Short-term wind power forecasting based on variational mode decomposition and generative adversarial networks

doi:10.3969/j.issn.2097-0706.2024.02.004

Abstract

Abstract:

Although certain progress has been achieved on wind power predictability and prediction accuracy， the short-term prediction accuracy is restricted by the strong nonlinearity of meteorological and wind power data. Hence， the improvement on short-term wind power prediction is a research hotspot. To deal with the nonlinearity and instability of wind power data， a short-term wind power forecasting method based on variational model decomposition and generative adversarial networks is proposed. The method introduces variational mode decomposition to disperse the nonlinearity of wind power data and decompose the complex sequence of prediction task into several simple sequences. Activation function and loss function are designed to solve the instability of traditional generative adversarial network models， and the key parameters of the designed activation function are analysed. The proposed method is tested on the data from a Bengaluru wind farm， showing a decent prediction result. Its forecasting mean square error is 79.65% and 51.83% lower than that of long short-term memory and variational mode decomposition-long short-term memory，respectively.

Key words: short-term wind power forecasting, variational mode decomposition, generative adversarial network, long short-term memory networks, activation function

CLC Number:

TM761

JIANG Shanhe, LI Wei, XU Xiaoyan, WANG Dekai. Short-term wind power forecasting based on variational mode decomposition and generative adversarial networks[J]. Integrated Intelligent Energy, 2024, 46(2): 28-35.

