Day-ahead electricity price prediction model based on GRU optimized by crossover optimization algorithm

doi:10.3969/j.issn.2097-0706.2022.11.005

Abstract

Abstract:

The penetration rate of renewable energy is rising with the development of new energy system.In view of the mounting number of features and fluctuation of electricity price， a combined prediction model based on gated recurrent unit （GRU） optimized by crossover optimization algorithm（CSO）is proposed，to improve the prediction accuracy of day-ahead electricity price. Firstly， the correlation between electricity price and features is analysed by mutual information （MI） method and those features with high correlation are extracted. The electricity price sequence is decomposed into several components by variational modal decomposition （VMD）. Features extracted from the superimposed components are sent to GRU prediction model for prediction. CSO algorithm can optimize some parameters of the model and avoid the model from falling into local optimum. The final predicted electricity price is obtained by superimposing the prediction results of different components. Verified by the experimental data of Nordpool， a power market leader in the Nordic region， the proposed prediction method is more accurate than others.

Key words: new power system, renewable energy, electricity price prediction, mutual information, variational mode decomposition, crossover optimization algorithm, gated recurrent unit

CLC Number:

TK01⁺9:TM715

ZHAI Guangsong, WANG Peng, LIANG Pengxun, XIE Zhifeng, YIN Hao. Day-ahead electricity price prediction model based on GRU optimized by crossover optimization algorithm[J]. Integrated Intelligent Energy, 2022, 44(11): 36-42.

