Electricity consumption prediction based on Deep AR neural network time-series model
1. School of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China; 2. School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China; 3. School of Mechanical and Automotive Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
Abstract：To predict the electricity consumption demand accurately, based on the data set of PJM company in the United States, the deep autoregressive recurrent networks (Deep AR) timeseries model was utilized to predict the electricity consumption of Commonwealth Edison Company at a certain 12hour interval in the future. Based on the distribution parameters of the data in the long short term memory network (LSTM), the predicted value was obtained by sampling in the distribution. Mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) were used as evaluation indexes for predicting short-term electricity consumption, and the model was compared with the time-series model of autoregressive integrated moving average (ARIMA) algorithm model and the Prophet algorithm model. The results show that the three performance indexes of MAE、RMSE and MAPE of the Deep AR algorithm model in predicting short-term electricity consumption are respective 1 070.01, 1 279.31 and 6.12% with high prediction accuracy. The proposed algorithm can not only predict electricity consumption in the future, but also can predict the probability distribution for further describing the globality of events.
KAYTEZ F. A hybrid approach based on autoregressive integrated moving average and leastsquare support vector machine for longterm forecasting of net electricity consumption[J]. Energy, doi: 10.1016/j.energy.2020.117200.
[2]
LIU X L, LIN Z, FENG Z M. Shortterm offshore wind speed forecast by seasonal ARIMAA comparison against GRU and LSTM[J]. Energy, doi.org/10.1016/j.energy.2021.120492.
ZHAO B J, LIU Q, FU Y X, et al. Boundary vorticity flux prediction of the centrifugal pump impeller based on artificial neural network[J]. Journal of Drainage and Irrigation Machinery Engineering, 2020,38(2):127-132.(in Chinese)
SHEN J, DOU W, XIANG P C, et al. Research on defrost control of cold storage cooling fan based on neural network model[J]. Fluid Machinery, 2020,48(4):72-77.(in Chinese)
CHEN Y J,LIU J Y, LI L, et al. Forecast and research on energy consumption based on analytical and combined model of BP neural network and regression[J]. Network and Information Technology, 2008, 27(11): 59-60.(in Chinese)
SONG J L, RUAN S Q, CHEN Y D, et al. Research on thermal performance of natural draft counterflow wet cooling tower based on artificial neural network[J]. Fluid Machinery, 2019,47(3):79-83.(in Chinese)
LEI L, WANG N, ZHENG H, et al. Prediction of supply air volume of variable air volume airconditioning based on deepconfidence neural network[J]. Fluid Machinery, 2021,49(3):85-90.(in Chinese)
[8]
程玉桂,戴爱明,黎明. 基于 BP 神经网络的城区基本用电负荷预测与验证[J]. 南昌航空大学学报(自然科学版), 2009, 23(3): 79-83.
CHENG Y G, DAI A M, LI M. Forecasts and verification of the basic electric load of city based on BP neural network[J]. Journal of Nanchang Hang Kong University(Natural Sciences), 2009, 23(3): 79-83.(in Chinese)
[9]
KIM N, KIM M, CHOI J K. LSTM based shortterm electricity consumption forecast with daily load profile sequences[C]∥2018 IEEE 7th Global Conference on Consumer Electronics. Piscataway, USA: IEEE, 2018: 136-137.
[10]
KANG K, SUN H B, ZHANG C K, et al. Shortterm electrical load forecasting method based on stacked autoencoding and GRU neural network[J]. Evolutionary Intelligence, 2019, 12: 385-394.
[11]
IBRAHIM M,ALSHEIKH A, AlHINDAWI Q, et al. Shorttime wind speed forecast using artificial learningbased algorithms[J]. Computational Intelligence and Neuroscience, doi:10.1155/2020/8439719.
[12]
SALINAS D, FLUNKERT V, GASTHAUS J, et al. Deep AR: probabilistic forecasting with autoregressive recurrent networks[J]. International Journal of Forecasting, 2020, 36: 1181-1191.
YUAN P, LEI Z L, ZENG Q H, et al. Research on load forecasting of refrigerated display cabinet based on data mining[J]. Fluid Machinery, 2021,49(5):87-90.(in Chinese)
[14]
LU D, ZHAO D B, LI Z Y. Shortterm nodal load forecasting based on machine learning techniques[J]. International Transactions on Electrical Energy Systems, doi:10.1002/2050-7038.13016.
