基于改进降噪自编码器和多元时序聚类的海上风电功率预测

周海; 刘建锋; 周健; 周勇良; 李美玉; 励晨阳

doi:10.19912/j.0254-0096.tynxb.2021-1240

PDF(1970 KB)

太阳能学报 ›› 2023, Vol. 44 ›› Issue (3) : 129-138. DOI: 10.19912/j.0254-0096.tynxb.2021-1240

基于改进降噪自编码器和多元时序聚类的海上风电功率预测

周海¹, 刘建锋¹, 周健², 周勇良³, 李美玉³, 励晨阳⁴

作者信息 +

OFFSHORE WIND POWER PREDICTION BASED ON IMPROVED DENOISING AUTO-ENCODER AND MULTIVARIATE TIME SERIES CLUSTERING

Zhou Hai¹, Liu Jianfeng¹, Zhou Jian², Zhou Yongliang³, Li Meiyu³, Li Chenyang⁴

Author information +

文章历史 +

摘要

针对海上数值天气预报（NWP）精度低、气象因素复杂等特点,提出一种基于改进的双向降噪自编码器（BDAE）和多元时序聚类的短期海上风电功率预测方法。首先,利用Toeplitz矩阵逆协方差聚类（TICC）进行风况相似性分类,即根据30、70、100 m海上NWP风速进行多元序列实时分割与聚类;然后,针对不同风况类型分别建立可提取过去、未来双向有效信息的改进BDAE修正模型,以修正轮毂高度风速的预测误差;最后,基于修正后的轮毂高度风速以及其余NWP数据,利用TICC算法划分气象相似类型,并在此基础上建立对应的海上风电功率预测模型。采用国内某海上风电场数据进行实验验证,结果表明所提方法能提高海上风电功率预测精度,具有一定工程实用价值。

Abstract

Aiming at the characteristics of low precision and complex meteorological factors of offshore numerical weather prediction (NWP), this paper proposes a short-term offshore wind power forecasting method based on improved bi-directional denoising auto-encoder (BDAE) and clustering of multivariate time series. Firstly, toeplitz inverse Covariance-based clustering (TICC) is used to classify the similarity of wind conditions, that is, the multivariate series are segmented and clustered in real time according to the wind speed of offshore NWP at 30, 70, and 100 m. Secondly, according to different wind conditions, an improved BDAE correction model which can extract the past and future bidirectional effective information is established to correct the prediction error of hub height wind speed.Finally, based on the modified hub height wind speed and other NWP data, the TICC algorithm is used to classify meteorological similarity types, and the corresponding offshore wind power prediction model is established on this basis. The experimental verification is carried out with the data of an offshore wind farm in China, and the results show that the proposed method can improve the accuracy of offshore wind powerforecasting, which have certain engineering practical value.

导出引用

周海, 刘建锋, 周健, 周勇良, 李美玉, 励晨阳. 基于改进降噪自编码器和多元时序聚类的海上风电功率预测[J]. 太阳能学报. 2023, 44(3): 129-138 https://doi.org/10.19912/j.0254-0096.tynxb.2021-1240

Zhou Hai, Liu Jianfeng, Zhou Jian, Zhou Yongliang, Li Meiyu, Li Chenyang. OFFSHORE WIND POWER PREDICTION BASED ON IMPROVED DENOISING AUTO-ENCODER AND MULTIVARIATE TIME SERIES CLUSTERING[J]. Acta Energiae Solaris Sinica. 2023, 44(3): 129-138 https://doi.org/10.19912/j.0254-0096.tynxb.2021-1240

