Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network

Shaohan Qi

doi:10.47297/taposatWSP2633-456923.20250609

PDF
导出
分享
收藏
专辑

Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network

6卷第9期

浏览量: 0 Downloads: 4

Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network
Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network
Hebei University of Technology, Tianjin, 300130, China
2025年6卷第9期页码：113-117
纸质出版：2025-09-30
DOI：10.47297/taposatWSP2633-456923.20250609
Accepted：

Scan QR Code

Shaohan Qi. (2025). Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network. 科技理论与实践（英文版）, 6(9), 113-117.

Qi Shaohan.Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network[J].Theory and Practice of Science and Technology,2025,06(09):113-117.
Shaohan Qi. (2025). Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network. 科技理论与实践（英文版）, 6(9), 113-117. DOI： 10.47297/taposatWSP2633-456923.20250609.

Qi Shaohan.Based on LSTM Study on Day-Ahead Power Load Forecasting in France Based on Neural Network[J].Theory and Practice of Science and Technology,2025,06(09):113-117. DOI： 10.47297/taposatWSP2633-456923.20250609.

摘要

Abstract

The French power system is facing challenges such as the increasing proportion of renewable energy

frequent extreme weather events

and increasingly complex cross-border power exchange. The accuracy of day-ahead load forecasting has become a key link in ensuring the security of the power grid and the efficient operation of the power market. This paper uses the actual load curve of France from January 1

2024 to January 1

2025 as a sample to construct a single-variable LSTM forecasting model and provides a complete MATLAB implementation process. The experimental results show that under the setting of 15-step input and 1-step output

the RMSE of the model test set is 0.398 GW

the MAE is 0.355 GW

and the R² is 0.943

which is significantly better than the traditional time series method. The paper further combines the French load characteristics with the code visualization results to demonstrate the necessity

advantages and disadvantages of the application of LSTM

and proposes future improvement directions.

关键词

Keywords

references

RTE (2025) French Electricity Load Open Data Platform. Available at: https://opendata.rte-france.com/pages/consumption.

ENTSO-E (2025) ENTSO-E Transparency Platform. Available at: https://transparency.entsoe.eu.

H2O.ai (2021) An Introduction to Time-Series Modeling: Traditional Time-Series Models and Their Limitations. Available at: https://h2o.ai/blog/2021/an-introduction-to-time-series-modeling-traditional-time-series-models-and-their-limitations.

Wiley Online Library (2025) Data Driven Long Short-Term Load Prediction: LSTM.

Wikipedia (2025) Long short-term memory. Available at: https://en.wikipedia.org/wiki/Long_short-term_memory.

Caicedo-Vivas, J.S. and Alfonso-Morales, W. (2023) 'Short-Term Load Forecasting Using an LSTM Neural Network for a Grid Operator', Energies, 16(23), p. 7878 .

MathWorks (2025) Sequence Input Layer and LSTM Layer for Regression. Available at: https://www.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.sequenceinputlayer.html.

Hochreiter, S. and Schmidhuber, J. (1997) 'Long short-term memory', Neural Computation, 9(8), pp. 1735 – 1780 .

Olah, C. (2015) Understanding LSTM Networks. colah's blog. Available at: https://colah.github.io/posts/2015-08-Understanding-LSTMs.

Wikipedia (2025) Réseau de Transport d'Électricité. Available at: https://en.wikipedia.org/wiki/R%C3%A9seau_de_Transport_d%27%C3%89lectricit%C3%A9.

Jing, J., Di, H., Wang, T. et al. (2025) 'Optimization of power system load forecasting and scheduling based on artificial neural networks', Energy Informatics, 8, p. 6 . Available at:

Liu, H., Fan, Z., Xie, H. and Wang, N. (2022) 'Distributionally Robust Joint Chance-Constrained Dispatch for Electricity–Gas–Heat Integrated Energy System Considering Wind Uncertainty', Energies, 15(5), p. 1796 .

文章被引用时，请邮件提醒。

Submit

Analysis of Fabric Selection Stability Model Based on Big Data Algorithm for Cultural Brand Building

Research and Case Study of Quantitative Investment Trading Strategy Model in Bitcoin and Gold

Related Author

Zhenghao Tong

Lu He

Da Rao

Related Institution

School of Art and Design, Fuzhou University of International Studies and Trade

Sichuan University

⁰