A new method for short-term load forecasting based on weighted fusion of base models

•Expanding economic and climate features as inputs to enhance the model’s ability to capture load changes.•Introducing singular value decomposition technology to remove noise from load data, while preserving key features and trends, to improve data quality.•Select multiple typical models as sub mode...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Applied thermal engineering Ročník 280; s. 128313
Hlavní autori: Bian, Kai, Xu, Jing, Wang, Haohao, Ding, Junfeng
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 01.12.2025
Predmet:
ISSN:1359-4311
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract •Expanding economic and climate features as inputs to enhance the model’s ability to capture load changes.•Introducing singular value decomposition technology to remove noise from load data, while preserving key features and trends, to improve data quality.•Select multiple typical models as sub model samples through the system, and choose the three with the best performance as the research focus.•Using the SFOA optimization algorithm to optimize the weights of the combined model and achieve higher prediction accuracy. In recent years, integrated energy systems have developed rapidly, integrating electricity, heat, cooling, and other energy sources into cross-energy coupling systems for efficient energy optimization. The setting and dynamic adjustment of their operating parameters depend heavily on power load forecasting, so improving forecasting accuracy is key to optimizing system operation. However, the load data is affected by interference from non-stationary signals and the introduction of a significant amount of noise. The traditional single-model approach struggles to handle complex and variable forecasting tasks, failing to meet the high-precision load forecasting demands. Given such problems, this research proposes a combined forecasting model utilizing SVD denoising preprocessing and multi-base model fusion to address the limitations of single-model approaches and enhance load forecasting precision. We used one year of load data from an Indian power plant to verify our proposed model. Firstly, SVD preprocesses the original data to remove noise and abnormal fluctuations. Unlike common modal decomposition methods that are prone to causing modal aliasing, SVD has stronger universality for nonlinear and non-stationary power load data and avoids such defects. Secondly, a weighted combination model integrating three base models is constructed, with SFOA searching for the minimum error weight to determine the optimal combination weight. The results showed that the MSE of the single BP model is 2.3709 MW, while the combined model proposed is just 1.2764 MW, which is 46 % lower. This intuitively proves the advantages of the combined model, providing data support for integrated energy system development and optimal resource allocation.
AbstractList •Expanding economic and climate features as inputs to enhance the model’s ability to capture load changes.•Introducing singular value decomposition technology to remove noise from load data, while preserving key features and trends, to improve data quality.•Select multiple typical models as sub model samples through the system, and choose the three with the best performance as the research focus.•Using the SFOA optimization algorithm to optimize the weights of the combined model and achieve higher prediction accuracy. In recent years, integrated energy systems have developed rapidly, integrating electricity, heat, cooling, and other energy sources into cross-energy coupling systems for efficient energy optimization. The setting and dynamic adjustment of their operating parameters depend heavily on power load forecasting, so improving forecasting accuracy is key to optimizing system operation. However, the load data is affected by interference from non-stationary signals and the introduction of a significant amount of noise. The traditional single-model approach struggles to handle complex and variable forecasting tasks, failing to meet the high-precision load forecasting demands. Given such problems, this research proposes a combined forecasting model utilizing SVD denoising preprocessing and multi-base model fusion to address the limitations of single-model approaches and enhance load forecasting precision. We used one year of load data from an Indian power plant to verify our proposed model. Firstly, SVD preprocesses the original data to remove noise and abnormal fluctuations. Unlike common modal decomposition methods that are prone to causing modal aliasing, SVD has stronger universality for nonlinear and non-stationary power load data and avoids such defects. Secondly, a weighted combination model integrating three base models is constructed, with SFOA searching for the minimum error weight to determine the optimal combination weight. The results showed that the MSE of the single BP model is 2.3709 MW, while the combined model proposed is just 1.2764 MW, which is 46 % lower. This intuitively proves the advantages of the combined model, providing data support for integrated energy system development and optimal resource allocation.
