A Two-Stage Short-Term Load Forecasting Method Using Long Short-Term Memory and Multilayer Perceptron

被引:16
|
作者
Xie, Yuhong [1 ]
Ueda, Yuzuru [2 ]
Sugiyama, Masakazu [1 ]
机构
[1] Univ Tokyo, Res Ctr Adv Sci & Technol, Sch Engn, Meguro Ku, 4-6-1 Komaba, Tokyo 1538904, Japan
[2] Tokyo Univ Sci, Sch Engn, Katsushika Ku, 6-3-1 Niijuku, Tokyo 1258585, Japan
关键词
short-term load forecast; hybrid model; long short-term memory; multilayer perceptron; sequence-to-sequence; ARTIFICIAL NEURAL-NETWORK; HYBRID;
D O I
10.3390/en14185873
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
Load forecasting is an essential task in the operation management of a power system. Electric power companies utilize short-term load forecasting (STLF) technology to make reasonable power generation plans. A forecasting model with low prediction errors helps reduce operating costs and risks for the operators. In recent years, machine learning has become one of the most popular technologies for load forecasting. In this paper, a two-stage STLF model based on long short-term memory (LSTM) and multilayer perceptron (MLP), which improves the forecasting accuracy over the entire time horizon, is proposed. In the first stage, a sequence-to-sequence (seq2seq) architecture, which can handle a multi-sequence of input to extract more features of historical data than that of single sequence, is used to make multistep predictions. In the second stage, the MLP is used for residual modification by perceiving other information that the LSTM cannot. To construct the model, we collected the electrical load, calendar, and meteorological records of Kanto region in Japan for four years. Unlike other LSTM-based hybrid architectures, the proposed model uses two independent neural networks instead of making the neural network deeper by concatenating a series of LSTM cells and convolutional neural networks (CNNs). Therefore, the proposed model is easy to be trained and more interpretable. The seq2seq module performs well in the first few hours of the predictions. The MLP inherits the advantage of the seq2seq module and improves the results by feeding artificially selected features both from historical data and information of the target day. Compared to the LSTM-AM model and single MLP model, the mean absolute percentage error (MAPE) of the proposed model decreases from 2.82% and 2.65% to 2%, respectively. The results demonstrate that the MLP helps improve the prediction accuracy of seq2seq module and the proposed model achieves better performance than other popular models. In addition, this paper also reveals the reason why the MLP achieves the improvement.
引用
收藏
页数:17
相关论文
共 50 条
  • [1] Short-Term Load Forecasting using A Long Short-Term Memory Network
    Liu, Chang
    Jin, Zhijian
    Gu, Jie
    Qiu, Caiming
    [J]. 2017 IEEE PES INNOVATIVE SMART GRID TECHNOLOGIES CONFERENCE EUROPE (ISGT-EUROPE), 2017,
  • [2] Short-term Load Forecasting with Distributed Long Short-Term Memory
    Dong, Yi
    Chen, Yang
    Zhao, Xingyu
    Huang, Xiaowei
    [J]. 2023 IEEE POWER & ENERGY SOCIETY INNOVATIVE SMART GRID TECHNOLOGIES CONFERENCE, ISGT, 2023,
  • [3] A Two-Stage Random Forest Method for Short-term Load Forecasting
    Wu, Xiaoyu
    He, Jinghan
    Yip, Tony
    Lu, Jian
    Lu, Ning
    [J]. 2016 IEEE POWER AND ENERGY SOCIETY GENERAL MEETING (PESGM), 2016,
  • [4] Short-term load forecasting using a two-stage sarimax model
    Tarsitano, Agostino
    Amerise, Ilaria L.
    [J]. ENERGY, 2017, 133 : 108 - 114
  • [5] Integrating Long Short-Term Memory and Genetic Algorithm for Short-Term Load Forecasting
    Santra, Arpita Samanta
    Lin, Jun-Lin
    [J]. ENERGIES, 2019, 12 (11)
  • [6] Short-term power load forecasting using integrated methods based on long short-term memory
    Zhang, WenJie
    Qin, Jian
    Mei, Feng
    Fu, JunJie
    Dai, Bo
    Yu, WenWu
    [J]. SCIENCE CHINA-TECHNOLOGICAL SCIENCES, 2020, 63 (04) : 614 - 624
  • [7] Short-term power load forecasting using integrated methods based on long short-term memory
    ZHANG WenJie
    QIN Jian
    MEI Feng
    FU JunJie
    DAI Bo
    YU WenWu
    [J]. Science China(Technological Sciences), 2020, 63 (04) - 624
  • [8] Short-term power load forecasting using integrated methods based on long short-term memory
    ZHANG WenJie
    QIN Jian
    MEI Feng
    FU JunJie
    DAI Bo
    YU WenWu
    [J]. Science China Technological Sciences, 2020, (04) : 614 - 624
  • [9] Short-term power load forecasting using integrated methods based on long short-term memory
    WenJie Zhang
    Jian Qin
    Feng Mei
    JunJie Fu
    Bo Dai
    WenWu Yu
    [J]. Science China Technological Sciences, 2020, 63 : 614 - 624
  • [10] Long Short Term Memory Networks for Short-Term Electric Load Forecasting
    Narayan, Apurva
    Hipel, Keith W.
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), 2017, : 2573 - 2578