Multi-step-ahead prediction of thermal load in regional energy system using deep learning method
[Display omitted] Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and multi-energy dispatch of building energy systems. Conventional approaches to future load time series prediction rely either on iterative one-step-ahea...
Uložené v:
| Vydané v: | Energy and buildings Ročník 233; s. 110658 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Lausanne
Elsevier B.V
15.02.2021
Elsevier BV |
| Predmet: | |
| ISSN: | 0378-7788, 1872-6178 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | [Display omitted]
Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and multi-energy dispatch of building energy systems. Conventional approaches to future load time series prediction rely either on iterative one-step-ahead predictors or direct predictors, which ignores the temporal dependency between successive loads in time-delay building energy systems. This study, therefore, proposes a temporal attention encoder-decoder network (TA-EDN) model to improve the accuracy of multi-step-ahead thermal load prediction with the following three functional modules: long short-term memory (LSTM) network, which is to process the intrinsic temporal relationships among input and output variables, encoder-decoder network (EDN), which is to realize the multi-input multi-output modeling, and attention mechanism, which is to improve the ability of processing variables with long sequences. An actual regional energy system is selected to perform the 24-hour-ahead prediction using the proposed model as a validation experiment. The results suggest that the TA-EDN model significantly improves the prediction accuracy of the future thermal loads time series, achieving a mean absolute percentage error of 7.4%, compared to that of 9.1% of EDN model, 12.4% of LSTM-IS (a model combining LSTM with iterative strategy) and 12.7% of LSTM-DS (a model combining LSTM with direct strategy). In addition, compared with the ideal benchmark of multi-step-ahead prediction, the proposed TA-EDN model has room for improvement in the prediction of low load or large fluctuation load. |
|---|---|
| AbstractList | [Display omitted]
Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and multi-energy dispatch of building energy systems. Conventional approaches to future load time series prediction rely either on iterative one-step-ahead predictors or direct predictors, which ignores the temporal dependency between successive loads in time-delay building energy systems. This study, therefore, proposes a temporal attention encoder-decoder network (TA-EDN) model to improve the accuracy of multi-step-ahead thermal load prediction with the following three functional modules: long short-term memory (LSTM) network, which is to process the intrinsic temporal relationships among input and output variables, encoder-decoder network (EDN), which is to realize the multi-input multi-output modeling, and attention mechanism, which is to improve the ability of processing variables with long sequences. An actual regional energy system is selected to perform the 24-hour-ahead prediction using the proposed model as a validation experiment. The results suggest that the TA-EDN model significantly improves the prediction accuracy of the future thermal loads time series, achieving a mean absolute percentage error of 7.4%, compared to that of 9.1% of EDN model, 12.4% of LSTM-IS (a model combining LSTM with iterative strategy) and 12.7% of LSTM-DS (a model combining LSTM with direct strategy). In addition, compared with the ideal benchmark of multi-step-ahead prediction, the proposed TA-EDN model has room for improvement in the prediction of low load or large fluctuation load. Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and multi-energy dispatch of building energy systems. Conventional approaches to future load time series prediction rely either on iterative one-step-ahead predictors or direct predictors, which ignores the temporal dependency between successive loads in time-delay building energy systems. This study, therefore, proposes a temporal attention encoder-decoder network (TA-EDN) model to improve the accuracy of multi-step-ahead thermal load prediction with the following three functional modules: long short-term memory (LSTM) network, which is to process the intrinsic temporal relationships among input and output variables, encoder-decoder network (EDN), which is to realize the multi-input multi-output modeling, and attention mechanism, which is to improve the ability of processing variables with long sequences. An actual regional energy system is selected to perform the 24-hour-ahead prediction using the proposed model as a validation experiment. The results suggest that the TA-EDN model significantly improves the prediction accuracy of the future thermal loads time series, achieving a mean absolute percentage error of 7.4%, compared to that of 9.1% of EDN model, 12.4% of LSTM-IS (a model combining LSTM with iterative strategy) and 12.7% of LSTM-DS (a model combining LSTM with direct strategy). In addition, compared with the ideal benchmark of multi-step-ahead prediction, the proposed TA-EDN model has room for improvement in the prediction of low load or large fluctuation load. |
| ArticleNumber | 110658 |
| Author | Zhou, Ruoyu Tian, Zhe Lu, Yakai Liu, Wenjing |
| Author_xml | – sequence: 1 givenname: Yakai surname: Lu fullname: Lu, Yakai – sequence: 2 givenname: Zhe surname: Tian fullname: Tian, Zhe email: tianzhe@tju.edu.cn – sequence: 3 givenname: Ruoyu surname: Zhou fullname: Zhou, Ruoyu – sequence: 4 givenname: Wenjing surname: Liu fullname: Liu, Wenjing |
| BookMark | eNqFkE1LAzEQhoMo2Ko_QQh43pqku5ssHkSKX6B40XPMJrNtyjZZk6zQf29Ke_LSU8jM-wwzzxSdOu8AoWtKZpTQ-nY9A9eOtjczRliuUVJX4gRNqOCsqCkXp2hC5lwUnAtxjqYxrgnJGU4n6Pt97JMtYoKhUCtQBg8BjNXJeod9h9MKwkb1uPe5ZR0OsMydXAAHYbnFcZvRDR6jdUtsAAbcgwpu99tAWnlzic461Ue4OrwX6Ovp8XPxUrx9PL8uHt4KPZ_zVDBWG9EIBS0pW2qYIV3ZqrLlwMqacF6WLJ9aMmhES4Q2tFW6URpMVXWcV2x-gW72c4fgf0aISa79GPKmUbKyoQ3PHnhO3e1TOvgYA3RS26R2x6agbC8pkTulci0PSuVOqdwrzXT1jx6C3aiwPcrd7znIAn4tBBm1BZe3twF0ksbbIxP-AFpEln4 |
| CitedBy_id | crossref_primary_10_1007_s11269_024_04033_1 crossref_primary_10_1016_j_enbuild_2022_111832 crossref_primary_10_1109_JIOT_2025_3549715 crossref_primary_10_1016_j_est_2024_114959 crossref_primary_10_1016_j_neucom_2024_128939 crossref_primary_10_1016_j_apenergy_2025_125273 crossref_primary_10_1007_s13042_023_01796_8 crossref_primary_10_1016_j_jobe_2024_111448 crossref_primary_10_3390_agronomy14020349 crossref_primary_10_1016_j_jobe_2021_103017 crossref_primary_10_1016_j_enbuild_2024_114661 crossref_primary_10_1016_j_egyai_2025_100480 crossref_primary_10_1016_j_ijhydene_2023_01_068 crossref_primary_10_1007_s10043_024_00873_9 crossref_primary_10_1016_j_scs_2021_103625 crossref_primary_10_1016_j_compag_2023_107932 crossref_primary_10_1109_TII_2025_3534419 crossref_primary_10_1016_j_enbuild_2022_112317 crossref_primary_10_1016_j_jobe_2022_105330 crossref_primary_10_1007_s00521_023_08878_2 crossref_primary_10_1016_j_apenergy_2022_118801 crossref_primary_10_1016_j_aei_2021_101357 crossref_primary_10_3390_s23146528 crossref_primary_10_1016_j_rineng_2024_103765 crossref_primary_10_1016_j_aei_2022_101854 crossref_primary_10_1016_j_apenergy_2024_123042 crossref_primary_10_1371_journal_pntd_0011587 crossref_primary_10_1016_j_petrol_2022_110844 crossref_primary_10_3390_su16177805 crossref_primary_10_1016_j_energy_2022_124915 crossref_primary_10_1109_ACCESS_2021_3129172 crossref_primary_10_1016_j_jobe_2022_105028 |