中图分类号： TK513.5

参考文献

[1] Global Wind Energy Council: GWEC GLOBAL WIND REPORT2018[EB/OL]. Brussels,Global Wind Energy Council,(2019-04-03)[2020-07-04].https://gwec.net/wp-content/uploads/2019/04/GWEC-Global-Wind-Report-2018.pdf.
[2] 迟永宁, 梁伟, 张占奎, 等. 大规模海上风电输电与并网关键技术研究综述[J]. 中国电机工程学报, 2016, 36(14): 3758-3771.
CHI Y N, LIANG W, ZHANG Z K, et al.An overview on key technologies regarding power transmission and grid integration of large scale offshore wind power[J]. Proceedings of the CSEE, 2016, 36(14): 3758-3771.
[3] 江长明, 杨健, 柳玉, 等. 基于数据驱动的风电功率预测误差解耦评价方法[J]. 电力系统自动化, 2021, 45(1): 105-113.
JIANG C M, YANG J, LIU Y, et al.Data-driven decoupling evaluation method of wind power prediction error[J]. Automation of electric power systems, 2021, 45(1): 105-113.
[4] XUE H L, SHEN X S, CHOU J F.A forecast error correction method in numerical weather prediction by using recent multiple-time evolution data[J]. Advances in atmospheric sciences, 2013, 30(5): 17-27.
[5] BUHAN S, ÖZKAZANÇ Y, ÇADIRCI. IŞIK Wind pattern recognition and reference wind mast data correlations with NWP for improved wind-electric power forecasts[J]. IEEE transactions on industrial informatics, 2017, 12(3): 991-1004.
[6] LIU Y Q, WANG Y M, LI L, et al.Numerical weather prediction wind correction methods and its impact on computational fluid dynamics based wind power forecasting[J]. Journal of renewable &sustainable energy, 2016, 8(3): 770-778.
[7] 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 C, LIU Y Q, ZHOU J K, 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.
[8] 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, 2015, 7(1): 1-13.
[9] 王勃, 冯双磊, 刘纯. 基于天气分型的风电功率预测方法[J]. 电网技术, 2014, 38(1): 93-98.
WANG B, FENG S L, LIU C.Study on weather typing based wind power prediction[J]. Power system technology, 2014, 38(1): 93-98.
[10] 杨家然, 王兴成, 罗晓芬, 等. 基于数值天气预报及模糊聚类的风电功率智能组合预测[J]. 太阳能学报, 2017, 38(3): 669-675.
YANG J R, WANG X C, LUO X F, et al.Intelligent combined prediction of wind power based on numerical weather prediction and fuzzy clustering[J]. Acta energiae solaris sinica, 2017, 38(3): 669-675.
[11] 熊音笛, 刘开培, 秦亮, 等. 基于时序数据动态天气划分的短期风电功率预测方法[J]. 电网技术, 2019, 43(9): 3353-3359.
XIONG Y D, LIU K P, QIN L, et al.Short-term wind powerprediction method based on time series data dynamic weather division[J]. Power system technology, 2019, 43(9): 3353-3359.
[12] 符杨, 郑紫宸, 时帅, 等. 考虑气象相似性与数值天气预报修正的海上风功率预测[J]. 电网技术, 2019, 43(4): 1253-1260.
FU Y, ZHENG Z C, SHI S, et al.Offshore wind power forecasting considering meteorological similarity and NWP correction[J]. Power system technology, 2019, 43(4): 1253-1260.
[13] 王铮, 王伟胜, 刘纯, 等. 基于风过程方法的风电功率预测结果不确定性估计[J]. 电网技术, 2013, 37(1): 242-247.
WANG Z, WANG W S, LIU C, et al.Uncertaintyestimation of wind power prediction result based on wind processmethod[J]. Power system technology, 2013, 37(1): 242-247.
[14] 阎洁, 许成志, 刘永前, 等. 基于风速云模型相似日的短期风电功率预测方法[J]. 电力系统自动化, 2018, 42(6): 53-59.
YAN J, XU C Z, LIU Y Q, et al.Short-term wind powerprediction method based on wind speed cloud model in similar day[J]. Automation of electric power systems, 2018, 42(6): 53-59.
[15] 丁志勇, 杨苹, 杨曦, 等. 基于连续时间段聚类的支持向量机风电功率预测方法[J]. 电力系统自动化, 2012, 36(14): 131-135.