Figures/Tables 12

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

Table 2

Fig.5

Fig.6

Table 3

Fig.7

Table 4

Fig.8

References 18

[1]	刘昊, 郭烨, 孙宏斌. 中国跨区跨省电力交易综述及展望[J/OL]. 电力系统自动化, 2022:1-13(2022-04-26)[2022-05-15]. http://kns.cnki.net/kcms/detail/32.1180.TP.20220425.1002.002.html.
	LIU Hao, GUO Ye, SUN Hongbin. Review and prospect of inter-regional and inter-provincial power trading in China[J/OL]. Automation of Electric Power Systems, 2022:1-13(2022-04-26)[2022-05-15]. http://kns.cnki.net/kcms/detail/32.1180.TP.20220425.1002.002.html.
[2]	张一泓, 朱国荣, 蔡永自, 等. 基于自回归积分滑动平均模型的日前电价预测[J]. 自动化技术与应用, 2020, 39(1):125-129,139.
	ZHANG Yihong, ZHU Guorong, CAI Yongzi, et al. Forecast of day-ahead price based on ARIMA model[J]. Techniques of Automation and Applications, 2020, 39(1):125-129,139.
[3]	MUNIAIN P, ZIEL F. Probabilistic forecasting in day-ahead electricity markets: Simulating peak and off-peak prices[J]. International Journal of Forecasting, 2020, 36(4):1193-1210. doi: 10.1016/j.ijforecast.2019.11.006
[4]	殷豪, 曾云, 孟安波, 等. 基于奇异谱分析的短期电价预测[J]. 电力系统保护与控制, 2019, 47(1):115-122.
	YIN Hao, ZENG Yun, MENG Anbo, et al. Short-term electricity price forecasting based on singular spectrum analysis[J]. Power System Protection and Control, 2019, 47(1):115-122.
[5]	陈杰尧, 陶春华, 马光文, 等. 基于数据挖掘与支持向量机的现货市场出清价预测方法[J]. 电网与清洁能源, 2020, 36(10):14-19,27.
	CHEN Jieyao, TAO Chunhua, MA Guangwen, et al. Forecasting method of spot market clearing price based on data mining and support vector machine[J]. Power System and Clean Energy, 2020, 36(10):14-19,27.
[6]	鲁娅楠, 王金梅, 孙帆. 基于粒子群算法的BP神经网络电价预测研究[J]. 科技创新与应用, 2018(28):15-17.
	LU Yanan, WANG Jinmei, SUN Fan. Research on BP neural network electricity price prediction based on particle swarm optimization[J]. Technology Innovation and Application, 2018(28):15-17.
[7]	UGURLU U, OKSUZ I, TAS O. Electricity price forecasting using recurrent neural networks[J]. Energies, 2018, 11(5):1255. doi: 10.3390/en11051255
[8]	殷豪, 丁伟锋, 陈顺, 等. 基于长短时记忆网络-纵横交叉算法的含高比例新能源电力市场日前电价预测[J]. 电网技术, 2022, 46(2):472-480.
	YIN Hao, DING Weifeng, CHEN Shun, et al. Day-ahead electricity price forecasting of power market with high proportion of new energy based on LSTM-CSO model[J]. Power System Technology, 2022, 46(2):472-480.
[9]	谢谦, 董立红, 厍向阳. 基于Attention-GRU的短期电价预测[J]. 电力系统保护与控制, 2020, 48(23):154-160.
	XIE Qian, DONG Lihong, SHE Xiangyang. Short-term electricity price forecasting based on Attention-GRU[J]. Power System Protection and Control, 2020, 48(23):154-160.
[10]	MACIEJOWSKA K, NITKA W, WERON T. Enhancing load,wind and solar generation for day-ahead forecasting of electricity prices[J]. Energy Economics, 2021, 99:105273. doi: 10.1016/j.eneco.2021.105273
[11]	LI W, BECKER D M. Day-ahead electricity price prediction applying hybrid models of LSTM-based deep learning methods and feature selection algorithms under consideration of market coupling[J]. Energy, 2021, 237:121543. doi: 10.1016/j.energy.2021.121543
[12]	刘达, 雷自强, 孙堃. 基于小波包分解和长短期记忆网络的短期电价预测[J]. 智慧电力, 2020, 48(4):77-83.
	LIU Da, LEI Ziqiang, SUN Kun. Short-term electricity price forecasting based on wavelet packet decomposition & long-term and short-term memory networks[J]. Smart Power, 2020, 48(4):77-83.
[13]	勾玄, 肖先勇. 基于经验模式分解与LSTM神经网络的短期电价预测模型[J]. 西安理工大学学报, 2020, 36(1):129-134.
	GOU Xuan, XIAO Xianyong. Short-term electricity price forecasting model based on empirical mode decomposition and LSTM neural network[J]. Journal of Xi'an University of Technology, 2020, 36(1):129-134.
[14]	张金良, 王明雪. 基于EEMD,SVM和ARMA组合模型的电价预测[J]. 电力需求侧管理, 2020, 22(3):63-68.
	ZHANG Jinliang, WANG Mingxue. Electricity price forecasting based on EEMD,SVM and ARMA combination model[J]. Power Demand Side Management, 2020, 22(3):63-68.
[15]	杨昭, 张钢, 赵俊杰, 等. 基于变分模态分解和改进粒子群算法优化最小二乘支持向量机的短期电价预测[J]. 电气技术, 2021, 22(10):11-16.
	YANG Zhao, ZHANG Gang, ZHAO Junjie, et al. Short term electricity price forecasting based on variational mode decomposition and improved particle swarm optimization-least square support vector machine[J]. Electrical Engineering, 2021, 22(10):11-16.
[16]	陈涛, 吕松, 任廷林, 等. 基于最小二乘支持向量机的周用电量预测方法[J]. 华电技术, 2020, 42(1): 35-40.
	CHEN Tao, LYU Song, REN Tinglin, et al. Prediction method for weekly electricity consumption based on LSSVM algorithm[J]. Huadian Technology, 2020, 42(1): 35-40.
[17]	闫来清, 董泽. 基于k-近邻互信息和WKOPLS的SCR脱硝系统动态预测模型[J]. 中国电机工程学报, 2019, 39(10):2970-2980.
	YAN Laiqing, DONG Ze. Dynamic prediction model of SCR denitrification system based on k-nearest neighbor mutual information and WKOPLS[J]. Proceedings of the CSEE, 2019, 39(10):2970-2980.
[18]	王陈恩, 殷豪, 陈顺, 等. 考虑空间耦合的少数据风电功率预测方法[J/OL]. 南方电网技术, 2022:1-6(2022-02-23)[2022-05-15]. http://kns.cnki.net/kcms/detail/44.1643.TK.20220222.1836.006.html.
	WANG Chenen, YIN Hao, CHEN Shun, et al. Wind power forecasting method with few data considering spatial coupling[J/OL]. Southern Power System Technology, 2022:1-6(2022-02-23)[2022-05-15]. http://kns.cnki.net/kcms/detail/44.1643.TK.20220222.1836.006.html.

特征		MI值
生产		0.587 8
负荷		0.415 3
风电		0.206 5
计划流量		0.574 1
市场容量		0.317 9
市场耦合因素	瑞典净交换电量	0.095 8
	挪威净交换电量	0.688 0
	丹麦净交换电量	0.287 6
	芬兰净交换电量	0.148 9
	立陶宛净交换电量	0.104 4
	拉脱维亚净交换电量	0.229 1
	爱沙尼亚净交换电量	0.317 5
	挪威价区2-荷兰交换电量	0.193 5
	瑞典价区4-德国交换电量	0.317 0
	丹麦价区1-德国交换电量	0.320 7
	丹麦价区2-德国交换电量	0.215 4

预测模型	X_RMSE	X_MAE	X_SMAPE
GRU	3.694	2.711	13.149
PCA-GRU	3.297	2.503	11.858
MI-GRU	3.282	2.450	11.696

预测模型	X_RMSE	X_MAE	X_SMAPE
GRU	3.282	2.450	11.696
EEMD-GRU	2.307	1.721	8.286
VMD-GRU	2.064	1.545	7.616

预测模型	X_RMSE	X_MAE	X_SMAPE
BP	2.321	1.736	8.438
RNN	2.529	1.873	9.155
LSTM	2.138	1.623	8.022
GRU	2.064	1.545	7.616
CSO-GRU	2.026	1.523	7.546