| Cites_doi | 10.1016/j.enconman.2015.07.009 10.1016/j.procs.2017.11.374 10.1016/j.neucom.2009.11.030 10.1007/11494669_124 10.1162/neco.1997.9.8.1735 10.1016/j.energy.2018.09.068 10.1016/j.apenergy.2019.03.187 10.18653/v1/D15-1166 10.1016/j.scs.2018.02.016 10.1016/j.apenergy.2018.12.004 10.1109/SIU.2018.8404313 10.1016/j.enbuild.2013.06.007 10.1016/j.energy.2019.116085 10.1016/j.energy.2017.04.045 10.1162/tacl_a_00105 10.1016/j.enbuild.2019.02.014 10.1016/j.apenergy.2019.02.052 10.21437/Interspeech.2012-65 10.1016/j.enconman.2013.12.060 10.1016/j.apenergy.2017.12.051 10.1016/j.neucom.2006.06.015 10.1016/j.enbuild.2018.04.008 10.1016/j.enbuild.2012.08.007 10.1016/j.rser.2017.07.018 10.1109/IECON.2016.7793413 10.1016/j.eswa.2012.01.039 10.1016/j.apenergy.2017.07.009 10.1016/j.applthermaleng.2017.09.007 10.1109/CAC.2017.8242744 |
| ContentType | Journal Article |
| Copyright | 2020 Copyright Elsevier BV Feb 15, 2021 |
| Copyright_xml | – notice: 2020 – notice: Copyright Elsevier BV Feb 15, 2021 |
| DBID | AAYXX CITATION 7ST 8FD C1K F28 FR3 KR7 SOI |
| DOI | 10.1016/j.enbuild.2020.110658 |
| DatabaseName | CrossRef Environment Abstracts Technology Research Database Environmental Sciences and Pollution Management ANTE: Abstracts in New Technology & Engineering Engineering Research Database Civil Engineering Abstracts Environment Abstracts |
| DatabaseTitle | CrossRef Civil Engineering Abstracts Engineering Research Database Technology Research Database Environment Abstracts ANTE: Abstracts in New Technology & Engineering Environmental Sciences and Pollution Management |
| DatabaseTitleList | Civil Engineering Abstracts |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1872-6178 |
| ExternalDocumentID | 10_1016_j_enbuild_2020_110658 S0378778820334447 |
| GroupedDBID | --M -~X .~1 0R~ 1B1 1~. 1~5 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JM 9JN AABNK AACTN AAEDT AAEDW AAHCO AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AARJD AAXUO ABFYP ABJNI ABLST ABMAC ABYKQ ACDAQ ACGFS ACIWK ACRLP ADBBV ADEZE ADTZH AEBSH AECPX AEKER AENEX AFKWA AFRAH AFTJW AFXIZ AGHFR AGUBO AGYEJ AHEUO AHHHB AHIDL AHJVU AIEXJ AIKHN AITUG AJOXV AKIFW ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AXJTR BELTK BJAXD BKOJK BLECG BLXMC CS3 DU5 EBS EFJIC EFLBG EO8 EO9 EP2 EP3 FDB FIRID FNPLU FYGXN G-Q GBLVA IHE J1W JARJE JJJVA KCYFY KOM LY6 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 RNS ROL SDF SDG SES SPC SPCBC SSJ SSR SST SSZ T5K ~02 ~G- --K 29G 9DU AAQXK AATTM AAXKI AAYWO AAYXX ABFNM ABWVN ABXDB ACLOT ACNNM ACRPL ACVFH ADCNI ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EJD FEDTE FGOYB G-2 HVGLF HZ~ R2- RPZ SAC SET SEW WUQ ZMT ZY4 ~HD 7ST 8FD AGCQF C1K F28 FR3 KR7 SOI |
| ID | FETCH-LOGICAL-c337t-226d898aeb04b1d2d0f4ba4b7e24607744201642e98b08cd1bac9aced55f77523 |
| ISICitedReferencesCount | 38 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000712412900005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0378-7788 |
| IngestDate | Wed Aug 13 04:27:09 EDT 2025 Tue Nov 18 21:51:39 EST 2025 Sat Nov 29 07:12:20 EST 2025 Fri Feb 23 02:48:40 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Deep learning Attention mechanism Sequence-to-sequence Multi-step-ahead prediction Temporal dependency |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c337t-226d898aeb04b1d2d0f4ba4b7e24607744201642e98b08cd1bac9aced55f77523 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| PQID | 2491971787 |
| PQPubID | 2045483 |
| ParticipantIDs | proquest_journals_2491971787 crossref_citationtrail_10_1016_j_enbuild_2020_110658 crossref_primary_10_1016_j_enbuild_2020_110658 elsevier_sciencedirect_doi_10_1016_j_enbuild_2020_110658 |
| PublicationCentury | 2000 |
| PublicationDate | 2021-02-15 |
| PublicationDateYYYYMMDD | 2021-02-15 |
| PublicationDate_xml | – month: 02 year: 2021 text: 2021-02-15 day: 15 |
| PublicationDecade | 2020 |
| PublicationPlace | Lausanne |
| PublicationPlace_xml | – name: Lausanne |
