A federated LSTM network for load forecasting using multi-source data with homomorphic encryption

Short-term load forecasting is of great significance to the operation of power systems. Various uncertain factors, such as meteorological social data, have already been combined with historical power data to create more accurate load forecasting models. In traditional systems, data from various indu...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengdi Wang, Rui Xin, Mingrui Xia, Zhifeng Zuo, Yinyin Ge, Pengfei Zhang, Hongxing Ye
Format: Article
Language:English
Published: AIMS Press 2025-03-01
Series:AIMS Energy
Subjects:
Online Access:https://www.aimspress.com/article/doi/10.3934/energy.2025011
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849730182019022848
author Mengdi Wang
Rui Xin
Mingrui Xia
Zhifeng Zuo
Yinyin Ge
Pengfei Zhang
Hongxing Ye
author_facet Mengdi Wang
Rui Xin
Mingrui Xia
Zhifeng Zuo
Yinyin Ge
Pengfei Zhang
Hongxing Ye
author_sort Mengdi Wang
collection DOAJ
description Short-term load forecasting is of great significance to the operation of power systems. Various uncertain factors, such as meteorological social data, have already been combined with historical power data to create more accurate load forecasting models. In traditional systems, data from various industries and regions are centralized for knowledge extraction. However, concerns regarding data security and privacy often prevent industries from sharing their data, limiting both the quantity and diversity of data available for forecasting models. These challenges drive the adoption of federated learning (FL) to address issues related to data silos and privacy. In this paper, a novel framework for short-term load forecasting was proposed using historical data from industries such as power, meteorology, and finance. Long short-term memory (LSTM) networks were utilized for forecasting, and federated learning (FL) was implemented to protect data privacy. FL allows clients in multiple regions to collaboratively train a shared model without exposing their data. To further enhance security, the homomorphic encryption (HE) using Paillier algorithm was introduced during the federated process. Experimental results demonstrate that the federated model, which extracts knowledge from different regions, outperforms locally trained models. Furthermore, longer HE keys have little effect on predictive performance but significantly slow down encryption and decryption, thereby increasing training time.
format Article
id doaj-art-77266cfd2ca84882935299b898f76314
institution DOAJ
issn 2333-8334
language English
publishDate 2025-03-01
publisher AIMS Press
record_format Article
series AIMS Energy
spelling doaj-art-77266cfd2ca84882935299b898f763142025-08-20T03:08:57ZengAIMS PressAIMS Energy2333-83342025-03-0113226528910.3934/energy.2025011A federated LSTM network for load forecasting using multi-source data with homomorphic encryptionMengdi Wang0Rui Xin1Mingrui Xia2Zhifeng Zuo3Yinyin Ge4Pengfei Zhang5Hongxing Ye6State Grid Hebei Electric Power Co., Ltd. Information and Communication Branch, Zhongtong City Square, Qiaoxi District, Shijiazhuang City, Hebei Province, ChinaState Grid Hebei Electric Power Co., Ltd. Information and Communication Branch, Zhongtong City Square, Qiaoxi District, Shijiazhuang City, Hebei Province, ChinaXi'an Jiaotong University, No. 28, Xianning West Road Xi'an, Shaanxi Province, ChinaXi'an Jiaotong University, No. 28, Xianning West Road Xi'an, Shaanxi Province, ChinaXi'an Jiaotong University, No. 28, Xianning West Road Xi'an, Shaanxi Province, ChinaState Grid Hebei Electric Power Co., Ltd. Information and Communication Branch, Zhongtong City Square, Qiaoxi District, Shijiazhuang City, Hebei Province, ChinaXi'an Jiaotong University, No. 28, Xianning West Road Xi'an, Shaanxi Province, ChinaShort-term load forecasting is of great significance to the operation of power systems. Various uncertain factors, such as meteorological social data, have already been combined with historical power data to create more accurate load forecasting models. In traditional systems, data from various industries and regions are centralized for knowledge extraction. However, concerns regarding data security and privacy often prevent industries from sharing their data, limiting both the quantity and diversity of data available for forecasting models. These challenges drive the adoption of federated learning (FL) to address issues related to data silos and privacy. In this paper, a novel framework for short-term load forecasting was proposed using historical data from industries such as power, meteorology, and finance. Long short-term memory (LSTM) networks were utilized for forecasting, and federated learning (FL) was implemented to protect data privacy. FL allows clients in multiple regions to collaboratively train a shared model without exposing their data. To further enhance security, the homomorphic encryption (HE) using Paillier algorithm was introduced during the federated process. Experimental results demonstrate that the federated model, which extracts knowledge from different regions, outperforms locally trained models. Furthermore, longer HE keys have little effect on predictive performance but significantly slow down encryption and decryption, thereby increasing training time.https://www.aimspress.com/article/doi/10.3934/energy.2025011short-term load forecastingfederated learninglong short-term memoryhomomorphic encryptionsmart grid
spellingShingle Mengdi Wang
Rui Xin
Mingrui Xia
Zhifeng Zuo
Yinyin Ge
Pengfei Zhang
Hongxing Ye
A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
AIMS Energy
short-term load forecasting
federated learning
long short-term memory
homomorphic encryption
smart grid
title A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
title_full A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
title_fullStr A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
title_full_unstemmed A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
title_short A federated LSTM network for load forecasting using multi-source data with homomorphic encryption
title_sort federated lstm network for load forecasting using multi source data with homomorphic encryption
topic short-term load forecasting
federated learning
long short-term memory
homomorphic encryption
smart grid
url https://www.aimspress.com/article/doi/10.3934/energy.2025011
work_keys_str_mv AT mengdiwang afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT ruixin afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT mingruixia afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT zhifengzuo afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT yinyinge afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT pengfeizhang afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT hongxingye afederatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT mengdiwang federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT ruixin federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT mingruixia federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT zhifengzuo federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT yinyinge federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT pengfeizhang federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption
AT hongxingye federatedlstmnetworkforloadforecastingusingmultisourcedatawithhomomorphicencryption