Figures/Tables 8

Fig.1

Fig.2

Table 1

Fig.3

Table 2

Table 3

Fig.4

Fig.5

References 28

[1]	LIU H, CHEN C. Data processing strategies in wind energy forecasting models and applications: A comprehensive review[J]. Applied Energy, 2019, 249: 392-408. doi: 10.1016/j.apenergy.2019.04.188
[2]	YANG J, FANG L Q, SONG D R, et al. Review of control strategy of large horizontal‐axis wind turbines yaw system[J]. Wind Energy, 2021, 24(2): 97-115. doi: 10.1002/we.v24.2
[3]	WANG Y, HU Q H, SRINIVASAN D, et al. Wind power curve modeling and wind power forecasting with inconsistent data[J]. IEEE Transactions on Sustainable Energy, 2018, 10(1): 16-25. doi: 10.1109/TSTE.2018.2820198
[4]	WANG Y, ZOU R M, LIU F, et al. A review of wind speed and wind power forecasting with deep neural networks[J]. Applied Energy, 2021, 304(1): 117766. doi: 10.1016/j.apenergy.2021.117766
[5]	薛禹胜, 郁琛, 赵俊华, 等. 关于短期及超短期风电功率预测的评述[J]. 电力系统自动化, 2015, 39(6):141-151.
	XUE Yusheng, YU Chen, ZHAO Junhua, et al. A review on short-term and ultra-short-term wind power prediction[J]. Automation of Electric Power Systems, 2015, 39(6):141-151.
[6]	王伟胜, 王铮, 董存, 等. 中国短期风电功率预测技术现状与误差分析[J]. 电力系统自动化, 2021, 45(1): 17-27.
	WANG Weisheng, WANG Zheng, DONG Cun, et al. Current status and error analysis of short-term wind power forecasting technology in China[J]. Automation of Electric Power System, 2021, 45(1): 17-27.
[7]	王玉. 基于深度学习的风电场短期风速预测组合模型[D]. 合肥: 中国科学技术大学, 2021.
	WANG Yu. Combination model of short-term wind speed prediction for wind farm based on deep learning[D]. Hefei: University of Science and Technology of China, 2021.
[8]	HU J M, HENG J N, WEN J M, et al. Deterministic and probabilistic wind speed forecasting with de-noisingreconstruction strategy and quantile regression based algorithm[J]. Renewable Energy, 2020, 162:1208-1226. doi: 10.1016/j.renene.2020.08.077
[9]	MA L, LUAN S Y, JIANG C W, et al. A review on the forecasting of wind speed and generated power[J]. Renewable and Sustainable Energy Reviews, 2009, 13(4): 915-920. doi: 10.1016/j.rser.2008.02.002
[10]	庞传军, 尚学伟, 张波, 等. 基于改进梯度提升算法的短期风电功率概率预测[J]. 电力系统自动化, 2022, 46(16): 198-206.
	PANG Chuanjun, SHAN Xuewei, ZHANG Bo, et al. Short-term wind power probability prediction based on improved gradient boosting machine algorithm[J]. Automation of Electric Power System, 2022, 46(16): 198-206.
[11]	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 Jie, XU Chengzhi, LIU Yongqian, et al. Short-term wind power prediction method based on wind speed cloud model in similar day[J]. Automation of Electric Power System, 2018, 42(6): 53-59.
[12]	SHI J, GUO J M, ZHENG S T. Evaluation of hybrid forecasting approaches for wind speed and power generation time series[J]. Renewable and Sustainable Energy Reviews, 2012, 16(5): 3471-3480. doi: 10.1016/j.rser.2012.02.044
[13]	SHUKUR O B, LEE M H. Daily wind speed forecasting through hybrid KF-ANN model based on ARIMA[J]. Renewable Energy, 2015, 76: 637-647. doi: 10.1016/j.renene.2014.11.084
[14]	DE GIORGI M G, CAMPILONGO S, FICARELLA A, et al. Comparison between wind power prediction models based on wavelet decomposition with least-squares support vector machine (LS-SVM) and artificial neural network (ANN)[J]. Energies, 2014, 7(8): 5251-5272. doi: 10.3390/en7085251
[15]	郑真, 朱峰, 马小丽, 等. 基于TL-LSTM的新能源功率短期预测[J]. 综合智慧能源, 2023, 45(1): 41-48. doi: 10.3969/j.issn.2097-0706.2023.01.005
	ZHENG Zhen, ZHU Feng, MA Xiaoli, et al. Short-term new energy power prediction based on TL-LSTM[J]. Integrated Intelligent Energy, 2023, 45(1): 41-48. doi: 10.3969/j.issn.2097-0706.2023.01.005
[16]	杨晶显, 张帅, 刘继春, 等. 基于VMD和双重注意力机制LSTM的短期光伏功率预测[J]. 电力系统自动化, 2021, 45(3):174-182.
	YANG Jingxian, ZHANG Shuai, LIU Jichun, et al. Short-term PV power prediction based on VMD and dual attention mechanism LSTM[J]. Automation of Electric Power Systems, 2021, 45(3):174-182.
[17]	苗长新, 李昊, 王霞, 等. 基于数据驱动和深度学习的超短期风电功率预测[J]. 电力系统自动化, 2021, 45(14): 22-29.
	MIAO Changxin, LI Hao, WANG Xia, et al. Data-driven and deep-learning-based ultra-short-term wind power prediction[J]. Automation of Electric Power Systems, 2021, 45(14):22-29.
[18]	GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]// Advances in Neural Information Processing Systems, 2014: 2672-2680.
[19]	王坤峰, 苟超, 段艳杰, 等. 生成式对抗网络 GAN 的研究进展与展望[J]. 自动化学报, 2017, 43(3): 321-332.
	WANG Kunfeng, GOU Chao, DUAN Yanjie, et al. Generative adversarial networks: The state of the art and beyond[J]. Acta Automatic Sinica, 2017, 43(3):321-332.
[20]	ALDAUSARI N, SOWMYA A, MARCUS N, et al. Video generative adversarial networks: A review[J]. ACM Computing Surveys (CSUR), 2022, 55(2): 1-25.
[21]	王静. 基于多元经验模态分解生成对抗网络的金融时间序列预测[D]. 大连: 大连海事大学, 2020.
	WANG Jing. Financial time series prediction of adversarial networks based on multivariate empirical mode decomposition[D]. Dalian: Dalian Maritime University, 2020.
[22]	DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2013, 62(3): 531-544.
[23]	李文武, 石强, 王凯, 等. 基于变分模态分解和深度门控网络的径流预测[J]. 水力发电学报, 2020, 39(3): 34-44.
	LI Wenwu, SHI Qiang, WANG Kai, et al. Runoff prediction based on variational mode decomposition and depth-gated networks[J]. Journal of Hydropower, 2020, 39(3): 34-44.
[24]	翟广松, 王鹏, 梁鹏勋, 等. 基于纵横交叉算法-门控循环单元的日前电价预测模型[J]. 综合智慧能源, 2022, 44(11): 36-42. doi: 10.3969/j.issn.2097-0706.2022.11.005
	ZHAI Guangsong, WANG Peng, LIANG Pengxun, et al. Day-ahead electricity price prediction model based on GRU optimized by crossover optimization algorithm[J]. Integrated Intelligent Energy, 2022, 44(11): 36-42. doi: 10.3969/j.issn.2097-0706.2022.11.005
[25]	简定辉, 李萍, 黄宇航. 基于GA-VMD-ResNet-LSTM网络的短期电力负荷预测[J]. 国外电子测量技术, 2022, 41(10): 15-22.
	JIAN Dinghui, LI Pin, HUANG Yuhang. Short-term power load forecasting based on GA-VMD-ResNet-LSTM network[J]. Foreign Electronic Measurement Technology, 2022, 41(10): 15-22.
[26]	KINGMA D P, BA J. Adam:A method for stochastic optimization[C]//The 3rd International Conference for Learning Representations, San Diego, 2015:1-15.
[27]	YUNUS K, THIRINGER T, CHEN P Y. ARIMA-based frequency-decomposed modeling of wind speed time series[J]. IEEE Transactions on Power Systems, 2015, 31(4): 2546-2556. doi: 10.1109/TPWRS.59
[28]	LIU H, MI X W, LI Y F. Smart multi-step deep learning model for wind speed forecasting based on variational mode decomposition, singular spectrum analysis, LSTM network and ELM[J]. Energy Conversion and Management, 2018, 159: 54-64. doi: 10.1016/j.enconman.2018.01.010

k	中心频率
2	0.000 40, 0.041 81
3	0.000 40, 0.041 24, 0.083 35
4	0.000 40, 0.041 22, 0.082 79, 0.161 03
5	0.000 39, 0.041 20, 0.081 78, 0.123 27, 0.167 87

l	MAE	MSE	RMSE	δmax
0.01	24.694 4	2 489.314 0	49.893 0	577.994 6
0.05	26.747 6	2 390.992 7	48.897 8	829.124 8
0.10	21.350 2	2 118.766 7	46.030 1	518.077 9
0.20	26.300 1	2 659.534 2	51.570 7	580.099 7
0.30	25.740 3	2 712.210 2	52.078 9	565.314 7
0.40	26.747 7	2 724.080 6	52.192 7	565.858 8
0.50	29.226 6	3 427.549 6	58.545 3	629.489 4

预测模型	MAE	MSE	RMSE	δmax
ARIMA^[27]	53.359 0	17 770.382 1	133.305 6	1 674.928 9
LSTM^[28]	72.260 1	10 412.7773	102.043 0	972.853 4
VMD-ARIMA^[27]	34.521 1	6 054.779 6	77.812 5	1 514.236 6
VMD-LSTM^[28]	39.224 3	4 398.470 7	66.321 0	1 008.752 3
VMD-GAN(l = 0.10)	21.350 2	2 118.766 7	46.030 1	518.077 9