Abstract
The load forecast plays an important role in the secure and economic operations of power systems. Due to the increase of the multivariate heterogeneous equipment in the power system, it is difficult to determine the key characteristics of the sample data for the load prediction. Therefore, a Bi-directional long short-term memory network (BiLSTM) combined with attention mechanism is proposed to improve the accuracy of prediction. BiLSTM can grasp the global trend of sample data, and the attention mechanism can extract the key features of sample data. Specifically, this model firstly uses convolutional neural network (CNN) to extract local features from the load data. Finally, a load sample data collected from a certain region of China is used to train the model, and the test results show that the model has a better prediction accuracy compared with the existing models, e.g., long short-term memory neural network (LSTM) and CNN-LSTM.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Jeong, D., Park, C., et al.: Short-term electric load forecasting for buildings using logistic mixture vector autoregressive model with curve registration. Appl. Energy 282, 1–14 (2021)
Feng, Z., Bo, S., et al.: Cooling, heating and electrical load forecasting method for CCHP system based on multivariate phase space reconstruction and Kalman Filter. Proc. CSEE 36(2), 399–406 (2016). (in Chinses)
Fang, L., Yan, L., et al.: Short-term parallel prediction of electric power load Based on balanced KNN algorithm. Electr. Power 51(10), 88–94 (2018)
Meng, X., Shao, J.F.: A review of deep learning methods applied on load forecasting. Video Eng. 46(1), 44–51 (2022). (in Chinese)
Xiao, L., Xi, W., et al.: Short-term wind load forecasting based on improved LSSVM and error forecasting correction. Power Syst. Prot. Control 43(11), 63–69 (2015). (in Chinese)
Wang, D., Zhuo, Y., et al.: Short-term load forecasting based on chaotic characteristic of loads and least squares support vector machines. Power Syst. Technol. 32(7), 66–71 (2008). (in Chinese)
Li, G., Liu, Z., et al.: Ultra short-term power load forecasting based on randomly distributive embedded framework and BP neural network. Power Syst. Technol. 44(2), 437–444 (2020). (in Chinese)
Gao, A., Li, G., et al.: Short-term load forecasting model based on Adam algorithm to optimize GRU neural network. Electron. Des. Eng. 30(9), 180–183 (2022). (in Chinese)
Shi, Q ., Lei, W., et al.: Short-term load forecasting by long and short-term temporal networks with attention based on modal decomposition. Power Syst. Technol. 1–13 (2022). https://doi.org/10.13335/j.1000-3673.pst.2022.0368. (in Chinese)
Liu, J., Jin, Y., et al.: Multi-scale short–term load forecasting based On VMD and TNC. J. Univ. Electron. Sci. Technol. China 51(4), 550–557 (2022). (in Chinese)
Chen, Z., Sun, L., et al.: Short-term electrical load forecasting based on deep learning LSTM networks. Electron. Des. Appl. 47(01), 39–41 (2018). (in Chinese)
Huan, W., Peng, L., et al.: Long-term and short-term power load online prediction method based on CNN-BiLSTM. Comput. Simul. 39(3), 96–103 (2022). (in Chinese)
Zhao, B., Wang, Z., et al.: A short-term power load forecasting method based on attention mechanism of CNN-GRU. Power Syst. Technol. 43(12), 4370–4376 (2022). (in Chinese)
Lu, B., Huo, Z., et al.: Multi load ultra short-term forecasting of integrated energy system based on LSTNet-Skip. Proc. CSEE 1–12 (2022) (in Chinese)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 State Grid Electric Power
About this paper
Cite this paper
Lu, C., Shao, L., Feng, C., Hu, J. (2023). Short-Term Load Forecasting Model Based on BiLSTM and Attention Mechanism. In: Zeng, P., Zhang, XP., Terzija, V., Ding, Y., Luo, Y. (eds) The 37th Annual Conference on Power System and Automation in Chinese Universities (CUS-EPSA). CUS-EPSA 2022. Lecture Notes in Electrical Engineering, vol 1030. Springer, Singapore. https://doi.org/10.1007/978-981-99-1439-5_64
Download citation
DOI: https://doi.org/10.1007/978-981-99-1439-5_64
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-1438-8
Online ISBN: 978-981-99-1439-5
eBook Packages: EnergyEnergy (R0)