FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination

Abstract Multivariate time series forecasting is widely used in various fields in real life. Many time series prediction models have been proposed. The current forecasting model lacks the mining of correlation between sequences based on different periods and correlation of periodical features betwee...

Full description

Saved in:
Bibliographic Details
Main Authors: WenChang Zhang, LuZhi Yuan, Yun Sha, LingLin Yang, XueJun Liu, Yong Yan
Format: Article
Language:English
Published: Wiley 2024-10-01
Series:Electronics Letters
Subjects:
Online Access:https://doi.org/10.1049/ell2.70069
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850265950610259968
author WenChang Zhang
LuZhi Yuan
Yun Sha
LingLin Yang
XueJun Liu
Yong Yan
author_facet WenChang Zhang
LuZhi Yuan
Yun Sha
LingLin Yang
XueJun Liu
Yong Yan
author_sort WenChang Zhang
collection DOAJ
description Abstract Multivariate time series forecasting is widely used in various fields in real life. Many time series prediction models have been proposed. The current forecasting model lacks the mining of correlation between sequences based on different periods and correlation of periodical features between different periods when dealing with data. In this paper, we propose a multivariate data prediction model FAFTransformer based on the reorganization of multi‐periodic features, which first extracts the multi‐periodic information of the time series using the method of frequency domain analysis. The temporal dependencies within sequences are then captured using convolution based on different periods, and the correlations between sequences are learned by combining the multivariate attention mechanism to obtain the intra‐sequence and inter‐sequence correlations under the same period. Finally, period fusion is proposed to capture the correlation of period characteristics between different periods. The experimental results show that the model achieves the best results on multiple datasets compared to the latest seven predictive models for time‐series data.
format Article
id doaj-art-b726912edbcf43fbacc677d5544f38f3
institution OA Journals
issn 0013-5194
1350-911X
language English
publishDate 2024-10-01
publisher Wiley
record_format Article
series Electronics Letters
spelling doaj-art-b726912edbcf43fbacc677d5544f38f32025-08-20T01:54:16ZengWileyElectronics Letters0013-51941350-911X2024-10-016020n/an/a10.1049/ell2.70069FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombinationWenChang Zhang0LuZhi Yuan1Yun Sha2LingLin Yang3XueJun Liu4Yong Yan5Information Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaInformation Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaInformation Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaInformation Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaInformation Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaInformation Engineering Institute Beijing Institute of Petrochemical Technology Beijing ChinaAbstract Multivariate time series forecasting is widely used in various fields in real life. Many time series prediction models have been proposed. The current forecasting model lacks the mining of correlation between sequences based on different periods and correlation of periodical features between different periods when dealing with data. In this paper, we propose a multivariate data prediction model FAFTransformer based on the reorganization of multi‐periodic features, which first extracts the multi‐periodic information of the time series using the method of frequency domain analysis. The temporal dependencies within sequences are then captured using convolution based on different periods, and the correlations between sequences are learned by combining the multivariate attention mechanism to obtain the intra‐sequence and inter‐sequence correlations under the same period. Finally, period fusion is proposed to capture the correlation of period characteristics between different periods. The experimental results show that the model achieves the best results on multiple datasets compared to the latest seven predictive models for time‐series data.https://doi.org/10.1049/ell2.70069artificial intelligencedata analysisneural net architecturePeriodic fusiontime series prediction
spellingShingle WenChang Zhang
LuZhi Yuan
Yun Sha
LingLin Yang
XueJun Liu
Yong Yan
FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
Electronics Letters
artificial intelligence
data analysis
neural net architecture
Periodic fusion
time series prediction
title FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
title_full FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
title_fullStr FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
title_full_unstemmed FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
title_short FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination
title_sort faftransformer multivariate time series prediction method based on multi period feature recombination
topic artificial intelligence
data analysis
neural net architecture
Periodic fusion
time series prediction
url https://doi.org/10.1049/ell2.70069
work_keys_str_mv AT wenchangzhang faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination
AT luzhiyuan faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination
AT yunsha faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination
AT linglinyang faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination
AT xuejunliu faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination
AT yongyan faftransformermultivariatetimeseriespredictionmethodbasedonmultiperiodfeaturerecombination