An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model

This paper proposes a hybrid forecasting method (VMD–KPCA–xLSTM–Informer) based on variational-mode decomposition (VMD), kernel principal component analysis (KPCA), extended long short-term memory network (xLSTM), and the Informer model. First, the method decomposes the original power load data and...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiawen You, Huafeng Cai, Dadian Shi, Liwei Guo
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/18/9/2240
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850279182300348416
author Jiawen You
Huafeng Cai
Dadian Shi
Liwei Guo
author_facet Jiawen You
Huafeng Cai
Dadian Shi
Liwei Guo
author_sort Jiawen You
collection DOAJ
description This paper proposes a hybrid forecasting method (VMD–KPCA–xLSTM–Informer) based on variational-mode decomposition (VMD), kernel principal component analysis (KPCA), extended long short-term memory network (xLSTM), and the Informer model. First, the method decomposes the original power load data and environmental parameter data using VMD to capture their multi-scale characteristics. Next, KPCA extracts nonlinear features and reduces the dimensionality of the decomposed modals to eliminate redundant information while retaining key features. The xLSTM network then models temporal dependencies to enhance the model’s memory capability and prediction accuracy. Finally, the Informer model processes long-sequence data to improve prediction efficiency. Experimental results demonstrate that the VMD–KPCA–xLSTM–Informer model achieves an average absolute percentage error (MAPE) as low as 2.432% and a coefficient of determination (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula>) of 0.9532 on dataset I, while, on dataset II, it attains a MAPE of 4.940% and an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.8897. These results confirm that the method significantly improves the accuracy and stability of short-term power load forecasting, providing robust support for power system optimization.
format Article
id doaj-art-af2bc2b6a3c24d36815451e3df59abbb
institution OA Journals
issn 1996-1073
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-af2bc2b6a3c24d36815451e3df59abbb2025-08-20T01:49:11ZengMDPI AGEnergies1996-10732025-04-01189224010.3390/en18092240An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer ModelJiawen You0Huafeng Cai1Dadian Shi2Liwei Guo3Normal School of Vocational Techniques, Hubei University of Technology, Wuhan 430068, ChinaSchool of Electrical and Electronic Engineering, Hubei University of Technology, Wuhan 430068, ChinaDetroit Green Technology Institute, Hubei University of Technology, Wuhan 430068, ChinaDetroit Green Technology Institute, Hubei University of Technology, Wuhan 430068, ChinaThis paper proposes a hybrid forecasting method (VMD–KPCA–xLSTM–Informer) based on variational-mode decomposition (VMD), kernel principal component analysis (KPCA), extended long short-term memory network (xLSTM), and the Informer model. First, the method decomposes the original power load data and environmental parameter data using VMD to capture their multi-scale characteristics. Next, KPCA extracts nonlinear features and reduces the dimensionality of the decomposed modals to eliminate redundant information while retaining key features. The xLSTM network then models temporal dependencies to enhance the model’s memory capability and prediction accuracy. Finally, the Informer model processes long-sequence data to improve prediction efficiency. Experimental results demonstrate that the VMD–KPCA–xLSTM–Informer model achieves an average absolute percentage error (MAPE) as low as 2.432% and a coefficient of determination (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula>) of 0.9532 on dataset I, while, on dataset II, it attains a MAPE of 4.940% and an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.8897. These results confirm that the method significantly improves the accuracy and stability of short-term power load forecasting, providing robust support for power system optimization.https://www.mdpi.com/1996-1073/18/9/2240short-term electricity load forecastingvariational-mode decompositionhybrid neural network
spellingShingle Jiawen You
Huafeng Cai
Dadian Shi
Liwei Guo
An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
Energies
short-term electricity load forecasting
variational-mode decomposition
hybrid neural network
title An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
title_full An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
title_fullStr An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
title_full_unstemmed An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
title_short An Improved Short-Term Electricity Load Forecasting Method: The VMD–KPCA–xLSTM–Informer Model
title_sort improved short term electricity load forecasting method the vmd kpca xlstm informer model
topic short-term electricity load forecasting
variational-mode decomposition
hybrid neural network
url https://www.mdpi.com/1996-1073/18/9/2240
work_keys_str_mv AT jiawenyou animprovedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT huafengcai animprovedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT dadianshi animprovedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT liweiguo animprovedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT jiawenyou improvedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT huafengcai improvedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT dadianshi improvedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel
AT liweiguo improvedshorttermelectricityloadforecastingmethodthevmdkpcaxlstminformermodel