Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Journal of Electrotechnology, Electrical Engineering and Management
  4. Vol 6, Issue 3, 2023

Short-term load prediction based on Pearson-optimized CNN-LSTM hybrid neural network

Download as PDF

DOI: 10.23977/jeeem.2023.060303 | Downloads: 9 | Views: 409

Author(s)

Xu Chen 1, Xinying Liu 2

Affiliation(s)

1 Department of Electrical and Mechanical Engineering, Xinjiang Agricultural University, Urumqi, 830000, China
2 Department of Electrical and Control Engineering, Liaoning University of Engineering and Technology, Huludao, 125105, China

Corresponding Author

Xinying Liu

ABSTRACT

Under the demand of new power system construction, it is important to establish a solid and reliable power grid structure with stable operation by accelerating the construction of a "double high" strategy with the goal of "double carbon". Bus load can reflect the operation of the power grid, so bus load forecasting is important to maintain the safety and stability of the power system. To solve the problems of low accuracy and inefficiency of existing load forecasting methods for power systems, this paper adopts a combined CNN-LSTM load forecasting model with Pearson optimization, which is machine learning combined with deep learning. Firstly, Pearson correlation analysis is used for data processing to extract the main features of load data. Then three neural networks, CNN, LSTM, and CNN-LSTM, are used for training and load prediction, respectively. The experimental results show that the load prediction accuracy of the hybrid CNN-LSTM neural network prediction model based on Pearson optimization is higher than that of CNN and LSTM alone and matches with the actual value, which is a load prediction method with higher accuracy.

KEYWORDS

Load forecasting, Power systems, CNN, LSTM, Pearson correlation analysis

CITE THIS PAPER

Xu Chen, Xinying Liu, Short-term load prediction based on Pearson-optimized CNN-LSTM hybrid neural network. Journal of Electrotechnology, Electrical Engineering and Management (2023) Vol. 6: 17-25. DOI: http://dx.doi.org/10.23977/jeeem.2023.060303.

REFERENCES

[1] Zhu Haohao, Zhu Jizhong, Li Shenglin, et al. Fractional-order power system current calculation and voltage analysis [J]. Journal of Electrical Machinery and Control, 2022, 26(04):38-46.
[2] Weicong Kong, Zhao Yang Dong, Youwei Jia, et al. Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network. [J]. IEEE Trans. Smart Grid,2019,10(1).
[3] Sheng Hongying, Zhao Weiguo, Chen Yang, et al. Short-term load forecasting for power systems based on deep data mining[J]. Journal of Jilin University (Information Science Edition), 2023, 41(01):131-137.
[4] Zhang Zhiwang. Application of BP network optimized by genetic algorithm in power system load forecasting [J]. Electronic World, 2021(17): 168-169.
[5] Wang Zhuyao, Zhang Zhisheng. Short-term load forecasting for power systems based on quantum-weighted GRU neural networks [J]. Journal of Power Systems and Automation, 2022, 34(01):1-7. 
[6] Gao Ao, Li Guoyu, Password Aoyang,et al. Short-term load forecasting model based on optimized GRU neural network with Adam algorithm [J]. Electronic Design Engineering, 2022, 30(9): 180-183, 188.
[7] Yin Chunjie, Xiao Dafang, Li Pengfei, Zhao Qin. Short-term load forecasting for regional microgrid based on LSTM neural network [J]. Computers and Modernization,2022(04):7-11.
[8] Gan Jingfu, Yan Kun, Ma Minghan, He Pengkang, Tian Xincheng. Research on short-term load prediction by artificial neural network based on improved clustering algorithm [J]. New Technology of Electric Power,2022,41(09):40-46.
[9] Meng S, Shao JF. A review of deep learning-based electric load forecasting methods [J]. TV Technology, 2022, 46(01): 44-51.
[10] Fan Jiangchuan, Yu Haozheng, Liu Huiting, et al. Short-term electric load forecasting based on the multi-branch gated residual convolutional neural network [J]. China Electric Power,2022,55(11):155-162+174.
[11]  Shang C, Gao J, Liu H, et al. Short-term Load Forecasting Based on PSO-KFCM Daily Load Curve Clustering and CNN-LSTM Model[J]. IEEE Access, 2021, PP(99):1-1.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 2072
Visits: 97479

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.