ArticleNumber 128313
Author Xu, Jing
Bian, Kai
Wang, Haohao
Ding, Junfeng
Author_xml – sequence: 1
  givenname: Kai
  surname: Bian
  fullname: Bian, Kai
  organization: State Key Laboratory of Digital Intelligent Technology for Unmanned Coal Mining, Anhui University of Science and Technology, Huainan 232001 Anhui, China
– sequence: 2
  givenname: Jing
  orcidid: 0009-0003-1489-966X
  surname: Xu
  fullname: Xu, Jing
  email: xuj053024@163.com
  organization: State Key Laboratory of Digital Intelligent Technology for Unmanned Coal Mining, Anhui University of Science and Technology, Huainan 232001 Anhui, China
– sequence: 3
  givenname: Haohao
  surname: Wang
  fullname: Wang, Haohao
  organization: State Key Laboratory of Digital Intelligent Technology for Unmanned Coal Mining, Anhui University of Science and Technology, Huainan 232001 Anhui, China
– sequence: 4
  givenname: Junfeng
  surname: Ding
  fullname: Ding, Junfeng
  organization: State Key Laboratory of Digital Intelligent Technology for Unmanned Coal Mining, Anhui University of Science and Technology, Huainan 232001 Anhui, China
BookMark eNqNkDtPwzAUhT0UibbwHzywJviRlyWWqqKAVImlzJZj3zSukriyDRX_HrdlYWO6j6NzdO-3QLPJTYDQAyU5JbR6POTqeBxiD35UA0z7nBFW5pQ1nPIZmlNeiqzglN6iRQgHQpJSF3O0W-EJTniE2DuDO-dx6J2PWUw5eHDqsgOtQrTTHrcqgMFuwiew-z6mvvsMNs2uu2h4dAaGcIduOjUEuP-tS_Sxed6tX7Pt-8vberXNNKt5zArGKaOKtoUgApqWlZS1HS2hVm2tuDJC8KYQja4IEU3XsLZipqyJYEy3pdF8iZ6uudq7EDx08ujtqPy3pESeqciD_EtFnqnIK5Vk31zt6WT4suBl0BYmDcaml6M0zv4v6Ac--3fg
Cites_doi 10.1016/j.applthermaleng.2025.125799
10.1016/j.jhydrol.2022.128463
10.70470/EDRAAK/2024/005
10.1016/j.engappai.2024.108375
10.1016/j.egyr.2022.09.188
10.1016/j.apenergy.2022.119269
10.1016/j.cie.2024.110114
10.1016/j.rineng.2024.103560
10.1016/j.energy.2025.134757
10.1016/j.enbuild.2022.111908
10.1007/s00521-024-10694-1
10.1016/j.apenergy.2024.124166
10.1016/j.eswa.2023.122012
10.1016/j.cma.2024.117718
10.1016/j.enconman.2020.112534
10.1016/j.applthermaleng.2025.127031
10.1016/j.apenergy.2022.118801
10.1016/j.energy.2023.128274
10.1016/j.enbuild.2023.113396
10.1016/j.asoc.2019.105548
10.1109/TNNLS.2023.3259149
10.3390/app12136647
10.1016/j.enbuild.2022.112666
10.1016/j.applthermaleng.2024.125224
10.1007/s00521-021-05958-z
10.1016/j.egyr.2023.05.048
10.1162/neco_a_01199
10.58496/BJAI/2024/010
10.1016/j.ref.2023.04.010
10.1007/s11063-024-11622-z
10.1016/j.apenergy.2021.118197
10.1016/j.energy.2022.124041
10.1007/s11042-022-13462-2
10.3390/su14127307
10.1016/j.enbuild.2024.114698
10.1080/15325008.2023.2181883
10.1016/j.engappai.2023.106773
10.1109/TDEI.2023.3269725
ContentType Journal Article
Copyright 2025 Elsevier Ltd
Copyright_xml – notice: 2025 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.applthermaleng.2025.128313
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
ExternalDocumentID 10_1016_j_applthermaleng_2025_128313
S1359431125029059
GroupedDBID --K
--M
.~1
0R~
1B1
1RT
1~.
1~5
23M
4.4
457
4G.
5GY
5VS
7-5
71M
8P~
AABNK
AAEDT
AAEDW
AAHCO
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AARJD
AATTM
AAXKI
AAXUO
AAYWO
ABFNM
ABJNI
ABMAC
ABNUV
ACDAQ
ACGFS
ACIWK
ACLOT
ACRLP
ACVFH
ADBBV
ADCNI
ADEWK
ADEZE
ADTZH
AEBSH
AECPX
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFTJW
AGHFR
AGUBO
AGYEJ
AHIDL
AHJVU
AHPOS
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKURH
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
APXCP
AXJTR
BELTK
BJAXD
BKOJK
BLXMC
CS3
EBS
EFJIC
EFKBS
EFLBG
ENUVR
EO8
EO9
EP2
EP3
FDB
FEDTE
FIRID
FNPLU
FYGXN
G-Q
GBLVA
HVGLF
IHE
J1W
JARJE
JJJVA
KOM
M41
MO0
MS~
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSG
SSR
SST
SSZ
T5K
TN5
~G-
~HD
9DU
AAQXK
AAYXX
ABWVN
ABXDB
ACNNM
ACRPL
ADMUD
ADNMO
AGQPQ
ASPBG
AVWKF
AZFZN
CITATION
EJD
FGOYB
HZ~
R2-
ID FETCH-LOGICAL-c273t-423121a1b4909e8b2512bf15e7ab7a3ad9938498c60098f82b62d570922cb5dc3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001573266800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1359-4311
IngestDate Sat Nov 29 06:51:16 EST 2025
Wed Dec 10 14:23:08 EST 2025
IsPeerReviewed true
IsScholarly true
Keywords Combined model
Feature selection
Data denoising
Short-term load forecasting
Parameter optimization
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c273t-423121a1b4909e8b2512bf15e7ab7a3ad9938498c60098f82b62d570922cb5dc3
ORCID 0009-0003-1489-966X
ParticipantIDs crossref_primary_10_1016_j_applthermaleng_2025_128313
elsevier_sciencedirect_doi_10_1016_j_applthermaleng_2025_128313
PublicationCentury 2000
PublicationDate 2025-12-01
2025-12-00
PublicationDateYYYYMMDD 2025-12-01
PublicationDate_xml – month: 12
  year: 2025
  text: 2025-12-01
  day: 01
PublicationDecade 2020
PublicationTitle Applied thermal engineering
PublicationYear 2025
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Li, Huang, Hu, Li (b0060) 2023; 279
Wang, Wang, Xu (b0100) 2019; 82
Tang (b0040) 2022; 252
Zhou (b0065) 2024; 321
Fan, Han, Li, Peng, Yeh, Hong (b0140) 2024; 238
Liu (b0130) 2025; 262
Sadeghi, Golshanfard, Eslami, Rahbar, Kari (b0160) 2023; 45
Li, Huang, Zhu (b0005) 2020; 207
Karijadi, Chou (b0135) 2022; 259
Kuyumani, Hasan, Shongwe (b0075) 2023; 51
Smyl, Dudek, Pełka (b0035) 2023; 35
Niu, Yu, Sun, Gao, Wang (b0010) 2022; 313
Cao, Wang, Xia (b0125) 2024; 132
Gong, Qu, Zhu, Xu, Yang (b0185) 2025; 318
Iqbal, Adnan, Mohamed, Tariq (b0115) 2024; 24
Lang, Gao (b0205) 2025; 436
Fan, Zhang, Huang, Zhu, Chen (b0080) 2023; 35
Balasubramani, Natarajan (b0145) 2024; 2024
Aderyani, Mousavi, Jafari (b0170) 2022; 614
Yang, Shi, Li, Song, Zhang, Chen (b0190) 2022; 307
Ziyu, Zeyu, Huiwei, Guo, Kun (b0090) 2023; 144
Poongadan, Lineesh (b0150) 2024; 56
Hong, Chan (b0095) 2023; 126
Lu, Wang (b0175) 2023; 82
Zulfiqar, Kamran, Rasheed, Alquthami, Milyani (b0045) 2022; 8
Yu, Si, Hu, Zhang (b0070) 2019; 31
Kachalla, Ghiaus, Baseer (b0030) 2025; 267
Yang, Peng, Zhu, Liang, Qi, Ben (b0110) 2025
Yan, Lu, Liu, He, Zhang, Guo (b0025) 2023; 297
Xu, Pan, Sun, Yuan (b0050) 2022
Zhong, Li, Meng, Li, Yildiz, Mirjalili (b0195) 2025; 37
Zhu, Dong, Zheng, Li, Huang, Xi (b0015) 2022; 321
Chen, Huang, Zheng, Yang (b0105) 2024; 375
Lei, Wang, Li (b0155) 2023; 30
Chen, Zhu, Hu, Wang, Sun, Yang (b0165) 2023; 9
Gu, Wang, Liu (b0120) 2024; 191
Nyangaresi (b0200) 2024; 2024
Kim, Tak, Lee (b0020) 2025
Wan, Chang, Khalil, He (b0085) 2023; 282
Wang, Cui, Niu (b0055) 2022; 14
Cai, Li, Su, Zhu, He (b0180) 2022; 12
Balasubramani (10.1016/j.applthermaleng.2025.128313_b0145) 2024; 2024
Lei (10.1016/j.applthermaleng.2025.128313_b0155) 2023; 30
Aderyani (10.1016/j.applthermaleng.2025.128313_b0170) 2022; 614
Yang (10.1016/j.applthermaleng.2025.128313_b0190) 2022; 307
Lang (10.1016/j.applthermaleng.2025.128313_b0205) 2025; 436
Poongadan (10.1016/j.applthermaleng.2025.128313_b0150) 2024; 56
Kachalla (10.1016/j.applthermaleng.2025.128313_b0030) 2025; 267
Hong (10.1016/j.applthermaleng.2025.128313_b0095) 2023; 126
Gu (10.1016/j.applthermaleng.2025.128313_b0120) 2024; 191
Gong (10.1016/j.applthermaleng.2025.128313_b0185) 2025; 318
Zhou (10.1016/j.applthermaleng.2025.128313_b0065) 2024; 321
Wan (10.1016/j.applthermaleng.2025.128313_b0085) 2023; 282
Lu (10.1016/j.applthermaleng.2025.128313_b0175) 2023; 82