| PublicationTitle | Energy and buildings |
| PublicationYear | 2021 |
| Publisher | Elsevier B.V Elsevier BV |
| Publisher_xml | – name: Elsevier B.V – name: Elsevier BV |
| References | Niu, Tian, Lu (b0020) 2019; 243 Werner (b0025) 2017; 137 Ding, Zhang, Yuan, Yang (b0065) 2018; 128 Wang, Wang, Zeng (b0045) 2018; 171 M.T. Luong, H. Pham, C.D. Manning, Effective approaches to attention-based neural machine translation[J]. arXiv preprint arXiv:1508.04025, 2015. Lu, Tian, Peng (b0170) 2019; 190 Koschwitz, Frisch, Van Treeck (b0040) 2018; 165 Fan, Wang, Gang (b0080) 2019; 236 A. Vaswani, N. Shazeer, N. Parmar, et al. Attention is all you need[C], in: Advances in Neural Information Processing Systems, 2017, pp. 5998–6008. Zhou, Cao, Wang (b0140) 2016; 4 Zhang, Cao, Romagnoli (b0075) 2018; 39 Sorjamaa, Hao, Reyhani (b0110) 2007; 70 M. Sundermeyer, R. Schlüter, H. Ney, LSTM neural networks for language modeling[C], in: Thirteenth Annual Conference of the International Speech Communication Association, 2012. Fan, Sun, Zhao (b0070) 2019; 240 Rahman, Srikumar, Smith (b0085) 2018; 212 D.L. Marino, K. Amarasinghe, M. Manic, Building energy load forecasting using deep neural networks[C], in: IECON 2016-42nd Annual Conference of the IEEE Industrial Electronics Society. IEEE, 2016, pp. 7046–7051. He (b0050) 2017; 122 Kumar, Aggarwal, Sharma (b0035) 2013; 65 L. Kuan, B. Zhenfu, W. Xin, et al., Short-term CHP heat load forecast method based on concatenated LSTMs[C], in: 2017 Chinese Automation Congress (CAC). IEEE, 2017, pp. 99–103. A. Tokgöz, G. Ünal, A RNN based time series approach for forecasting turkish electricity load[C], in: 2018 26th Signal Processing and Communications Applications Conference (SIU). IEEE, 2018, pp. 1–4. Y. Ji, J. Hao, N. Reyhani, et al., Direct and recursive prediction of time series using mutual information selection[C], in: International Work-Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2005, pp. 1010–1017. Ma, Fang, Liu (b0030) 2017; 204 Yun, Luck, Mago (b0055) 2012; 54 Guo, Nazarian, Ko (b0060) 2014; 80 I. Sutskever, O. Vinyals, Q.V. Le, Sequence to sequence learning with neural networks[C], in: Advances in Neural Information Processing Systems, 2014, pp. 3104–3112. Di Somma, Yan, Bianco (b0015) 2015; 103 Bontempi (b0120) 2008 Hochreiter, Schmidhuber (b0165) 1997; 9 Zafar, Mahmood, Razzaq (b0010) 2018; 82 Taieb, Bontempi, Atiya (b0130) 2012; 39 Xue, Jiang, Zhou (b0115) 2019; 188 Taieb, Sorjamaa, Bontempi (b0125) 2010; 73 Liu (b0005) 2018; 99 D. Bahdanau, K. Cho, Y. Bengio, Neural machine translation by jointly learning to align and translate[J]. arXiv preprint arXiv:1409.0473, 2014. Niu (10.1016/j.enbuild.2020.110658_b0020) 2019; 243 Kumar (10.1016/j.enbuild.2020.110658_b0035) 2013; 65 Fan (10.1016/j.enbuild.2020.110658_b0080) 2019; 236 Rahman (10.1016/j.enbuild.2020.110658_b0085) 2018; 212 10.1016/j.enbuild.2020.110658_b0105 Zhang (10.1016/j.enbuild.2020.110658_b0075) 2018; 39 Fan (10.1016/j.enbuild.2020.110658_b0070) 2019; 240 10.1016/j.enbuild.2020.110658_b0160 Zafar (10.1016/j.enbuild.2020.110658_b0010) 2018; 82 10.1016/j.enbuild.2020.110658_b0100 Bontempi (10.1016/j.enbuild.2020.110658_b0120) 2008 10.1016/j.enbuild.2020.110658_b0145 Guo (10.1016/j.enbuild.2020.110658_b0060) 2014; 80 Taieb (10.1016/j.enbuild.2020.110658_b0125) 2010; 73 Zhou (10.1016/j.enbuild.2020.110658_b0140) 2016; 4 Yun (10.1016/j.enbuild.2020.110658_b0055) 2012; 54 Ding (10.1016/j.enbuild.2020.110658_b0065) 2018; 128 Ma (10.1016/j.enbuild.2020.110658_b0030) 2017; 204 10.1016/j.enbuild.2020.110658_b0090 He (10.1016/j.enbuild.2020.110658_b0050) 2017; 122 Lu (10.1016/j.enbuild.2020.110658_b0170) 2019; 190 Di Somma (10.1016/j.enbuild.2020.110658_b0015) 2015; 103 Sorjamaa (10.1016/j.enbuild.2020.110658_b0110) 2007; 70 Taieb (10.1016/j.enbuild.2020.110658_b0130) 2012; 39 Hochreiter (10.1016/j.enbuild.2020.110658_b0165) 1997; 9 10.1016/j.enbuild.2020.110658_b0135 10.1016/j.enbuild.2020.110658_b0095 10.1016/j.enbuild.2020.110658_b0150 Liu (10.1016/j.enbuild.2020.110658_b0005) 2018; 99 Xue (10.1016/j.enbuild.2020.110658_b0115) 2019; 188 10.1016/j.enbuild.2020.110658_b0155 Werner (10.1016/j.enbuild.2020.110658_b0025) 2017; 137 Koschwitz (10.1016/j.enbuild.2020.110658_b0040) 2018; 165 Wang (10.1016/j.enbuild.2020.110658_b0045) 2018; 171 |
