梁超, 刘永前, 周家慷, 等. 基于卷积循环神经网络的风电场内多点位风速预测方法[J]. 电网技术, 2021, 45(2): 534-542.
	LIANG Chao, LIU Yongqian, ZHOU Jiakang, et al. Wind speed prediction at multi-locations based on combination of recurrent and convolutional neural networks[J]. Power System Technology, 2021, 45(2): 534-542.
[9]	HU Yalan, CHEN Liang. A nonlinear hybrid wind speed forecasting model using LSTM network, hysteretic ELM and differential evolution algorithm[J]. Energy Conversion and Management, 2018, 173: 123-142. doi: 10.1016/j.enconman.2018.07.070 URL
[10]	史加荣, 赵丹梦, 王琳华, 等. 基于RR-VMD-LSTM的短期风电功率预测[J]. 电力系统保护与控制, 2021, 49(21):63-70.
	SHI Jiarong, ZHAO Danmeng, WANG Linhua, et al. Short-term wind power prediction based on RR-VMD-LSTM[J]. Power System Protection and Control, 2021, 49(21):63-70
[11]	刘铨. 基于集成学习的风电功率预测系统设计与实现[D]. 武汉: 中南财经政法大学, 2021.
	LIU Quan. Design and implementation of wind power prediction system based on integrated learning[D]. Wuhan: Zhongnan University of Economics and Law, 2021.
[12]	葛滨. 基于集成学习算法的风机有功功率预测及风电机组优化调度研究[D]. 天津: 河北工业大学, 2020.
	GE Bin. Active power prediction and optimal scheduling of wind turbine based on integrated learning algorithm[D]. Tianjin:Hebei University of Technology, 2020.
[13]	WANG Shuai, LI Bin, LI Guanzheng, et al. Short-term wind power prediction based on multidimensional data cleaning and feature reconfiguration[J]. Applied Energy, 2021, 292: 116851. doi: 10.1016/j.apenergy.2021.116851 URL
[14]	KO M S, LEE K, KIM J K, et al. Deep concatenated residual network with bidirectional LSTM for one-hour-ahead wind power forecasting[J]. IEEE Transactions on Sustainable Energy, 2020, 12(02): 1321-1335.. doi: 10.1109/TSTE.2020.3043884 URL
[15]	王渝红, 史云翔, 周旭, 等. 基于时间模式注意力机制的BiLSTM多风电机组超短期功率预测[J]. 高电压技术, 2022, 48(5): 1884-1892.
	WANG Yuhong, SHI Yunxiang, ZHOU Xu, et al. Ultra-short-term power prediction for BiLSTM multi wind turbines based on temporal pattern attention[J]. High Voltage Engineering, 2022, 48(5): 1884-1892.
[16]	KUANG H H, GUO Q, LI S Q, et al. Short-term wind power forecasting model based on multi-feature extraction and CNN-LSTM[J]. IOP Conference Series: Earth and Environmental Science, 2021, 702(1): 012019. doi: 10.1088/1755-1315/702/1/012019
[17]	MA Z J, MEI G. A hybrid attention-based deep learning approach for wind power prediction[J]. Applied Energy, 2022, 323: 119608. doi: 10.1016/j.apenergy.2022.119608 URL
[18]	郭子梦. 风电场集群短期风功率预测方法研究[D]. 包头: 内蒙古科技大学, 2020.
	GUO Zimeng. Research on short-term wind power prediction method of wind farm cluster[D]. Baotou: Inner Mongolia University of Science & Technology, 2020.
[19]	DA SILVA R G, et al.DAL MOLIN RIBEIRO M H, MORENO S R, A novel decomposition-ensemble learning framework for multi-step ahead wind energy forecasting[J]. Energy, 2021, 216: 119174. doi: 10.1016/j.energy.2020.119174 URL
[20]	高金兰, 李豪, 段玉波, 等. 基于Stacking多GRU模型的风电场短期功率预测[J]. 吉林大学学报(信息科学版), 2020, 38(4): 482-490.