Yu (10.1016/j.applthermaleng.2025.128313_b0070) 2019; 31
Wang (10.1016/j.applthermaleng.2025.128313_b0055) 2022; 14
Fan (10.1016/j.applthermaleng.2025.128313_b0080) 2023; 35
Yan (10.1016/j.applthermaleng.2025.128313_b0025) 2023; 297
Kuyumani (10.1016/j.applthermaleng.2025.128313_b0075) 2023; 51
Niu (10.1016/j.applthermaleng.2025.128313_b0010) 2022; 313
Cai (10.1016/j.applthermaleng.2025.128313_b0180) 2022; 12
Li (10.1016/j.applthermaleng.2025.128313_b0005) 2020; 207
Iqbal (10.1016/j.applthermaleng.2025.128313_b0115) 2024; 24
Liu (10.1016/j.applthermaleng.2025.128313_b0130) 2025; 262
Xu (10.1016/j.applthermaleng.2025.128313_b0050) 2022
Tang (10.1016/j.applthermaleng.2025.128313_b0040) 2022; 252
Zhu (10.1016/j.applthermaleng.2025.128313_b0015) 2022; 321
Smyl (10.1016/j.applthermaleng.2025.128313_b0035) 2023; 35
Ziyu (10.1016/j.applthermaleng.2025.128313_b0090) 2023; 144
Nyangaresi (10.1016/j.applthermaleng.2025.128313_b0200) 2024; 2024
Yang (10.1016/j.applthermaleng.2025.128313_b0110) 2025
Cao (10.1016/j.applthermaleng.2025.128313_b0125) 2024; 132
Zhong (10.1016/j.applthermaleng.2025.128313_b0195) 2025; 37
Chen (10.1016/j.applthermaleng.2025.128313_b0105) 2024; 375
Zulfiqar (10.1016/j.applthermaleng.2025.128313_b0045) 2022; 8
Wang (10.1016/j.applthermaleng.2025.128313_b0100) 2019; 82
Chen (10.1016/j.applthermaleng.2025.128313_b0165) 2023; 9
Sadeghi (10.1016/j.applthermaleng.2025.128313_b0160) 2023; 45
Fan (10.1016/j.applthermaleng.2025.128313_b0140) 2024; 238
Kim (10.1016/j.applthermaleng.2025.128313_b0020) 2025
Karijadi (10.1016/j.applthermaleng.2025.128313_b0135) 2022; 259
Li (10.1016/j.applthermaleng.2025.128313_b0060) 2023; 279
References_xml – volume: 14
  start-page: 7307
  year: 2022
  ident: b0055
  article-title: Wind power forecasting based on LSTM improved by EMD-PCA-RF
  publication-title: Sustainability
– volume: 238
  year: 2024
  ident: b0140
  article-title: A hybrid model for deep learning short-term power load forecasting based on feature extraction statistics techniques
  publication-title: Expert Syst. Appl.
– volume: 45
  start-page: 242
  year: 2023
  end-page: 258
  ident: b0160
  article-title: Improving PV power plant forecast accuracy: a hybrid deep learning approach compared across short, medium, and long-term horizons
  publication-title: Renewable Energy Focus
– volume: 9
  start-page: 1022
  year: 2023
  end-page: 1031
  ident: b0165
  article-title: Research on short-term load forecasting of new-type power system based on GCN-LSTM considering multiple influencing factors
  publication-title: Energy Rep.
– volume: 375
  year: 2024
  ident: b0105
  article-title: A load forecasting approach for integrated energy systems based on aggregation hybrid modal decomposition and combined model
  publication-title: Appl. Energy
– volume: 2024
  start-page: 73
  year: 2024
  end-page: 82
  ident: b0145
  article-title: Improving bus passenger flow prediction using Bi-LSTM fusion model and SMO algorithm
  publication-title: Babylonian J. Artificial Intelligence
– volume: 30
  start-page: 2107
  year: 2023
  end-page: 2116
  ident: b0155
  article-title: A denoising method of partial discharge signal based on improved SVD-VMD
  publication-title: IEEE Trans. Dielectr. Electr. Insul.
– volume: 614
  year: 2022
  ident: b0170
  article-title: Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN
  publication-title: J. Hydrol.
– volume: 2024
  start-page: 32
  year: 2024
  end-page: 38
  ident: b0200
  article-title: AI-driven energy forecasting enhancing smart grid efficiency with LSTM networks
  publication-title: EDRAAK J.
