A novel outlier calendrical heterogeneity reconstruction deep learning model for electricity demand forecasting

被引:0
|
作者
Huan Songhua [1 ,2 ]
机构
[1] Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
[2] Univ Chinese Acad Sci, Beijing, Peoples R China
关键词
Outlier reconstruction; deep learning; electricity demand; forecasting model; calendrical heterogeneity;
D O I
10.3233/JIFS-235218
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The development of an accurate electricity demand forecasting model is of paramount importance for promoting global energy efficiency and sustainability. Nonetheless, the presence of outliers and inappropriate model training can result in suboptimal performance. To tackle these challenges, this study explores the potential of Convolutional Neural Network (CNN) and active learning theory as forecasting solutions, offering high efficiency and advantages for long time series. In this study, a hybrid model that combines Isolation Forest (IF), Outlier Reconstruction (OR), CNN and Random Forest (RF) is conducted to mitigate computational complexity and enhance the accuracy of electricity demand forecasting in the presence of outliers. IF is employed to detect outliers in electricity demand time series, while OR is used to reconstruct subsequences based on calendrical heterogeneity for training. CNN is applied for both training and forecasting, and the final output is combined using RF. The effectiveness of the proposed IF-OR-CNN-RF model is validated using electricity data collected from recent sources in Australia at different sampling frequency. The experimental results demonstrate that, in comparison with other popular CNN-based electricity demand forecasting models, IF-OR-CNN-RF model outperforms with significantly improved performance metrics. Specifically, Mean Absolute Error (MAE), Root Mean Squared Error (RMSE) and R-squared values are 77.92, 179.18 and 0.9769 in 5-minute frequency; 162.67, 353.96 and 0.9775 in 10-minute frequency; 841.27, 1374.79 and 0.9622 in 30-minute frequency; 2746.01, 3824.00 and 0.9262 in 60-minute frequency; 9106.08, 12269.04 and 0.8044 in 120-minute frequency. IF-OR-CNN-RF model represents a valuable framework for future electricity demand forecasting, particularly in scenarios involving outliers.
引用
收藏
页码:3363 / 3394
页数:32
相关论文
共 50 条
  • [1] Deep Learning for Forecasting Electricity Demand in Taiwan
    Yang, Cheng-Hong
    Chen, Bo-Hong
    Wu, Chih-Hsien
    Chen, Kuo-Chang
    Chuang, Li-Yeh
    MATHEMATICS, 2022, 10 (14)
  • [2] Electricity Demand Forecasting with a Novel Hybrid Model
    Wu, L. P.
    Yang, Q. S.
    Du, G. Q.
    Wang, J. H.
    INTERNATIONAL CONFERENCE ON AUTOMATION, MECHANICAL AND ELECTRICAL ENGINEERING (AMEE 2015), 2015, : 655 - 662
  • [3] A novel deep learning framework with a COVID-19 adjustment for electricity demand forecasting
    Cui, Zhesen
    Wu, Jinran
    Lian, Wei
    Wang, You-Gan
    ENERGY REPORTS, 2023, 9 : 1887 - 1895
  • [4] Forecasting electricity demand by hybrid machine learning model
    Fan, Shu
    Mao, Chengxiong
    Zhang, Jiadong
    Chen, Luonan
    NEURAL INFORMATION PROCESSING, PT 2, PROCEEDINGS, 2006, 4233 : 952 - 963
  • [5] Empirical Mode Decomposition Based Deep Learning for Electricity Demand Forecasting
    Bedi, Jatin
    Toshniwal, Durga
    IEEE ACCESS, 2018, 6 : 49144 - 49156
  • [6] An Improved Deep Learning Model for Electricity Price Forecasting
    Iqbal, Rashed
    Mokhlis, Hazlie
    Khairuddin, Anis Salwa Mohd
    Muhammad, Munir Azam
    INTERNATIONAL JOURNAL OF INTERACTIVE MULTIMEDIA AND ARTIFICIAL INTELLIGENCE, 2024, 9 (01):
  • [7] A Hybrid Deep Learning Forecasting Model Using GPU Disaggregated Function Evaluations Applied For Household Electricity Demand Forecasting
    Coelho, Vitor N.
    Coelho, Igor M.
    Rios, Eyder
    Filho, Alexandre S. T.
    Reis, Agnaldo J. R.
    Coelho, Bruno N.
    Alves, Alysson
    Netto, Guilherme G.
    Souza, Marcone J. F.
    Guimaraes, Frederico G.
    PROCEEDINGS OF RENEWABLE ENERGY INTEGRATION WITH MINI/MICROGRID (REM2016), 2016, 103 : 280 - 285
  • [8] A Novel Robust Grey Model for Forecasting Chinese Electricity Demand
    Yao, Riquan
    Jin, Shaojun
    Wei, Cong
    Kong, Jiayang
    DISCRETE DYNAMICS IN NATURE AND SOCIETY, 2022, 2022
  • [9] A novel approach for electricity demand forecasting
    Li, Caihong
    Mub, Senhui
    Wang, Jianzhou
    Yang, Yi
    Li, Lian
    IEICE NONLINEAR THEORY AND ITS APPLICATIONS, 2014, 5 (02): : 184 - 197
  • [10] A Deep Learning Approach to Forecasting Monthly Demand for Residential-Sector Electricity
    Son, Hyojoo
    Kim, Changwan
    SUSTAINABILITY, 2020, 12 (08)