数据驱动的时间注意力卷积风电功率预测模型

张浩田; 温蜜; 李晋国; 田英杰

doi:10.19912/j.0254-0096.tynxb.2021-0453

PDF(32666 KB)

太阳能学报 ›› 2022, Vol. 43 ›› Issue (10) : 167-176. DOI: 10.19912/j.0254-0096.tynxb.2021-0453

数据驱动的时间注意力卷积风电功率预测模型

张浩田¹, 温蜜¹, 李晋国¹, 田英杰²

作者信息 +

DATA DRIVEN TIME ATTENTION CONVOLUTION WIND POWER PREDICTION MODEL

Zhang Haotian¹, Wen Mi¹, Li Jinguo¹, Tian Yingjie²

Author information +

文章历史 +

摘要

由于风电受气象特征影响大,风能波动性和间歇性强,导致快速、精准的风电预测成为一个难题。对此,该文提出一种基于数据驱动的时间注意力卷积网络的风电功率预测方法。首先,将来自风力机和数据采集（SCADA）系统的数据进行清洗;然后采用可并行计算的时间卷积网络,并加入Attention机制突出关键特征的影响,使模型训练速度和预测精度得到有效提升。实验结果表明,该文所提方法与其他方法相比可更准确地减少数据噪声,同时有更高的预测精度和更快的训练速度。

Abstract

Because wind power is greatly affected by meteorological characteristics, and ther wind energy has strong volatility and intermittence, so that fast and accurate wind power prediction becomes a difficult problem. Therefore, a data-driven time attention convolution network wind power prediction method is proposed. Firstly, the data from the wind turbine and SCADA system are cleaned. Then the Temporal convolutional network which can be calculated in parallel is adopted, and the attention mechanism is added to highlight the influence of key features, so that the training speed and prediction accuracy of the model are effectively improved. Experimental results show that compared with other methods, the proposed method can reduce data noise more accurately, and has higher prediction accuracy and faster training speed.

导出引用

张浩田, 温蜜, 李晋国, 田英杰. 数据驱动的时间注意力卷积风电功率预测模型[J]. 太阳能学报. 2022, 43(10): 167-176 https://doi.org/10.19912/j.0254-0096.tynxb.2021-0453

Zhang Haotian, Wen Mi, Li Jinguo, Tian Yingjie. DATA DRIVEN TIME ATTENTION CONVOLUTION WIND POWER PREDICTION MODEL[J]. Acta Energiae Solaris Sinica. 2022, 43(10): 167-176 https://doi.org/10.19912/j.0254-0096.tynxb.2021-0453