– volume: 279
  year: 2023
  ident: b0060
  article-title: Ultra-short term power load forecasting based on CEEMDAN-SE and LSTM neural network
  publication-title: Energ. Buildings
– volume: 282
  year: 2023
  ident: b0085
  article-title: Short-term power load forecasting for combined heat and power using CNN-LSTM enhanced by attention mechanism
  publication-title: Energy
– volume: 8
  start-page: 13333
  year: 2022
  end-page: 13352
  ident: b0045
  article-title: Hyperparameter optimization of support vector machine using adaptive differential evolution for electricity load forecasting
  publication-title: Energy Rep.
– volume: 191
  year: 2024
  ident: b0120
  article-title: A combined system based on data preprocessing and optimization algorithm for electricity load forecasting
  publication-title: Comput. Ind. Eng.
– volume: 56
  start-page: 164
  year: 2024
  ident: b0150
  article-title: Non-linear time series prediction using improved CEEMDAN, SVD and LSTM
  publication-title: Neural Process. Lett.
– volume: 24
  year: 2024
  ident: b0115
  article-title: A hybrid deep learning framework for short-term load forecasting with improved data cleansing and preprocessing techniques
  publication-title: Results Eng.
– volume: 51
  start-page: 746
  year: 2023
  end-page: 760
  ident: b0075
  article-title: A hybrid model based on CNN-LSTM to detect and forecast harmonics: a case study of an eskom substation in South Africa
  publication-title: Electric Power Components and Systems
– volume: 262
  year: 2025
  ident: b0130
  article-title: Proposing an innovative model for solar irradiance and wind speed forecasting
  publication-title: Appl. Therm. Eng.
– volume: 267
  year: 2025
  ident: b0030
  article-title: Comparative analysis of machine learning models for prediction and forecasting of electric water boilers energy consumption
  publication-title: Appl. Therm. Eng.
– volume: 318
  year: 2025
  ident: b0185
  article-title: Ensemble models of TCN-LSTM-LightGBM based on ensemble learning methods for short-term electrical load forecasting
  publication-title: Energy
– start-page: 1353
  year: 2022
  end-page: 1356
  ident: b0050
  article-title: Short-term load forecasting of power system based on genetic algorithm improved BP neural network algorithm
  publication-title: 2022 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA)
– volume: 12
  start-page: 6647
  year: 2022
  ident: b0180
  article-title: Short-term electrical load forecasting based on VMD and GRU-TCN hybrid network
  publication-title: Appl. Sci.
– year: 2025
  ident: b0020
  article-title: Empirical analysis of the demand-based heat supply system in district heating buildings
  publication-title: Appl. Therm. Eng.
– volume: 82
  year: 2019
  ident: b0100
  article-title: A novel combined model based on hybrid optimization algorithm for electrical load forecasting
  publication-title: Appl. Soft Comput.
– volume: 31
  start-page: 1235
  year: 2019
  end-page: 1270
  ident: b0070
  article-title: A review of recurrent neural networks: LSTM cells and network architectures
  publication-title: Neural Comput.
– volume: 207
  year: 2020
  ident: b0005
  article-title: Gradient descent iterative method for energy flow of integrated energy system considering multiple modes of compressors
  publication-title: Energ. Conver. Manage.
– volume: 313
  year: 2022
  ident: b0010
  article-title: Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism
  publication-title: Appl. Energy
– volume: 436
  year: 2025
  ident: b0205
  article-title: Dream Optimization Algorithm (DOA): a novel metaheuristic optimization algorithm inspired by human dreams and its applications to real-world engineering problems
  publication-title: Comput. Methods Appl. Mech. Eng.
– volume: 132
  year: 2024
  ident: b0125
  article-title: Combined electricity load-forecasting system based on weighted fuzzy time series and deep neural networks
  publication-title: Eng. Appl. Artif. Intel.
– volume: 35
  start-page: 13109
  year: 2023
  end-page: 13118
  ident: b0080
  article-title: Parallel spatio-temporal attention-based TCN for multivariate time series prediction
  publication-title: Neural Comput. & Applic.
– volume: 321
  year: 2022
  ident: b0015
  article-title: Review and prospect of data-driven techniques for load forecasting in integrated energy systems
  publication-title: Appl. Energy
– volume: 35
  start-page: 11346
  year: 2023
  end-page: 11358
  ident: b0035
  article-title: ES-dRNN: a hybrid exponential smoothing and dilated recurrent neural network model for short-term load forecasting
  publication-title: IEEE Trans. Neural Networks Learn. Syst.
– volume: 126
  year: 2023
  ident: b0095
  article-title: Short-term electric load forecasting using particle swarm optimization-based convolutional neural network
  publication-title: Eng. Appl. Artif. Intel.
– volume: 37
  start-page: 3641
  year: 2025
  end-page: 3683
  ident: b0195
  article-title: Starfish optimization algorithm (SFOA): a bio-inspired metaheuristic algorithm for global optimization compared with 100 optimizers
  publication-title: Neural Comput. & Applic.
