Multivariate Load Forecasting for Integrated Energy Systems Based on Error-compensated LSTM-GRU

doi:10.11985/2023.04.034

Abstract

Abstract:

To facilitate the dispatch of energy systems, joint forecasting of electric, thermal, and cooling loads of integrated energy systems is required. Aiming at the complex influencing factors of multi-load in integrated energy systems, firstly, the factors that have a greater impact on the multi-load load are screened out through grey correlation analysis. The load is then predicted using a traditional long short-term memory(LSTM) neural network. Secondly, the prediction error is trained by the gated recurrent unit(GRU) to obtain the error compensation value. Finally, through the reconstruction of the load forecast value and the error forecast value, a more accurate load forecast value is obtained. The feasibility of the error compensation model is verified by comparing the effect of error compensation on prediction accuracy through examples, and the superiority of the proposed prediction method is demonstrated by comparing it with other two prediction models, which can improve the accuracy of multivariate load prediction.

Key words: Integrated energy, load forecasting, long short-term memory neural network, gated recurrent units, error compensation

CLC Number:

TM715

GENG Yang, WANG Hailong, ZHANG Nan, FU Ming. Multivariate Load Forecasting for Integrated Energy Systems Based on Error-compensated LSTM-GRU[J]. Journal of Electrical Engineering, 2023, 18(4): 320-330.

