Short-term prediction of electric demand in building sector via hybrid support vector regression

被引:108
作者
Chen, Yibo [1 ,2 ]
Tan, Hongwei [1 ,2 ,3 ]
机构
[1] Tongji Univ, Sch Mech Engn, Siping Rd 1239, Shanghai 200092, Peoples R China
[2] Tongji Univ, UNEP Tongji Inst Environm Sustainable Dev, Zonghe Bldg,Siping Rd 1239, Shanghai 200092, Peoples R China
[3] Tongji Univ, Res Ctr Green Bldg & New Energy, Ruian Bldg,Siping Rd 1239, Shanghai 200092, Peoples R China
关键词
Short-term prediction; Electric demand intensity; Commercial buildings; Support vector regression; Wavelet decomposition; COMMERCIAL BUILDINGS; WAVELET TRANSFORM; ENERGY-CONSUMPTION; OPTIMAL OPERATION; NEURAL-NETWORK; LOAD; MANAGEMENT; CONSUMERS; MODELS; SVR;
D O I
10.1016/j.apenergy.2017.03.070
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Reliable and highly-generalized prediction models of short-term electric demand are urgently needed for the building sector, as the crucial basis of sophisticated building energy management. Advances in metering technologies and machine learning methods provide both opportunities and challenges for modified approaches. With multi-resolution wavelet decomposition (MWD) as a preprocessing from the point of view of signal analysis, the hybrid support vector regression (SVR) model was applied in two case study buildings to predict the hourly electric demand intensity. Taking ten-dimensional parameters of 29 workdays as the training sample, this model was carried out in a mall and a hotel, the consumed electric demand sequences of which represented the stationary and non-stationary series respectively. By comparisons between the hybrid SVR and the pure SVR, results indicated that the introduction of MWD can always improve the predicting accuracy for the hotel, while it is not necessary for the mall. Specifically, the similar steady level around 0.65 W/m(2) of absolute error was obtained for the mall and the hotel buildings, when a was lower than 0.1. At the same time, the steady quantitative values of relative errors tended to be around 4% and 6% respectively for the hotel and the mall. Based on the limited historical readings, this paper offers an on-line prediction method of short-term electric demand, which is applicable for the further smart energy management. (C) 2017 Elsevier Ltd. All rights reserved.
引用
收藏
页码:1363 / 1374
页数:12
相关论文
共 35 条
[1]   Electricity consumption forecasting models for administration buildings of the UK higher education sector [J].
Amber, K. P. ;
Aslam, M. W. ;
Hussain, S. K. .
ENERGY AND BUILDINGS, 2015, 90 :127-136
[2]  
[Anonymous], 1999, WAVELET TOUR SIGNAL
[3]   Artificial neural network model for forecasting sub-hourly electricity usage in commercial buildings [J].
Chae, Young Tae ;
Horesh, Raya ;
Hwang, Youngdeok ;
Lee, Young M. .
ENERGY AND BUILDINGS, 2016, 111 :184-194
[4]   Short term load forecast using fuzzy logic and wavelet transform integrated generalized neural network [J].
Chaturvedi, D. K. ;
Sinha, A. P. ;
Malik, O. P. .
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, 2015, 67 :230-237
[5]   Neural network based short-term load forecasting using weather compensation [J].
Chow, TWS ;
Leung, CT .
IEEE TRANSACTIONS ON POWER SYSTEMS, 1996, 11 (04) :1736-1742
[6]  
CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023/A:1022627411411
[7]   Forecasting diurnal cooling energy load for institutional buildings using Artificial Neural Networks [J].
Deb, Chirag ;
Eang, Lee Siew ;
Yang, Junjing ;
Santamouris, Mattheos .
ENERGY AND BUILDINGS, 2016, 121 :284-297
[8]  
Dimitris L., 2014, RENEWABLE SUSTAINABL, V39, P587
[9]  
EPA, 2016, COMM BUILD EN CONS S
[10]   Development of prediction models for next-day building energy consumption and peak power demand using data Mining techniques [J].
Fan, Cheng ;
Xiao, Fu ;
Wang, Shengwei .
APPLIED ENERGY, 2014, 127 :1-10