– year: 2025
  ident: b0110
  article-title: Building heat load forecasting considering mechanism data fusion: a stacking ensemble approach for district heating systems
  publication-title: Appl. Therm. Eng.
– volume: 82
  start-page: 9939
  year: 2023
  end-page: 9959
  ident: b0175
  article-title: A load forecasting model based on support vector regression with whale optimization algorithm
  publication-title: Multimed. Tools Appl.
– volume: 252
  year: 2022
  ident: b0040
  article-title: GM (1, 1) based improved seasonal index model for monthly electricity consumption forecasting
  publication-title: Energy
– volume: 297
  year: 2023
  ident: b0025
  article-title: An improved feature-time Transformer encoder-Bi-LSTM for short-term forecasting of user-level integrated energy loads
  publication-title: Energ. Buildings
– volume: 307
  year: 2022
  ident: b0190
  article-title: A combined deep learning load forecasting model of single household resident user considering multi-time scale electricity consumption behavior
  publication-title: Appl. Energy
– volume: 321
  year: 2024
  ident: b0065
  article-title: ISSA-LSTM: a new data-driven method of heat load forecasting for building air conditioning
  publication-title: Energ. Buildings
– volume: 144
  year: 2023
  ident: b0090
  article-title: Residual LSTM based short-term load forecasting
  publication-title: Appl. Soft Comput. J.
– volume: 259
  year: 2022
  ident: b0135
  article-title: A hybrid RF-LSTM based on CEEMDAN for improving the accuracy of building energy consumption prediction
  publication-title: Energ. Buildings
– volume: 267
  year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0030
  article-title: Comparative analysis of machine learning models for prediction and forecasting of electric water boilers energy consumption
  publication-title: Appl. Therm. Eng.
  doi: 10.1016/j.applthermaleng.2025.125799
– volume: 614
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0170
  article-title: Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN
  publication-title: J. Hydrol.
  doi: 10.1016/j.jhydrol.2022.128463
– volume: 2024
  start-page: 32
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0200
  article-title: AI-driven energy forecasting enhancing smart grid efficiency with LSTM networks
  publication-title: EDRAAK J.
  doi: 10.70470/EDRAAK/2024/005
– volume: 132
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0125
  article-title: Combined electricity load-forecasting system based on weighted fuzzy time series and deep neural networks
  publication-title: Eng. Appl. Artif. Intel.
  doi: 10.1016/j.engappai.2024.108375
– volume: 8
  start-page: 13333
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0045
  article-title: Hyperparameter optimization of support vector machine using adaptive differential evolution for electricity load forecasting
  publication-title: Energy Rep.
  doi: 10.1016/j.egyr.2022.09.188
– volume: 321
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0015
  article-title: Review and prospect of data-driven techniques for load forecasting in integrated energy systems
  publication-title: Appl. Energy
  doi: 10.1016/j.apenergy.2022.119269
– volume: 191
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0120
  article-title: A combined system based on data preprocessing and optimization algorithm for electricity load forecasting
  publication-title: Comput. Ind. Eng.
  doi: 10.1016/j.cie.2024.110114
– volume: 24
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0115
  article-title: A hybrid deep learning framework for short-term load forecasting with improved data cleansing and preprocessing techniques
  publication-title: Results Eng.
  doi: 10.1016/j.rineng.2024.103560
– volume: 318
  year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0185
  article-title: Ensemble models of TCN-LSTM-LightGBM based on ensemble learning methods for short-term electrical load forecasting
  publication-title: Energy
  doi: 10.1016/j.energy.2025.134757
– volume: 259
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0135
  article-title: A hybrid RF-LSTM based on CEEMDAN for improving the accuracy of building energy consumption prediction
  publication-title: Energ. Buildings
  doi: 10.1016/j.enbuild.2022.111908
– volume: 144
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0090
  article-title: Residual LSTM based short-term load forecasting
  publication-title: Appl. Soft Comput. J.
– volume: 37
  start-page: 3641
  issue: 5
  year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0195
  article-title: Starfish optimization algorithm (SFOA): a bio-inspired metaheuristic algorithm for global optimization compared with 100 optimizers
  publication-title: Neural Comput. & Applic.
  doi: 10.1007/s00521-024-10694-1
– volume: 375
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0105
  article-title: A load forecasting approach for integrated energy systems based on aggregation hybrid modal decomposition and combined model
  publication-title: Appl. Energy
  doi: 10.1016/j.apenergy.2024.124166
– volume: 238
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0140
  article-title: A hybrid model for deep learning short-term power load forecasting based on feature extraction statistics techniques
  publication-title: Expert Syst. Appl.
  doi: 10.1016/j.eswa.2023.122012
– volume: 436
  year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0205
  article-title: Dream Optimization Algorithm (DOA): a novel metaheuristic optimization algorithm inspired by human dreams and its applications to real-world engineering problems
  publication-title: Comput. Methods Appl. Mech. Eng.
