Short-term power load forecasting method based on improved hierarchical transfer learning and multi-scale CNN-BiLSTM-Attention

被引:0
作者
Ouyang F. [1 ]
Wang J. [1 ]
Zhou H. [1 ]
机构
[1] College of Information Engineering, China Jiliang University, Hangzhou
来源
Dianli Xitong Baohu yu Kongzhi/Power System Protection and Control | 2023年 / 51卷 / 02期
关键词
attention mechanism; BiLSTM; CNN; power load forecasting; transfer learning;
D O I
10.19783/j.cnki.pspc.220422
中图分类号
TM7 [输配电工程、电力网及电力系统];
学科分类号
080802 ;
摘要
Insufficient power load data samples in the target domain result in inadequate model training and low prediction accuracy. Thus an improved hierarchical transfer learning strategy combined with a multi-scale CNN-BiLSTM-Attention model is proposed for short-term power load forecasting. A multi-scale CNN superimposed, linked and in parallel is designed as a feature extractor, and the features are then passed as input to two BiLSTM structures for further learning. Then an attention mechanism is introduced to adjust the weight of the captured information vector. This paper divides the layers according to the structure of the basic model, and inputs the source data into the model according to the level of the goodness-of-fit value to perform hierarchical transfer learning training. It then retains the optimal training weight of each layer, and uses the target domain data to carry out model training and obtain the final predictive model after fine-tuning. Experiments show that the proposed multi-scale CNN-BiLSTM-Attention model can effectively improve the accuracy of load prediction. When the load data samples are insufficient, the improved hierarchical transfer learning strategy can effectively reduce the prediction error compared with traditional transfer learning. Taking six months of data in the target domain as an example, compared with traditional transfer learning, the MAPE is reduced by 13.31%, MAE is reduced by 15.16%, and RMSE is reduced by 14.37%. © 2023 Power System Protection and Control Press. All rights reserved.
引用
收藏
页码:132 / 140
页数:8
相关论文
共 24 条
[1]  
FAN Shixiong, LI Lixin, WANG Songyan, Et al., Application analysis and exploration of artificial intelligence technology in power grid dispatch and control, Power System Technology, 44, 2, pp. 401-411, (2020)
[2]  
WU F, CATTANI C, SONG W Q, Et al., Fractional ARIMA with an improved cuckoo search optimization for the efficient short-term power load forecasting, AEJ-Alexandria Engineering Journal, 59, 5, pp. 3111-3118, (2020)
[3]  
WANG Y, ZHANG N, CHEN X, Et al., A short-term residential load forecasting model based on LSTM recurrent neural network considering weather features, Energies, 14, 10, (2021)
[4]  
LI Yan, JIA Yajun, LI Lei, Et al., Short term power load forecasting based on a stochastic forest algorithm, Power System Protection and Control, 48, 21, pp. 117-124, (2020)
[5]  
WU J R, WANG G W, TIAN Y C, Et al., Support vector regression with asymmetric loss for optimal electric load forecasting, Energy, (2021)
[6]  
JIANG Jian, LIU Haiqiong, LI Heng, Et al., Peak load forecasting method of distribution network lines based on XGBoost, Power System Protection and Control, 49, 16, pp. 119-127, (2021)
[7]  
HOU Hui, WANG Qing, ZHAO Bo, Et al., Power load forecasting without key information based on phase space reconstruction and machine learning, Power System Protection and Control, 50, 4, pp. 75-82, (2022)
[8]  
HU Y, QU BY, WANG J, Et al., Short-term load forecasting using multimodal evolutionary algorithm and random vector functional link network based ensemble learning, Applied Energy, (2021)
[9]  
KIM J, MOON J, HWANG E, Et al., Recurrent inception convolution neural network for multi short-term load forecasting, Energy and Buildings, 194, pp. 328-341, (2019)
[10]  
WEN Lulu, ZHOU Kaile, YANG Shanlin, Et al., Optimal load dispatch of community microgrid with deep learning based solar power and load forecasting, Energy, 171, pp. 1053-1065, (2019)