| References_xml | – volume: 212 start-page: 372 year: 2018 end-page: 385 ident: b0085 article-title: Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks[J] publication-title: Appl. Energy – volume: 70 start-page: 2861 year: 2007 end-page: 2869 ident: b0110 article-title: Methodology for long-term prediction of time series[J] publication-title: Neurocomputing – volume: 204 start-page: 181 year: 2017 end-page: 205 ident: b0030 article-title: Modeling of district load forecasting for distributed energy system[J] publication-title: Appl. Energy – volume: 54 start-page: 225 year: 2012 end-page: 233 ident: b0055 article-title: Building hourly thermal load prediction using an indexed ARX model[J] publication-title: Energy Build. – volume: 65 start-page: 352 year: 2013 end-page: 358 ident: b0035 article-title: Energy analysis of a building using artificial neural network: a review[J] publication-title: Energy Build. – volume: 188 start-page: 116085 year: 2019 ident: b0115 article-title: Multi-step ahead forecasting of heat load in district heating systems using machine learning algorithms[J] publication-title: Energy – reference: A. Vaswani, N. Shazeer, N. Parmar, et al. Attention is all you need[C], in: Advances in Neural Information Processing Systems, 2017, pp. 5998–6008. – volume: 122 start-page: 308 year: 2017 end-page: 314 ident: b0050 article-title: Load forecasting via deep neural networks[J] publication-title: Procedia Comput. Sci. – reference: Y. Ji, J. Hao, N. Reyhani, et al., Direct and recursive prediction of time series using mutual information selection[C], in: International Work-Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2005, pp. 1010–1017. – start-page: 145 year: 2008 end-page: 154 ident: b0120 article-title: Long term time series prediction with multi-input multi-output local learning[J] publication-title: Proc. 2nd ESTSP – volume: 243 start-page: 274 year: 2019 end-page: 287 ident: b0020 article-title: Flexible dispatch of a building energy system using building thermal storage and battery energy storage[J] publication-title: Appl. Energy – volume: 39 start-page: 508 year: 2018 end-page: 518 ident: b0075 article-title: On the feature engineering of building energy data mining[J] publication-title: Sustain. Cities Soc. – volume: 128 start-page: 225 year: 2018 end-page: 234 ident: b0065 article-title: Effect of input variables on cooling load prediction accuracy of an office building publication-title: Appl. Therm. Eng. – volume: 39 start-page: 7067 year: 2012 end-page: 7083 ident: b0130 article-title: A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition[J] publication-title: Expert Syst. Appl. – reference: D.L. Marino, K. Amarasinghe, M. Manic, Building energy load forecasting using deep neural networks[C], in: IECON 2016-42nd Annual Conference of the IEEE Industrial Electronics Society. IEEE, 2016, pp. 7046–7051. – reference: D. Bahdanau, K. Cho, Y. Bengio, Neural machine translation by jointly learning to align and translate[J]. arXiv preprint arXiv:1409.0473, 2014. – volume: 236 start-page: 700 year: 2019 end-page: 710 ident: b0080 article-title: Assessment of deep recurrent neural network-based strategies for short-term building energy predictions[J] publication-title: Appl. Energy – volume: 4 start-page: 371 year: 2016 end-page: 383 ident: b0140 article-title: Deep recurrent models with fast-forward connections for