  doi: 10.1016/j.cma.2024.117718
– start-page: 1353
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0050
  article-title: Short-term load forecasting of power system based on genetic algorithm improved BP neural network algorithm
– volume: 207
  year: 2020
  ident: 10.1016/j.applthermaleng.2025.128313_b0005
  article-title: Gradient descent iterative method for energy flow of integrated energy system considering multiple modes of compressors
  publication-title: Energ. Conver. Manage.
  doi: 10.1016/j.enconman.2020.112534
– year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0020
  article-title: Empirical analysis of the demand-based heat supply system in district heating buildings
  publication-title: Appl. Therm. Eng.
  doi: 10.1016/j.applthermaleng.2025.127031
– volume: 313
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0010
  article-title: Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism
  publication-title: Appl. Energy
  doi: 10.1016/j.apenergy.2022.118801
– volume: 282
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0085
  article-title: Short-term power load forecasting for combined heat and power using CNN-LSTM enhanced by attention mechanism
  publication-title: Energy
  doi: 10.1016/j.energy.2023.128274
– volume: 297
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0025
  article-title: An improved feature-time Transformer encoder-Bi-LSTM for short-term forecasting of user-level integrated energy loads
  publication-title: Energ. Buildings
  doi: 10.1016/j.enbuild.2023.113396
– volume: 82
  year: 2019
  ident: 10.1016/j.applthermaleng.2025.128313_b0100
  article-title: A novel combined model based on hybrid optimization algorithm for electrical load forecasting
  publication-title: Appl. Soft Comput.
  doi: 10.1016/j.asoc.2019.105548
– volume: 35
  start-page: 11346
  issue: 8
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0035
  article-title: ES-dRNN: a hybrid exponential smoothing and dilated recurrent neural network model for short-term load forecasting
  publication-title: IEEE Trans. Neural Networks Learn. Syst.
  doi: 10.1109/TNNLS.2023.3259149
– volume: 12
  start-page: 6647
  issue: 13
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0180
  article-title: Short-term electrical load forecasting based on VMD and GRU-TCN hybrid network
  publication-title: Appl. Sci.
  doi: 10.3390/app12136647
– volume: 279
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0060
  article-title: Ultra-short term power load forecasting based on CEEMDAN-SE and LSTM neural network
  publication-title: Energ. Buildings
  doi: 10.1016/j.enbuild.2022.112666
– volume: 262
  year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0130
  article-title: Proposing an innovative model for solar irradiance and wind speed forecasting
  publication-title: Appl. Therm. Eng.
  doi: 10.1016/j.applthermaleng.2024.125224
– volume: 35
  start-page: 13109
  issue: 18
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0080
  article-title: Parallel spatio-temporal attention-based TCN for multivariate time series prediction
  publication-title: Neural Comput. & Applic.
  doi: 10.1007/s00521-021-05958-z
– volume: 9
  start-page: 1022
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0165
  article-title: Research on short-term load forecasting of new-type power system based on GCN-LSTM considering multiple influencing factors
  publication-title: Energy Rep.
  doi: 10.1016/j.egyr.2023.05.048
– volume: 31
  start-page: 1235
  issue: 7
  year: 2019
  ident: 10.1016/j.applthermaleng.2025.128313_b0070
  article-title: A review of recurrent neural networks: LSTM cells and network architectures
  publication-title: Neural Comput.
  doi: 10.1162/neco_a_01199
– volume: 2024
  start-page: 73
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0145
  article-title: Improving bus passenger flow prediction using Bi-LSTM fusion model and SMO algorithm
  publication-title: Babylonian J. Artificial Intelligence
  doi: 10.58496/BJAI/2024/010
– volume: 45
  start-page: 242
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0160
  article-title: Improving PV power plant forecast accuracy: a hybrid deep learning approach compared across short, medium, and long-term horizons
  publication-title: Renewable Energy Focus
  doi: 10.1016/j.ref.2023.04.010
– volume: 56
  start-page: 164
  issue: 3
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0150
  article-title: Non-linear time series prediction using improved CEEMDAN, SVD and LSTM
  publication-title: Neural Process. Lett.
  doi: 10.1007/s11063-024-11622-z
– volume: 307
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0190
  article-title: A combined deep learning load forecasting model of single household resident user considering multi-time scale electricity consumption behavior
  publication-title: Appl. Energy
  doi: 10.1016/j.apenergy.2021.118197
– year: 2025
  ident: 10.1016/j.applthermaleng.2025.128313_b0110
  article-title: Building heat load forecasting considering mechanism data fusion: a stacking ensemble approach for district heating systems
  publication-title: Appl. Therm. Eng.