Figures/Tables 16

References 20

[1]	宫飞翔, 李德智, 田世明, 等. 综合能源系统关键技术综述与展望[J]. 可再生能源, 2019, 37(8):1229-1235.
	GONG Feixiang, LI Dezhi, TIAN Shiming, et al. Overview and prospects of key technologies for integrated energy systems[J]. Renewable Energy, 2019, 37(8):1229-1235.
[2]	程浩忠, 胡枭, 王莉, 等. 区域综合能源系统规划研究综述[J]. 电力系统自动化, 2019, 43(7):2-13.
	CHENG Haozhong, HU Xiao, WANG Li, et al. A review of regional integrated energy system planning research[J]. Automation of Electric Power Systems, 2019, 43(7):2-13.
[3]	ZHANG Zhijun, YUE Dong, DOU Chunxia. DMPC based coordinated voltage control for integrated hybrid energy system[J]. IEEE Transactions on Industrial Informatics, 2021, 17(10):6786-6797. doi: 10.1109/TII.2020.3046633
[4]	朱继忠, 董瀚江, 李盛林, 等. 数据驱动的综合能源系统负荷预测综述[J]. 中国电机工程学报, 2021, 41(23):7905-7923.
	ZHU Jizhong, DONG Hanjiang, LI Shenglin, et al. Overview of data-driven comprehensive energy system load forecasting[J]. Proceedings of the CSEE, 2021, 41(23):7905-7923.
[5]	胡瑞琨, 吴煜宇, 郑豪丰, 等. 综合能源系统整体架构设计及负荷预测[J]. 宁夏电力, 2021(4):9-14,21.
	HU Ruikun, WU Yuyu, ZHENG Haofeng, et al. Integrated energy system overall architecture design and load forecasting[J]. Ningxia Electric Power, 2021(4):9-14,21.
[6]	段成俊. 区域综合能源系统负荷预测和优化调度研究[D]. 镇江: 江苏大学, 2020.
	DUAN Chengjun. Research on load forecast and optimal dispatch of regional integrated energy system[D]. Zhenjiang: Jiangsu University, 2020.
[7]	RENDON-SANCHEZ J F, DE MENEZES L M. Structural combination of seasonal exponential smoothing forecasts applied to load forecasting[J]. European Journal of Operational Research, 2019, 275(3):916-924. doi: 10.1016/j.ejor.2018.12.013
[8]	刘鑫, 滕欢, 宫毓斌, 等. 基于改进卡尔曼滤波算法的短期负荷预测[J]. 电测与仪表, 2019, 56(3):42-46.
	LIU Xin, TENG Huan, GONG Yubin, et al. Short-term load forecasting based on improved Kalman filter algorithm[J]. Electrical Measurement & Instrumentation, 2019, 56(3):42-46.
[9]	ZHANG Zhijun, ZHANG Yudi, YUE Dong, et al. Economic-driven hierarchical voltage regulation of incremental distribution networks:A cloud-edge collaboration based perspective[J]. IEEE Transactions on Industrial Informatics, 2022, 18(3):1746-1757. doi: 10.1109/TII.2021.3085670
[10]	WANG Yi, GAN Dahua, ZHANG Ning, et al. Feature selection for probabilistic load forecasting via sparse penalized quantile regression[J]. Journal of Modern Power Systems and Clean Energy, 2019, 7(5):1200-1209. doi: 10.1007/s40565-019-0552-3
[11]	DING Jia, WANG Maolin, PING Zuowei, et al. An integrated method based on relevance vector machine for short-term load forecasting[J]. European Journal of Operational Research, 2020, 287(2):497-510. doi: 10.1016/j.ejor.2020.04.007
[12]	谭风雷, 陈梦涛, 汪龙龙. 基于积温效应和优化支持向量机的短期电力负荷预测[J]. 电力需求侧管理, 2018, 20(5):33-36.
	TAN Fenglei, CHEN Mengtao, WANG Longlong. Short-term power load forecasting based on accumulated temperature effect and optimized support vector machine[J]. Power Demand Side Management, 2018, 20(5):33-36.
[13]	李焱, 贾雅君, 李磊, 等. 基于随机森林算法的短期电力负荷预测[J]. 电力系统保护与控制, 2020, 48(21):117-124.
	LI Yan, JIA Yajun, LI Lei, et al. Short-term power load forecasting based on random forest algorithm[J]. Power System Protection and Control, 2020, 48(21):117-124.
[14]	LI Jinghua, WEI Shanyang, DAI Wei. Combination of manifold learning and deep learning algorithms for mid-term electrical load forecasting[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(5):2584-2593. doi: 10.1109/TNNLS.2021.3106968
[15]	FARSI B, AMAYRI M, BOUGUILA N, et al. On short-term load forecasting using machine learning techniques and a novel parallel deep LSTM-CNN approach[J]. IEEE Access, 2021, 9:31191-31212. doi: 10.1109/Access.6287639
[16]	PARIZAD A, HATZIADONIU C. Deep learning algorithms and parallel distributed computing techniques for high-resolution load forecasting applying hyperparameter optimization[J]. IEEE Systems Journal, 2022, 16(3):3758-3769. doi: 10.1109/JSYST.2021.3130080
[17]	陈锦植, 刘海琼, 梁宏池, 等. 基于LSTM的配电网线路超短期负荷预测方法[J]. 能源工程, 2021(5):66-70.
	CHEN Jinzhi, LIU Haiqiong, LIANG Hongchi, et al. Super short-term load forecasting method of distribution network based on LSTM[J]. Energy Engineering, 2021(5):66-70.
[18]	朱凌建, 荀子涵, 王裕鑫, 等. 基于CNN-BiLSTM的短期电力负荷预测[J]. 电网技术, 2021, 45(11):4532-4539.
	ZHU Lingjian, XUN Zihan, WANG Yuxin, et al. Short-term power load forecasting based on CNN-BiLSTM[J]. Power System Technology, 2021, 45(11):4532-4539.
[19]	王克杰, 张瑞. 基于改进BP神经网络的短期电力负荷预测方法研究[J]. 电测与仪表, 2019, 56(24):115-121.
	WANG Kejie, ZHANG Rui. Research on short-term power load forecasting method based on improved BP neural network[J]. Electrical Measurement & Instrumentation, 2019, 56(24):115-121.
[20]	金山红, 朱想, 赫卫国, 等. 基于相空间重构ARIMA和SVR的母线净负荷预测[J]. 电力需求侧管理, 2018, 20(2):20-24.
	JIN Shanhong, ZHU Xiang, HE Weiguo, et al. Bus net load forecast based on phase space reconstruction ARIMA and SVR[J]. Power Demand Side Management, 2018, 20(2):20-24.