neural machine translation[J] publication-title: Trans. Assoc. Comput. Linguist. – volume: 165 start-page: 134 year: 2018 end-page: 142 ident: b0040 article-title: Data-driven heating and cooling load predictions for non-residential buildings based on support vector machine regression and NARX Recurrent Neural Network: a comparative study on district scale[J] publication-title: Energy – volume: 80 start-page: 46 year: 2014 end-page: 53 ident: b0060 article-title: Hourly cooling load forecasting using time-indexed ARX models with two-stage weighted least squares regression[J] publication-title: Energy Convers. Manage. – reference: A. Tokgöz, G. Ünal, A RNN based time series approach for forecasting turkish electricity load[C], in: 2018 26th Signal Processing and Communications Applications Conference (SIU). IEEE, 2018, pp. 1–4. – volume: 73 start-page: 1950 year: 2010 end-page: 1957 ident: b0125 article-title: Multiple-output modeling for multi-step-ahead time series forecasting[J] publication-title: Neurocomputing – volume: 190 start-page: 49 year: 2019 end-page: 60 ident: b0170 article-title: GMM clustering for heating load patterns in-depth identification and prediction model accuracy improvement of district heating system[J] publication-title: Energy Build. – reference: M.T. Luong, H. Pham, C.D. Manning, Effective approaches to attention-based neural machine translation[J]. arXiv preprint arXiv:1508.04025, 2015. – volume: 103 start-page: 739 year: 2015 end-page: 751 ident: b0015 article-title: Operation optimization of a distributed energy system considering energy costs and exergy efficiency[J] publication-title: Energy Convers. Manage. – reference: L. Kuan, B. Zhenfu, W. Xin, et al., Short-term CHP heat load forecast method based on concatenated LSTMs[C], in: 2017 Chinese Automation Congress (CAC). IEEE, 2017, pp. 99–103. – reference: I. Sutskever, O. Vinyals, Q.V. Le, Sequence to sequence learning with neural networks[C], in: Advances in Neural Information Processing Systems, 2014, pp. 3104–3112. – volume: 82 start-page: 1675 year: 2018 end-page: 1684 ident: b0010 article-title: Prosumer based energy management and sharing in smart grid[J] publication-title: Renew. Sustain. Energy Rev. – volume: 171 start-page: 11 year: 2018 end-page: 25 ident: b0045 article-title: Random Forest based hourly building energy prediction[J] publication-title: Energy Build. – volume: 99 start-page: 1 year: 2018 ident: b0005 article-title: Day-ahead optimal operation for multi-energy residential systems with renewables publication-title: IEEE Trans. Sustain. Energy – reference: M. Sundermeyer, R. Schlüter, H. Ney, LSTM neural networks for language modeling[C], in: Thirteenth Annual Conference of the International Speech Communication Association, 2012. – volume: 9 start-page: 1735 year: 1997 end-page: 1780 ident: b0165 article-title: Long short-term memory[J] publication-title: Neural Comput. – volume: 137 start-page: 617 year: 2017 end-page: 631 ident: b0025 article-title: International review of district heating and cooling[J] publication-title: Energy – volume: 240 start-page: 35 year: 2019 end-page: 45 ident: b0070 article-title: Deep learning-based feature engineering methods for improved building energy prediction[J] publication-title: Appl. Energy – volume: 103 start-page: 739 year: 2015 ident: 10.1016/j.enbuild.2020.110658_b0015 article-title: Operation optimization of a distributed energy system considering energy costs and exergy efficiency[J] publication-title: Energy Convers. Manage. doi: 10.1016/j.enconman.2015.07.009 – volume: 122 start-page: 308 year: 2017 ident: 10.1016/j.enbuild.2020.110658_b0050 article-title: Load forecasting via deep neural networks[J] publication-title: Procedia Comput. Sci. doi: 10.1016/j.procs.2017.11.374 – volume: 73 start-page: 1950 issue: 10–12 year: 2010 ident: 