– volume: 252
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0040
  article-title: GM (1, 1) based improved seasonal index model for monthly electricity consumption forecasting
  publication-title: Energy
  doi: 10.1016/j.energy.2022.124041
– volume: 82
  start-page: 9939
  issue: 7
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0175
  article-title: A load forecasting model based on support vector regression with whale optimization algorithm
  publication-title: Multimed. Tools Appl.
  doi: 10.1007/s11042-022-13462-2
– volume: 14
  start-page: 7307
  issue: 12
  year: 2022
  ident: 10.1016/j.applthermaleng.2025.128313_b0055
  article-title: Wind power forecasting based on LSTM improved by EMD-PCA-RF
  publication-title: Sustainability
  doi: 10.3390/su14127307
– volume: 321
  year: 2024
  ident: 10.1016/j.applthermaleng.2025.128313_b0065
  article-title: ISSA-LSTM: a new data-driven method of heat load forecasting for building air conditioning
  publication-title: Energ. Buildings
  doi: 10.1016/j.enbuild.2024.114698
– volume: 51
  start-page: 746
  issue: 8
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0075
  article-title: A hybrid model based on CNN-LSTM to detect and forecast harmonics: a case study of an eskom substation in South Africa
  publication-title: Electric Power Components and Systems
  doi: 10.1080/15325008.2023.2181883
– volume: 126
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0095
  article-title: Short-term electric load forecasting using particle swarm optimization-based convolutional neural network
  publication-title: Eng. Appl. Artif. Intel.
  doi: 10.1016/j.engappai.2023.106773
– volume: 30
  start-page: 2107
  issue: 5
  year: 2023
  ident: 10.1016/j.applthermaleng.2025.128313_b0155
  article-title: A denoising method of partial discharge signal based on improved SVD-VMD
  publication-title: IEEE Trans. Dielectr. Electr. Insul.
  doi: 10.1109/TDEI.2023.3269725
SSID ssj0012874
Score 2.4699326
Snippet •Expanding economic and climate features as inputs to enhance the model’s ability to capture load changes.•Introducing singular value decomposition technology...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 128313
SubjectTerms Combined model
Data denoising
Feature selection
Parameter optimization
Short-term load forecasting
Title A new method for short-term load forecasting based on weighted fusion of base models
URI https://dx.doi.org/10.1016/j.applthermaleng.2025.128313
Volume 280
WOSCitedRecordID wos001573266800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 1359-4311
  databaseCode: AIEXJ
  dateStart: 19960101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0012874
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9wwEB7RparKoaIPVKBFPnCrsiJ2HNunaoWoKEgIqVt1b5GTOAIECSK7wM_v-JFNoK1EVfUSZZ3Ik_V8tj-PxzMAuzkz0hjsgAInB1yglEmEv0VU0JLHTOMUWCYu2YQ4OZGzmToN4Qlal05A1LW8v1fX_1XVWIbKtkdn_0Ldy0qxAO9R6XhFteP1SYqf2CzhITO0cyJsz5BiR3YI_nTZaFdmCt06f2c7iZV2w-DOmUjxvlq0gUPaZz5TTjuksB1vtczxChVs-oiGvcXdm1WP9XlXNFs4vAxe-hEs1Ye6OdPNklCHHCtHi7oy4eVglKD8kYPH8rRM75pkB1fGVYSEJR6OvtQncvplJPdGhYux3ccPfwiljq2wMc6pzJ9gfRQr-5sVYSUgsaMKmeMzWKWCKzmC1cnXg9nRcoPJhvl3a_HwSS9gt3f9-7PM37OXASOZrsOrsJQgEw-B17Bi6jewNggw-RamE4JgIB4MBBVPejAQCwYyAANxYCBNTTowEA8G0lTuGfFgeAffvxxM9w-jkEcDe5xg8wgZc0xjHeeJ2lNG5pbS5lXMjdC50EyXyFFlomSR2uiylaR5ij1V7ClKi5yXBduAUd3U5j0QIyoVJzo1TKdJWSQ5T1Jd8IoaRpVMzSbwrn2yax8uJev8CC-yh-2a2XbNfLtuwueuMbNA_TylyxAPT6ph659r2IaXPZQ_wGh-szAf4XlxOz9vb3YCgH4C56-ILw
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+new+method+for+short-term+load+forecasting+based+on+weighted+fusion+of+base+models&rft.jtitle=Applied+thermal+engineering&rft.au=Bian%2C+Kai&rft.au=Xu%2C+Jing&rft.au=Wang%2C+Haohao&rft.au=Ding%2C+Junfeng&rft.date=2025-12-01&rft.pub=Elsevier+Ltd&rft.issn=1359-4311&rft.volume=280&rft_id=info:doi/10.1016%2Fj.applthermaleng.2025.128313&rft.externalDocID=S1359431125029059
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1359-4311&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1359-4311&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1359-4311&client=summon