Short-term wind power forecasting based on DenseNet convolutional neural networks

doi:10.3969/j.issn.2097-0706.2024.07.002

Abstract

Abstract:

Wind energy， as a clean and renewable energy source， plays a crucial role in the energy transformation. And accurate prediction on wind power output is important for the safe and efficient operation of the power system. However， the volatility and randomness of wind speed challenges the wind power prediction. To improve the accuracy of the prediction， a short-term wind power prediction model based on DenseNet convolutional neural network is proposed. The DenseNet160 network obtained from a simplified DenseNet160 network is of an outstanding densely connected structure，and proper depth and width，capable of solving vanishing gradient in the training process and realizing deep supervision by sending information from a upper layer to a deeper layer. Based on the 378-day wind power dataset collected from Natal in Brazil， the wind power output of the next day was predicted by DenseNet160 network and other 27 algorithms. The prediction results show that the mean absolute error （MAE）， mean squared error （MSE） and mean absolute percentage error （MAPE） of the DenseNet160 network is 10.89%，4.98% and 8.68% smaller than that of the second best algorithm， respectively. Meanwhile， the MAE of the DenseNet160 network is 25.56% smaller than that of the hybrid economy model using the same dataset. This results indicate that the proposed prediction model can fit the wind power data more accurately and obtain more reliable wind power prediction results.

Key words: wind power prediction, new energy, DenseNet, convolutional neural network, dense connection, vanishing gradient

CLC Number:

TK01⁺8:TM614

YIN Linfei, MENG Yujie. Short-term wind power forecasting based on DenseNet convolutional neural networks[J]. Integrated Intelligent Energy, 2024, 46(7): 12-20.

Figures/Tables 13

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Table 2

Table 3

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Table 4

References 33

[1]	OCHOA-BARRAGÁN R, PONCE-ORTEGA J M, TOVAR-FACIO J. Long-term energy transition planning: Integrating battery system degradation and replacement for sustainable power systems[J]. Sustainable Production and Consumption, 2023, 42: 335-350.
[2]	LEAL FILHO W, AZEITEIRO U M, BALOGUN A L, et al. The influence of ecosystems services depletion to climate change adaptation efforts in Africa[J]. Science of the Total Environment, 2021, 779: 146414.
[3]	MSIGWA G, IGHALO J O, YAP P S. Considerations on environmental, economic, and energy impacts of wind energy generation:Projections towards sustainability initiatives[J]. Science of the Total Environment, 2022, 849: 157755.
[4]	赵建立, 向佳霓, 王隗东, 等. 考虑风电不确定性的数据中心平抑风电功率波动的调度方法[J]. 综合智慧能源, 2022, 44(11): 70-78. doi: 10.3969/j.issn.2097-0706.2022.11.010
	ZHAO Jianli, XIANG Jiani, WANG Weidong, et al. A scheduling method for suppressing wind power fluctuation of data centers considering wind power uncertainty[J]. Integrated Intelligent Energy, 2022, 44(11): 70-78. doi: 10.3969/j.issn.2097-0706.2022.11.010
[5]	MEENAL R, BINU D, RAMYA K C, et al. Weather forecasting for renewable energy system:A review[J]. Archives of Computational Methods in Engineering, 2022, 29(5): 2875-2891.
[6]	WANG H, HAN S, LIU Y Q, et al. Sequence transfer correction algorithm for numerical weather prediction wind speed and its application in a wind power forecasting system[J]. Applied Energy, 2019, 237: 1-10.
[7]	PEARRE N S, SWAN L G. Statistical approach for improved wind speed forecasting for wind power production[J]. Sustainable Energy Technologies and Assessments, 2018, 27: 180-191.
[8]	ZHANG Y G, ZHAO Y A, KONG C H, et al. A new prediction method based on VMD-PRBF-ARMA-E model considering wind speed characteristic[J]. Energy Conversion and Management, 2020, 203: 112254.
[9]	LIU J P, WANG P, CHEN H Y, et al. A combination forecasting model based on hybrid interval multi-scale decomposition: Application to interval-valued carbon price forecasting[J]. Expert Systems with Applications, 2022, 191: 116267.
[10]	ALY H H. A novel deep learning intelligent clustered hybrid models for wind speed and power forecasting[J]. Energy, 2020, 213: 118773.
[11]	YIN X X, ZHAO X W. Big data driven multi-objective predictions for offshore wind farm based on machine learning algorithms[J]. Energy, 2019, 186: 115704.
[12]	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: 117766.
[13]	SAYEED A, CHOI Y S, JUNG J, et al. A deep convolutional neural network model for improving WRF simulations[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021.
[14]	SAEED A, LI C S, GAN Z H, et al. A simple approach for short-term wind speed interval prediction based on independently recurrent neural networks and error probability distribution[J]. Energy, 2022, 238: 122012.
[15]	LIU M D, DING L, BAI Y L. Application of hybrid model based on empirical mode decomposition, novel recurrent neural networks and the ARIMA to wind speed prediction[J]. Energy Conversion and Management, 2021, 233:113917.
[16]	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
[17]	WEI J Q, WU X J, YANG T M, et al. Ultra-short-term forecasting of wind power based on multi-task learning and LSTM[J]. International Journal of Electrical Power & Energy Systems, 2023, 149: 109073.
[18]	XU L, OU Y X, CAI J J, et al. Offshore wind speed assessment with statistical and attention-based neural network methods based on STL decomposition[J]. Renewable Energy, 2023, 216: 119097.
[19]	CHENG J M, LIANG RY, ZHAO L. DNN-based speech enhancement with self-attention on feature dimension[J]. Multimedia Tools and Applications, 2020, 79:32449-32470.
[20]	KHODAYAR M, KAYNAK O, KHODAYAR M E. Rough deep neural architecture for short-term wind speed forecasting[J]. IEEE Transactions on Industrial Informatics, 2017, 13(6): 2770-2779.
[21]	XIONG B R, MENG X Y, WANG R H, et al. Combined model for shor-term wind power prediction based on deep neural network and long short-term memory[J]. Journal of Physics: Conference Series, 2021, 1757(1): 012095.
[22]	QURESHI A S, KHAN A, ZAMEER A, et al. Wind power prediction using deep neural network based meta regression and transfer learning[J]. Applied Soft Computing, 2017, 58: 742-755.
[23]	MEDHAT S, ABDEL-GALIL H, ABOUTABL A E, et al. Iterative magnitude pruning-based light-version of AlexNet for skin cancer classification[J]. Neural Computing and Applications, 2024, 36(3): 1413-1428.
[24]	ZHU H Y, SUN M Y, FU H H, et al. Training a seismogram discriminator based on ResNet[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 59(8):7076-7085.
[25]	HUANG G, LIU Z, MAATEN L, et al. Densely connected convolutional networks[C]// Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
[26]	ZHANG Q S, WANG X, WU Y N, et al. Interpretable CNNs for object classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 43(10): 3416-3431.
[27]	IMANI M. Electrical load-temperature CNN for residential load forecasting[J]. Energy, 2021, 227: 120480.
[28]	YAN B W, SHEN R F, LI K, et al. Spatio-temporal correlation for simultaneous ultra-short-term wind speed prediction at multiple locations[J]. Energy, 2023, 284: 128418.
[29]	栗然, 罗东晖, 李鹏程, 等. 基于宽度和深度模型以及残差网络的综合能源负荷短期预测[J]. 华北电力大学学报(自然科学版), 2023, 50(6): 21-30.
	SU Ran, LUO Donghui, LI Pengcheng, et al. Comprehensive energy load short-term forecasting based on wide＆deep and ResNet network framework[J]. Journal of North China Electric Power University (Natural Science Edition), 2023, 50(6): 21-30.
[30]	余珊, 温蜜, 顾春华, 等. 基于VMD-DenseNet的短期电力负荷预测[J]. 计算机应用与软件, 2023, 40(7): 34-40, 70.
	YU Shan, WEN Mi, GU Chunhua, et al. Short-term electric load forecasting based on VMD-DenseNet[J]. Computer Applications and Software, 2023, 40(7): 34-40, 70.
[31]	SILVA E D P, SALLES J L F, FARDIN J F, et al. Management of an island and grid-connected microgrid using hybrid economic model predictive control with weather data[J]. Applied Energy, 2020, 278: 115581.
[32]	LI T, JIAO W C, WANG L N, et al. Automatic DenseNet sparsification[J]. IEEE Access, 2020, 8: 62561-62571.
[33]	SILVA E D P, SALLES J L F, FARDIN J F, et al. Measured and forecasted weather and power dataset for management of an island and grid-connected microgrid[J]. Data in Brief, 2021, 39: 107513.