10.1016/j.enbuild.2020.110658_b0125 article-title: Multiple-output modeling for multi-step-ahead time series forecasting[J] publication-title: Neurocomputing doi: 10.1016/j.neucom.2009.11.030 – ident: 10.1016/j.enbuild.2020.110658_b0105 doi: 10.1007/11494669_124 – volume: 9 start-page: 1735 issue: 8 year: 1997 ident: 10.1016/j.enbuild.2020.110658_b0165 article-title: Long short-term memory[J] publication-title: Neural Comput. doi: 10.1162/neco.1997.9.8.1735 – volume: 165 start-page: 134 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0040 article-title: Data-driven heating and cooling load predictions for non-residential buildings based on support vector machine regression and NARX Recurrent Neural Network: a comparative study on district scale[J] publication-title: Energy doi: 10.1016/j.energy.2018.09.068 – volume: 243 start-page: 274 year: 2019 ident: 10.1016/j.enbuild.2020.110658_b0020 article-title: Flexible dispatch of a building energy system using building thermal storage and battery energy storage[J] publication-title: Appl. Energy doi: 10.1016/j.apenergy.2019.03.187 – ident: 10.1016/j.enbuild.2020.110658_b0150 doi: 10.18653/v1/D15-1166 – volume: 39 start-page: 508 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0075 article-title: On the feature engineering of building energy data mining[J] publication-title: Sustain. Cities Soc. doi: 10.1016/j.scs.2018.02.016 – ident: 10.1016/j.enbuild.2020.110658_b0155 – volume: 236 start-page: 700 year: 2019 ident: 10.1016/j.enbuild.2020.110658_b0080 article-title: Assessment of deep recurrent neural network-based strategies for short-term building energy predictions[J] publication-title: Appl. Energy doi: 10.1016/j.apenergy.2018.12.004 – ident: 10.1016/j.enbuild.2020.110658_b0100 doi: 10.1109/SIU.2018.8404313 – ident: 10.1016/j.enbuild.2020.110658_b0160 – ident: 10.1016/j.enbuild.2020.110658_b0145 – volume: 65 start-page: 352 year: 2013 ident: 10.1016/j.enbuild.2020.110658_b0035 article-title: Energy analysis of a building using artificial neural network: a review[J] publication-title: Energy Build. doi: 10.1016/j.enbuild.2013.06.007 – volume: 188 start-page: 116085 year: 2019 ident: 10.1016/j.enbuild.2020.110658_b0115 article-title: Multi-step ahead forecasting of heat load in district heating systems using machine learning algorithms[J] publication-title: Energy doi: 10.1016/j.energy.2019.116085 – volume: 137 start-page: 617 year: 2017 ident: 10.1016/j.enbuild.2020.110658_b0025 article-title: International review of district heating and cooling[J] publication-title: Energy doi: 10.1016/j.energy.2017.04.045 – volume: 4 start-page: 371 year: 2016 ident: 10.1016/j.enbuild.2020.110658_b0140 article-title: Deep recurrent models with fast-forward connections for neural machine translation[J] publication-title: Trans. Assoc. Comput. Linguist. doi: 10.1162/tacl_a_00105 – volume: 190 start-page: 49 year: 2019 ident: 10.1016/j.enbuild.2020.110658_b0170 article-title: GMM clustering for heating load patterns in-depth identification and prediction model accuracy improvement of district heating system[J] publication-title: Energy Build. doi: 10.1016/j.enbuild.2019.02.014 – volume: 240 start-page: 35 year: 2019 ident: 10.1016/j.enbuild.2020.110658_b0070 article-title: Deep learning-based feature engineering methods for improved building energy prediction[J] publication-title: Appl. Energy doi: 10.1016/j.apenergy.2019.02.052 – ident: 10.1016/j.enbuild.2020.110658_b0135 doi: 10.21437/Interspeech.2012-65 – volume: 80 start-page: 46 year: 2014 ident: 10.1016/j.enbuild.2020.110658_b0060 article-title: Hourly cooling load forecasting using time-indexed ARX models with two-stage weighted least squares regression[J] publication-title: Energy Convers. Manage. doi: 10.1016/j.enconman.2013.12.060 – volume: 212 start-page: 372 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0085 