DING Z Y, YANG P, YANG X, et al.Wind power predictionmethod based on sequential time clustering support vector machine[J]. Automation of electric power systems, 2012, 36(14): 131-135.
[16] DAVID H, SAGAR V, STEPHEN B, et al.Toeplitz inverse covariance-based clustering of multivariate time series data[C]//Knowledge Discovery and Data Mining(KDD). Halifax, Nove Scotia, Canda: Association for Computing Machinery Google, 2017: 1-9.
[17] 杨茂, 周宜. 计及风电场状态的风电功率超短期预测[J]. 中国电机工程学报, 2019, 39(5): 1259-1267.
YANG M, ZHOU Y.Ultra-short-term prediction of wind power considering wind farm status[J]. Proceedings of the CSEE, 2019, 39(5): 1259-1267.
[18] 封宇, 何焱, 朱启昊, 等. 近海及海上风资源时空特性研究[J]. 清华大学学报(自然科学版), 2016, 56(5): 522-529.
FENG Y, HE Y, ZHU Q H, et al.Temporal and spatial characteristics of offshore and offshore wind resources[J]. Journal of Tsinghua University(science and technology)(Sci&Technol), 2016, 56(5): 522-529.
[19] 郁琛, 吕友杰, 段荣华, 等. 基于偏互信息法与支持向量机的覆冰闪络故障预警[J]. 电力系统自动化, 2018, 42(2): 92-98.
YU C, LYU Y J, DUAN R H, et al.A forecast method of icing flashover fault based on partial mutual information method and support vector machine[J]. Automation of electric power systems, 2018, 42(2): 92-98.
[20] 张金艺, 秦政, 林羽晨, 等. 面向伴随型机器人同步运动的递进式步态时相检测算法[J]. 仪器仪表学报, 2020, 41(1): 113-120.
ZHANG J Y, QIN Z, LIN Y C, et al.Progressive gait phase detection algorithm targeting to synchronous motion of companion robots[J]. Chinese journal of scientific instrument, 2020, 41(1): 113-120.
[21] 高亚静, 刘栋, 程华新, 等. 基于数据驱动的短期风电出力预估-校正预测模型[J].中国电机工程学报, 2015, 35(11): 2645-2653.
GAO Y J, LIU D, CHENG H X, et al.Predictor-corrector model of wind power forecast based on data-driven[J]. Proceedings of the CSEE, 2015, 35(11): 2645-2653.
[22] 陈修鹏, 李庚银, 周明, 等. 考虑新能源不确定性和点对点交易的配网产消者分布式优化调度[J]. 电网技术, 2020, 44(9): 3331-3340.
CHEN X P, LI G Y, ZHOU M, et al.Distributed optimal scheduling for prosumers in distribution network consideringuncertainty of renewable sources and P2P trading[J]. Power system technology, 2020, 44(9): 3331-3340.
[23] 代杰杰, 宋辉, 杨祎, 等. 基于栈式降噪自编码器的输变电设备状态数据清洗方法[J]. 电力系统自动化, 2017, 41(12): 224-230.
DAI J J, SONG H, YANG Y, et al.Cleaning method for status data of power transmission and transformation equipment based on stacked denoising autoencoders[J]. Automation of electric power systems, 2017, 41(12): 224-230.
[24] 朱乔木, 李弘毅, 王子琪, 等. 基于长短期记忆网络的风电场发电功率超短期预测[J]. 电网技术, 2017, 41(12): 3797-3802.
ZHU Q M, LI H Y, WANG Z Q, et al.Ultra-short-term prediction of wind farm power generation based on long-term and short-termmemory networks[J]. Power system technology, 2017, 41(12): 3797-3802.
[25] NB/T 31046—2013, 风电功率预测系统规范[S].
[26] 张斌, 庄池杰, 胡军, 等. 结合降维技术的电力负荷曲线集成聚类算法[J].中国电机工程学报, 2015, 35(15): 3741-3749.
ZHANG B, ZHUANG C J, HU J, et al.Ensemble clustering algorithm combined with dimension reduction techniques for power load profiles[J]. Proceedings of the CSEE, 2015, 35(15): 3741-3749.
[27] 周勇良, 余光正, 刘建锋, 等. 基于改进长期循环卷积神经网络的海上风电功率预测[J]. 电力系统自动化, 2021, 45(3): 183-191.
ZHOU Y L, YU G Z, LIU J F, et al.Offshore wind power prediction based on improved long-term recurrent convolutional neural network[J]. Automation of electric power systems, 2021, 45(3): 183-191.