	GAO Jinlan, LI Hao, DUAN Yubo, et al. Wind farm short-term power prediction based on stacking and fusion of multiple GRU models[J]. Journal of Jilin University (Information Science Edition), 2020, 38(4): 482-490.
[21]	杨帆. 基于深度学习的短期风电功率预测技术研究[D]. 武汉: 华中科技大学, 2020.
	YANG Fan. Research on short-term wind power prediction based on deep learning methodology[D]. Wuhan: Huazhong University of Science and Technology, 2020.
[22]	LIANG T, XIE G F, FAN S R, et al. A combined model based on CEEMDAN, permutation entropy, gated recurrent unit network, and an improved bat algorithm for wind speed forecasting[J]. IEEE Access, 2020, 8: 165612-165630. doi: 10.1109/Access.6287639 URL
[23]	周志华. 集成学习:基础与算法[M]. 北京: 电子工业出版社, 2020.
	ZHOU Zhihua. Ensemble methods:foundations and algorithms[M]. Beijing: Publishing House of Electronics Industry, 2020.
[24]	MA S H. A hybrid deep meta-ensemble networks with application in electric utility industry load forecasting[J]. Information Sciences, 2021, 544: 183-196. doi: 10.1016/j.ins.2020.07.054 URL
[25]	王轲, 钟海旺, 余南鹏, 等. 基于seq2seq和Attention机制的居民用户非侵入式负荷分解[J]. 中国电机工程学报, 2019, 39(1): 75-83.
	WANG Ke, ZHONG Haiwang, YU Nanpeng, et al. Nonintrusive load monitoring based on sequence-to-sequence model with attention mechanism[J]. Proceedings of the CSEE, 2019, 39(1): 75-83.
[26]	彭文, 王金睿, 尹山青. 电力市场中基于Attention-LSTM的短期负荷预测模型[J]. 电网技术, 2019, 43(5): 1745-1751.
	PENG Wen, WANG Jinrui, YIN Shanqing. Short-term load forecasting model based on attention-LSTM in electricity market[J]. Power System Technology, 2019, 43(5): 1745-1751.
[27]	WANG Shouxiang, WANG Xuan, WANG Shaomin, et al. Bi-directional long short-term memory method based on attention mechanism and rolling update for short-term load forecasting[J]. International Journal of Electrical Power & Energy Systems, 2019, 109: 470-479. doi: 10.1016/j.ijepes.2019.02.022 URL
[28]	WANG Shaomin, WANG Shouxiang, CHEN Haiwen, et al. Multi-energy load forecasting for regional integrated energy systems considering temporal dynamic and coupling characteristics[J]. Energy, 2020, 195: 116964. doi: 10.1016/j.energy.2020.116964 URL
[29]	VAZ J, PINHO JT, MESQUITA A. An extension of BEM method applied to horizontal-axis wind turbine design[J]. Renewable Energy, 2011, 36(6): 1734-1740. doi: 10.1016/j.renene.2010.11.018 URL
[30]	殷明慧, 顾伟, 陈载宇, 等. 面向风电机组有功功率控制的风轮变速运行模式比较研究[J]. 中国电机工程学报, 2022, 42(12): 4419-4430.
	YIN Minghui, GU Wei, CHEN Zaiyu, et al. Comparative study of rotor speed variation modes for active power control of wind turbine generators[J]. Proceedings of the CSEE, 2022, 42(12): 4419-4430.
[31]	LIU J H, LIU J Y, CHEN H X, et al. Abnormal energy identification of variable refrigerant flow air-conditioning systems based on data mining techniques[J]. Applied Thermal Engineering, 2019, 150: 398-411. doi: 10.1016/j.applthermaleng.2018.12.133 URL
[32]	MASHUD R, SUBBU S, RAHMAT H, et al. Solar thermal generation forecast via deep learning and application to buildings cooling system control[J]. Renewable Energy, 2022, 196: 694-706. doi: 10.1016/j.renene.2022.07.005 URL