Multi-layer Decomposition-Accumulation Principle-Based Monthly Electricity Consumption Forecasting Method for Urban Complex

doi:10.12204/j.issn.1000-7229.2021.02.004

Abstract

Abstract:

Traditional one-step monthly electricity consumption forecasting methods for urban complex cannot provide accurate forecasting results due to the serious challenge of model overfitting. In this paper, a monthly electricity forecasting method based on multi-layer decomposition-accumulation principle is proposed. The loads of urban complex are firstly subdivided into three categories according to their characteristics. Subsequently, for each category of load, the historical hourly electricity consumption data is collected and decomposed according to its week label so as to improve the forecasting performance of the multi-step forecasting model. Then, improved empirical mode decomposition (IEMD) algorithm is used to separate the fluctuations and trends of different scales in the electricity consumption series, and extreme gradient boosting algorithm (XGBoost) is utilized to establish multi-step forecasting model for each component. Finally, all of the forecasting results are accumulated to obtain monthly electricity consumption forecasting results. Results show that the proposed method can effectively capture the changes in the electricity consumption series. Its forecasting accuracy is improved by 18.2%-34.9% compared with the traditional method.

Key words: urban complex, multi-step forecast, improved empirical mode decomposition, extreme gradient boosting algorithm

CLC Number:

TM715

YANG Dezhou, WEI Yong, LI Wanwei, PENG Jing, LI Min, LI Zhenghui, ZHEN Zhao, WANG Fei. Multi-layer Decomposition-Accumulation Principle-Based Monthly Electricity Consumption Forecasting Method for Urban Complex[J]. ELECTRIC POWER CONSTRUCTION, 2021, 42(2): 27-34.

Figures/Tables 7

Fig.1 Typical daily load curves of three types of building loads in winter

Fig.2 Schematic diagram of historical hourly electricity consumption series decomposition

Fig.3 Decomposition results from the improved EMD algorithm

Fig.4 Forecasting model based on multi-layer decomposition-accumulation principle

Table 1 Parameters of XGBoost algorithm

Table 2 Forecasting performance metrics

Table 3 Forecasting results of monthly electricity consumption of urban complex

References 19

[1]	马宗国. 我国城市综合体发展途径探讨[J]. 城市发展研究, 2011,18(6):138-140.
	MA Zongguo. Study on development strategy of Chinese city complex[J]. Urban Studies, 2011,18(6):138-140.
[2]	刘贵文, 曹健宁. 城市综合体业态选择及组合比例[J]. 城市问题, 2010(5):41-45.
	LIU Guiwen, CAO Jianning. Trade selection and mix proportion of HOPSCA[J]. Urban Problems, 2010(5):41-45.
[3]	覃志凝, 叶红. TOD模式下东京城市综合体开发策略研究[J]. 智能建筑与智慧城市, 2020(3):24-26.
	QIN Zhining, YE Hong. Research on development strategy of Tokyo urban complex under TOD mode[J]. Intelligent Building & Smart City, 2020(3):24-26.
[4]	魏梅芳, 屈少青, 隆辉, 等. 城市综合体负荷特性建模与实例分析[J]. 中国电力, 2016,49(S1):116-122.
	WEI Meifang, QU Shaoqing, LONG Hui, et al. The modeling and analysis of load characteristics in urban complex[J]. Electric Power, 2016,49(S1):116-122.
[5]	张强, 王毅, 李鼎睿, 等. 基于X-12-ARIMA季节分解与年度电量校正的月度电量预测[J]. 电力建设, 2017,38(1):76-83.
	ZHANG Qiang, WANG Yi, LI Dingrui, et al. Monthly electricity forecast based on X-12-ARIMA seasonal decomposition and annual electricity correction[J]. Electric Power Construction, 2017,38(1):76-83.
[6]	宋建, 束洪春, 董俊, 等. 基于GM(1, 1)与BP神经网络的综合负荷预测[J]. 电力建设, 2020,41(5):75-80.
	SONG Jian, SHU Hongchun, DONG Jun, et al. Comprehensive load forecast based on GM(1, 1) and BP neural network[J]. Electric Power Construction, 2020,41(5):75-80.
[7]	赵腾, 王林童, 张焰, 等. 采用互信息与随机森林算法的用户用电关联因素辨识及用电量预测方法[J]. 中国电机工程学报, 2016,36(3):604-614. doi: 10.13334/j.0258-8013.pcsee.2016.03.003 URL
	ZHAO Teng, WANG Lintong, ZHANG Yan, et al. Relation factor identification of electricity consumption behavior of users and electricity demand forecasting based on mutual information and random forests[J]. Proceedings of the CSEE, 2016,36(3):604-614.
[8]	姚程文, 杨苹, 刘泽健. 基于CNN-GRU混合神经网络的负荷预测方法[J]. 电网技术, 2020,44(9):3416-3424.
	YAO Chengwen, YANG Ping, LIU Zejian. Load forecasting method based on CNN-GRU hybrid neural network[J]. Power System Technology, 2020,44(9):3416-3424.
[9]	ZHAO H X, MAGOULÈS F. A review on the prediction of building energy consumption[J]. Renewable and Sustainable Energy Reviews, 2012,16(6):3586-3592.
[10]	HONG W C, DONG Y C, ZHANG W Y, et al. Cyclic electric load forecasting by seasonal SVR with chaotic genetic algorithm[J]. International Journal of Electrical Power & Energy Systems, 2013,44(1):604-614.
[11]	GONZÁLEZ-ROMERA E, JARAMILLO-MORÁN M Á, CARMONA-FERNÁNDEZ D. Monthly electric energy demand forecasting based on trend extraction[J]. IEEE Transactions on Power Systems, 2006,21(4):1946-1953.
[12]	HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of The Royal Society A Mathematical Physical and Engineering Sciences, 1998,454(1971):903-995. doi: 10.1098/rspa.1998.0193 URL
[13]	李婉娉, 殷梓恒, 李鹏, 等. 基于改进Hilbert-Huang变换的谐波检测新方法[J]. 华东电力, 2013,41(3):636-640.
	LI Wanping, YIN Ziheng, LI Peng, et al. New method of harmonic detection based on the improved HHT[J]. East China Electric Power, 2013,41(3):636-640.
[14]	CHEN T Q, GUESTRIN C. XGBoost: A scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2016: 785-794.
[15]	LI K P, WANG F, MI Z Q, et al. Capacity and output power estimation approach of individual behind-the-meter distributed photovoltaic system for demand response baseline estimation[J]. Applied Energy, 2019,253:113595. doi: 10.1016/j.apenergy.2019.113595 URL
[16]	WANG F, XUAN Z M, ZHEN Z, et al. A day-ahead PV power forecasting method based on LSTM-RNN model and time correlation modification under partial daily pattern prediction framework[J]. Energy Conversion and Management, 2020,212:112766.
[17]	LU X X, LI K P, XU H C, et al. Fundamentals and business model for resource aggregator of demand response in electricity markets[J]. Energy, 2020,204:117885.
[18]	XUAN Z M, GAO X, LI K P, et al. PV-load decoupling based demand response baseline load estimation approach for residential customer with distributed PV system[J]. IEEE Transactions Industry Applications, 2020,56(6):6128-6137.
[19]	WANG F, XIANG B, LI K P, et al. Smart households’ aggregated capacity forecasting for load aggregators under incentive-based demand response programs[J]. IEEE Transactions on Industry Application, 2020,56(2):1086-1097.