影响因素	电负荷	冷负荷	热负荷
电负荷	1	0.86	0.63
冷负荷	0.86	1	0.26
热负荷	0.63	0.26	1
温度	0.81	0.82	0.81
湿度	0.21	0.29	0.22
风速	0.05	0.03	0.09
太阳辐射	0.59	0.66	0.68

模型	负荷类型	工作日			休息日
模型	负荷类型	MAPE(%)	RMSE	ME	MAPE(%)	RMSE	ME
1	电负荷	0.95	431.8	270.5	1.19	502.1	327.4
	热负荷	0.80	93.9	74.1	1.01	111.1	88.9
	冷负荷	0.99	171.5	127.3	1.17	203.5	151.7
2	电负荷	0.95	429.5	268.0	1.19	491.5	316.7
	热负荷	0.79	92.5	73.0	0.97	108.2	85.6
	冷负荷	0.94	162.7	126.1	1.09	191.4	142.2
3	电负荷	1.03	461.6	282.9	1.28	545.4	332.3
	热负荷	0.85	99.6	77.9	1.02	116.3	93.6
	冷负荷	0.99	171.2	135.2	1.18	206.2	157.2
4	电负荷	1.24	508.3	323.4	1.47	586.7	383.1
	热负荷	0.93	104.1	85.8	1.08	124.6	101.8
	冷负荷	1.09	189.3	147.7	1.37	223.3	175.6

模型	负荷类型	工作日			休息日
模型	负荷类型	MAPE(%)	RMSE	ME	MAPE(%)	RMSE	ME
5	电负荷	0.95	429.5	268.0	1.19	491.5	316.7
	热负荷	0.79	92.5	73.0	0.97	108.2	85.6
	冷负荷	0.94	162.7	126.1	1.09	191.4	142.2
6	电负荷	2.00	781.2	565.2	2.38	856.3	663.6
	热负荷	1.32	151.4	121.8	1.57	183.5	144.2
	冷负荷	1.39	244.8	186.5	1.68	296.6	224.6
7	电负荷	2.96	995.6	836.7	3.48	1 188.1	993.3
	热负荷	2.66	276.8	229.3	3.18	328.6	275.5
	冷负荷	1.71	297.8	244.2	2.04	358.5	291.0

模型	负荷类型	工作日			休息日
模型	负荷类型	MAPE(%)	RMSE	ME	MAPE(%)	RMSE	ME
8	电负荷	1.92	668.3	421.9	2.31	792.4	509.5
	热负荷	1.62	209.3	113.4	1.94	245.3	134.9
	冷负荷	1.81	279.9	208.7	2.18	331.6	250.5
9	电负荷	0.95	429.5	268.0	1.19	491.5	316.7
	热负荷	0.79	92.5	73.0	0.97	108.2	85.6
	冷负荷	0.94	162.7	126.1	1.09	191.4	142.2

模型	负荷类型	工作日			休息日
模型	负荷类型	MAPE(%)	RMSE	ME	MAPE(%)	RMSE	ME
10	电负荷	0.95	429.5	268.0	1.19	491.5	316.7
	热负荷	0.79	92.5	73.0	0.97	108.2	85.6
	冷负荷	0.94	162.7	126.1	1.09	191.4	142.2
11	电负荷	0.98	453.0	284.9	1.21	548.4	344.7
	热负荷	0.99	113.8	89.0	1.18	134.1	105.7
	冷负荷	0.99	170.7	137.7	1.17	202.2	163.4
12	电负荷	1.33	573.6	360.3	1.60	613.5	424.0
	热负荷	1.06	125.8	100.4	1.26	151.2	121.6
	冷负荷	1.20	199.7	173.3	1.43	240.3	209.6