Multi-Factor Short-Term Power Load Forecasting Based on Adversarial Training

doi:10.3969/j.issn.1008-0198.2025.04.020

Abstract

Abstract: In order to improve the accuracy and stability of power load forecasting, a multi-factor power load forecasting model based on adversarial training is proposed for short-term power load forecasting. This method combines historical load data and the weather and other characteristics of the forecast day to predict the power load and enhances the robustness of the prediction model to adversarial samples through adversarial training. Experimental results on a public dataset show that this method outperforms similar methods that only consider historical load data in terms of prediction accuracy and shows better robustness to adversarial samples.

Key words: power load forecasting, deep learning, adversarial sample, multi-factor analysis, adversarial training

CLC Number:

TM715.1

LI Xiaoping, HE Lubing, SHANG Longkang. Multi-Factor Short-Term Power Load Forecasting Based on Adversarial Training[J]. Hunan Electric Power, 2025, 45(4): 143-150.

References

[1] PARK C,HEO W G.Review of the changing electricity industry value chain in the ICT convergence era[J]. Journal of Cleaner Production,2020,258(4):120743.
[2] ALMESHAIEI E,SOLTAN H.A methodology for electric power load forecasting[J]. Alexandria Engineering Journal, 2011,50(2):137-144.
[3] PRABADEVI B,PHAM Q V,LIYANAGE M,et al.Deep learning for intelligent demand response and smart grids:a comprehensive survey[J]. Computer Science Review,2024,51:100617.
[4] MAHMUD M A.Isolated Area Load Forecasting using linear regression analysis:practical approach[J]. Energy and Power Engineering,2011,3(4):547-550.
[5] 唐祥玲,王平,李思岑,等. 基于方差-协方差组合预测的中长期电力负荷预测研究[J]. 电气技术,2015,16(1):15-18.
[6] 韩富尧,刘亚伟. 基于计量经济-灰色理论的多变量电力负荷预测方法[J]. 电气技术,2017,18(7):37-40.
[7] AHMAD A S,HASSAN M Y,ABDULLAH M P,et al.A review on applications of ANN and SVM for building electrical energy consumption forecasting[J]. Renewable and Sustainable Energy Reviews,2014,33:102-109.
[8] 吴润泽,包正睿,王文韬,等. Hadoop架构下基于模式匹配的短期电力负荷预测方法[J]. 电工技术学报,2018,33(7):1542-1551.
[9] 周华鑫. 基于PSOEM-LSSVM的中长期电力负荷预测及其应用研究[D]. 重庆:重庆大学,2013.
[10] 詹仁俊. 基于K-means聚类的小波支持向量机配电网短期负荷预测及应用[J]. 供用电,2019,36(4):64-70.
[11] QIN J,LIU H Z,MENG H F,et al.Robust dynamic economic dispatch in smart grids using an intelligent learning technology[J]. IEEE Transactions on Network Science and Engineering,2024,11(4):3759-3770.
[12] ZHANG W,LI M,WANG X M,et al.Study on the Mechanical Properties of Novel Composite Materials[J]. Journal of Materials Science,2023,58(12):4567-4579.
[13] 赵洋,王瀚墨,康丽,等. 基于时间卷积网络的短期电力负荷预测[J]. 电工技术学报,2022,37(5):1242-1251.
[14] 谷云东,张素杰,冯君淑. 大用户电力负荷的多模型模糊综合预测[J]. 电工技术学报,2015,30(23):110-115.
[15] 刘伟,王洪志. 基于改进注意力机制的时间卷积网络-长短期记忆网络短期电力负荷预测[J]. 电气技术,2024,25(10):8-14.
[16] 陈吕鹏,殷林飞,余涛,等. 基于深度森林算法的电力系统短期负荷预测[J]. 电力建设,2018,39(11):42-50.
[17] NGUYEN V B,DUONG M T,LE M H.Electricity demand forecasting for smart grid based on deep learning approach[C]//2020 5th International Conference on Green Technology and Sustainable Development(GTSD). Ho Chi Minh City,Vietnam. IEEE,2020:353-357.
[18] SON N,YANG S,NA J.Deep neural network and long short-term memory for electric power load forecasting[J].Applied Sciences,2020,10(18):6489.
[19] SHAO X R,KIM C S,SONTAKKE P,et al.Accurate deep model for electricity consumption forecasting using multi-channel and multi-scale feature fusion CNN-LSTM[J]. Energies,2020,13(8):1881.
[20] HUANG S T,SHEN J,LV Q Q,et al.A novel NODE approach combined with LSTM for short-term electricity load forecasting[J]. Future Internet,2023,15(1):22.
[21] YANG Y B,HAQ E U,JIA Y W.A novel deep learning approach for short and medium-term electrical load forecasting based on pooling LSTM-CNN model[C]//2020 IEEE/IAS Industrial and Commercial Power System Asia(I&CPS Asia). Weihai,China. IEEE,2020:26-34.
[22] 程定芳,任永建,陈正洪. 精细化气象因子对短期电力负荷预测的影响研究[J]. 华中师范大学学报(自然科学版),2020,54(5):792-797.
[23] 吴怡. 基于改进Transformer和时间卷积网络的电力负荷预测方法研究[D]. 秦皇岛:燕山大学,2023.
[24] ZACHARY C.(Deep Learning′s Deep Flaws)′s Deep Flaws[EB/OL]. (2015-01-26)[2025-02-22]. https://www.kdnuggets.com/2015/01/deep-learning-flaws-universal-machine-learning.html.
[25] SHAHAM U,YAMADA Y,NEGAHBAN S. Understanding adversarial training:Increasing local stability of neural nets through robust optimization[EB/OL].(2016-01-16)[2025-05-16]. https://arxiv.org/abs/1511.05432.
[26] 中国电机工程学会电工数学专委会. 第九届“中国电机工程学会杯”全国大学生电工数学建模竞赛题目[EB/OL].(2016-04-25)[2025-02-22]. http://shumo.nedu.edu.cn.
[27] 任建吉,位慧慧,邹卓霖,等. 基于CNN-BiLSTM-Attention的超短期电力负荷预测[J]. 电力系统保护与控制,2022,50(8):108-116.
[28] GOODFELLOW I J,SHLENS J,SZEGEDY C. Explaining and Harnessing Adversarial Examples[EB/OL]. (2015-03-20)[2025-05-16]. https://arxiv.org/abs/1412.6572.