A novel short receptive field based dilated causal convolutional network integrated with Bidirectional LSTM for short-term load forecasting
作者：
Highlights：
• A novel hybrid Encoder-Decoder (ED) model is proposed for STLF problem.
• In ED model, the SRDCC and BiLSTM are cascaded to improve STLF accuracy.
• The proposed SRDCC-BiLSTM model mitigates overfitting issue of the STLF problem.
• A detailed comparative analysis of ED model is conducted with known ML and DL models.
• The computational efficiency and time complexity of ED model is computed and compared.
摘要
•A novel hybrid Encoder-Decoder (ED) model is proposed for STLF problem.•In ED model, the SRDCC and BiLSTM are cascaded to improve STLF accuracy.•The proposed SRDCC-BiLSTM model mitigates overfitting issue of the STLF problem.•A detailed comparative analysis of ED model is conducted with known ML and DL models.•The computational efficiency and time complexity of ED model is computed and compared.
论文关键词：Data analysis,Load forecasting,Learning (artificial intelligence),Machine learning,Power engineering computing,Time series analysis
论文评审过程：Received 24 February 2022, Revised 20 April 2022, Accepted 28 May 2022, Available online 4 June 2022, Version of Record 9 June 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117689