中图分类号： TM614

参考文献

[1] YANG D X, JING Y Q, WANG C, et al.Analysis of renewable energy subsidy in China under uncertainty: feed-in tariff vs. renewable portfolio standard[J]. Energy strategy reviews, 2021, 34: 100628.
[2] AZAM M, KHAN A Q, OZTURK I.The effects of energy on investment, human health, environment and economic growth: empirical evidence from China[J]. Environmental science and pollution research, 2019, 26(11): 10816-10825.
[3] LEE J, ZHAO F.Global wind report 2021[R]. Brussels: Global Wind Energy Council, 2021.
[4] 韩自奋, 景乾明, 张彦凯, 等. 风电预测方法与新趋势综述[J]. 电力系统保护与控制, 2019, 47(24): 12-18.
HAN Z F, JING Q M, ZHANG Y K, et al.Overview of wind power forecasting methods and new trends[J]. Power system protection and control, 2019, 47(24): 12-18.
[5] 李莉, 刘永前, 杨勇平, 等. 基于CFD流场预计算的短期风速预测方法[J]. 中国电机工程学报, 2013, 33(7): 27-32.
LI L, LIU Y Q, YANG Y P, et al.Short term wind speed prediction method based on CFD flow field pre calculation[J]. Chinese journal of electrical engineering, 2013, 33(7): 27-32.
[6] ELDALI F A, HANSEN T M, SURYANARAYANAN S, et al.Employing ARIMA models to improve wind power forecasts: a case study in ERCOT[C]//North American Power Symposium, Denver, CO, USA, 2016: 1-6.
[7] WANG C, ZHANG H L, MA P.Wind power forecasting based on singular spectrum analysis and a new hybrid laguerre neural network[J]. Applied energy, 2020, 259:114139.
[8] LIU J, WANG X, LU Y.A novel hybrid methodology for short-term wind power forecasting based on adaptive neuro-fuzzy inference system[J] Renew energy, 2017, 103: 620-629.
[9] 崔杨, 李莉, 陈德荣. 基于最小二乘支持向量机的超短期风电负荷预测[J]. 电气自动化, 2014(5): 35-37.
CUI Y, LI L, CHEN D R.Ultra short term wind power load forecasting based on least squares support vector machine[J]. Electrical automation, 2014(5): 35-37.
[10] KISVARI A, LIN Z, LIU X.Wind power forecasting——a data-driven method along with gated recurrent neural network[J]. Renewable energy, 2021, 163: 1895-1909.
[11] 陈禹帆, 温蜜, 张凯, 等. 基于相似日匹配及TCN-Attention的短期光伏出力预测[J]. 电测与仪表,2022,59(10): 108-116.
CHEN Y F, WEN M, ZHANG K, et al.Short term photovoltaic output prediction based on similar day matching and TCN-attention[J]. Electrical measurement & instrumentation, 2022, 59(10): 108-116.
[12] SONG Z Y, BROWN L.Multi-dimensional evaluation of temporal neural networks on solar irradiance forecasting[C]//2019 IEEE Innovative Smart Grid Technologies-Asia(ISGT Asia), Chengdu, China, 2019: 4192-4197.
[13] GAN Z H, LI C S, ZHOU J Z, et al.Temporal convolutional networks interval prediction model for wind speed forecasting, electric power systems research[J]. Electric power systems research, 2021, 191(4): 106865.
[14] LIN Z, LIU X L.Wind power forecasting of an offshore wind turbine based on high-frequency SCADA data and deep learning neural network[J]. Energy, 2020, 201: 117693.
[15] 娄建楼, 胥佳, 陆恒, 等. 基于功率曲线的风电机组数据清洗算法[J]. 电力系统自动化, 2016, 40(10): 116-121.
LOU J L, XU J, LU H, et al.Wind turbine data cleaning algorithm based on power curve[J]. Power system automation, 2016, 40(10): 116-121.
[16] 王一妹, 刘辉, 宋鹏, 等. 基于多阶段递进识别的风电机组异常运行数据清洗方法[J]. 可再生能源, 2020, 38(11): 1470-1476.
WANG Y M, LIU H, SONG P, et al.Data cleaning method for abnormal operation of wind turbines based on multi-stage progressive identification[J]. Renewable energy, 2020, 38(11): 1470-1476.
[17] 邹同华, 高云鹏, 伊慧娟, 等. 基于Thompson tau-四分位和多点插值的风电功率异常数据处理[J]. 电力系统自动化, 2020, 44(15): 156-162.
ZOU T H, GAO Y P, YI H J, et al.Abnormal data processing of wind power based on Thompson tau quartile and multipoint interpolation[J]. Power system automation, 2020, 44(15): 156-162.
[18] 周玉, 孙红玉, 朱文豪, 等. 基于<inline-formula><mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="Mml1-0254-0096-43-10-167"><mml:mi>K</mml:mi></mml:math></inline-formula>均值聚类的分段样本数据选择方法[J]. 计算机应用研究, 2021, 38(6): 1683-1688.
ZHOU Y, SUN H Y, ZHU W H, et al.Segmented sample data selection method based on <inline-formula><mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="Mml2-0254-0096-43-10-167"><mml:mi>K</mml:mi></mml:math></inline-formula>-means clustering[J]. Computer application research, 2021, 38(6): 1683-1688.
[19] 朱壮壮, 周治平. 高斯混合生成模型检测健康数据异常[J]. 计算机科学与探索, 2022, 16(5): 1128-1135.
ZHU Z Z, ZHOU Z P.Detection of abnormal health data using Gaussian mixture generation model[J]. Journal of frontiers of compater science and technology, 2022, 16(5): 1128-1135.
[20] YANG L M, BI X H, WANG L M, et al.A method of two-stage clustering learning based on improved DBSCAN and density peak algorithm[J]. Computer communications, 2021, 167: 75-84.
[21] 杨建, 王力, 宋冬然, 等. 基于孤立森林与稀疏高斯过程回归的风电机组偏航角零点漂移诊断方法[J]. 中国电机工程学报, 2021, 41(18): 6198-6212.
YANG J, WANG L, SONG D R,et al.Wind turbine yaw angle zero drift diagnosis method based on isolated forest and sparse gaussian process regression[J]. Chinese journal of electrical engineering, 2021, 41(18): 6198-6212.
[22] 黄秋娟. 基于数据驱动的风电机组功率曲线异常识别方法研究[D]. 沈阳: 沈阳工业大学, 2019.
HUANG Q J.Research on abnormal identification method of wind turbine power curve based on data driven[D]. Shenyang: Shenyang University of Technology, 2019.
[23] 王黎明, 吴香华, 赵天良, 等. 基于距离相关系数和支持向量机回归的PM_2.5浓度滚动统计预报方案[J]. 环境科学学报, 2017, 37(4): 1268-1276.
WANG L M, WU X H, ZHAO T L, et al.PM based on distance correlation coefficient and support vector machine regression PM_2.5 concentration rolling statistical prediction scheme[J]. Journal of environmental science, 2017, 37(4): 1268-1276.