变量	内容
空气温度/℃	标准温度、测量温度、最高温度、最低温度、预测温度
湿度/%	标准湿度、测量湿度、最大湿度、最小湿度
露点温度/℃	标准露点温度、最高露点温度、最低露点温度
大气压力/hPa	标准气压、测量气压、最大气压、最低气压
风速/(m·s^-1)	真实风速、最大风速、预测风速
风向/(°)	测量风向、预测风向
全球水平辐照度/(kJ·m^-2)	短波辐射能、短波辐射区域、预测短波辐射、测量全球水平辐照度
降水量/mm	降水量、测量降水量

算法	编号	算法名称	层数	连接数
人工智能算法	1	DenseNet160	619	704
	2	AlexNet	25	24
	3	DarkNet19	64	63
	4	DarkNet53	184	206
	5	DenseNet201	708	805
	6	EffiientNetb0	290	363
	7	GoogLeNet	144	170
	8	InceptionResNetV2	823	920
	9	InceptionV3	315	349
	10	MobileNetV2	154	163
	11	Place365GoogLeNet	144	170
	12	ResNet101	71	78
	13	ResNet18	177	192
	14	ResNet50	349	379
	15	ShuffleNet	68	75
	16	SqueezeNet	41	40
	17	VGG16	47	46
	18	VGG19	170	181
	19	Xception	172	187
统计算法	20	AdaBoost Regression
	21	Automatic Relevance Determination
	22	Bayesian Ridge Regression
	23	Extra-trees Regression
	24	Gradient Boosting Regression
	25	Huber Regression
	26	Kernel Ridge Regression
	27	Nearest Neighbors Regression
	28	Ordinary Least Squares

网络	层数	连接数	δ_MAE/W
DenseNet160	619	704	50.1
DenseNet201	708	805	60.4