article-title: Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks[J] publication-title: Appl. Energy doi: 10.1016/j.apenergy.2017.12.051 – volume: 70 start-page: 2861 issue: 16–18 year: 2007 ident: 10.1016/j.enbuild.2020.110658_b0110 article-title: Methodology for long-term prediction of time series[J] publication-title: Neurocomputing doi: 10.1016/j.neucom.2006.06.015 – volume: 171 start-page: 11 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0045 article-title: Random Forest based hourly building energy prediction[J] publication-title: Energy Build. doi: 10.1016/j.enbuild.2018.04.008 – volume: 54 start-page: 225 year: 2012 ident: 10.1016/j.enbuild.2020.110658_b0055 article-title: Building hourly thermal load prediction using an indexed ARX model[J] publication-title: Energy Build. doi: 10.1016/j.enbuild.2012.08.007 – volume: 82 start-page: 1675 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0010 article-title: Prosumer based energy management and sharing in smart grid[J] publication-title: Renew. Sustain. Energy Rev. doi: 10.1016/j.rser.2017.07.018 – start-page: 145 year: 2008 ident: 10.1016/j.enbuild.2020.110658_b0120 article-title: Long term time series prediction with multi-input multi-output local learning[J] publication-title: Proc. 2nd ESTSP – ident: 10.1016/j.enbuild.2020.110658_b0090 doi: 10.1109/IECON.2016.7793413 – volume: 39 start-page: 7067 issue: 8 year: 2012 ident: 10.1016/j.enbuild.2020.110658_b0130 article-title: A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition[J] publication-title: Expert Syst. Appl. doi: 10.1016/j.eswa.2012.01.039 – volume: 204 start-page: 181 year: 2017 ident: 10.1016/j.enbuild.2020.110658_b0030 article-title: Modeling of district load forecasting for distributed energy system[J] publication-title: Appl. Energy doi: 10.1016/j.apenergy.2017.07.009 – volume: 128 start-page: 225 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0065 article-title: Effect of input variables on cooling load prediction accuracy of an office building publication-title: Appl. Therm. Eng. doi: 10.1016/j.applthermaleng.2017.09.007 – volume: 99 start-page: 1 year: 2018 ident: 10.1016/j.enbuild.2020.110658_b0005 article-title: Day-ahead optimal operation for multi-energy residential systems with renewables publication-title: IEEE Trans. Sustain. Energy – ident: 10.1016/j.enbuild.2020.110658_b0095 doi: 10.1109/CAC.2017.8242744 |
| SSID | ssj0006571 |
| Score | 2.4950461 |
| Snippet | [Display omitted]
Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and... Accurate and reliable multi-step-ahead thermal load prediction is generally considered as the basis of model predictive control and multi-energy dispatch of... |
| SourceID | proquest crossref elsevier |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 110658 |
| SubjectTerms | Attention mechanism Coders Deep learning Encoders-Decoders Energy Iterative methods Load Long short-term memory Model accuracy Multi-step-ahead prediction Predictions Predictive control Sequence-to-sequence Sequences Temporal dependency Thermal analysis Time series |
| Title | Multi-step-ahead prediction of thermal load in regional energy system using deep learning method |
| URI | https://dx.doi.org/10.1016/j.enbuild.2020.110658 https://www.proquest.com/docview/2491971787 |
| Volume | 233 |
| WOSCitedRecordID | wos000712412900005&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 customDbUrl: eissn: 1872-6178 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0006571 issn: 0378-7788 databaseCode: AIEXJ dateStart: 19950301 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Nb9QwELWg5QAHxKcoFOQDN5SSOE7sHCtUBKiqEBSx6iXEjt1mKcmqu4vKv2fssbOwBRWQuEQra8eb9XuxZ0aZN4Q8ZblwSmBtYvMsT7goikQqmyYKDmctIGDIrEd6XxwcyMmkehvKFee-nYDoe3l-Xs3-K9QwBmC70tm_gHucFAbgM4AOV4Adrn8EvC-pTQC8WdLATutlANpOR8_QOXxfAJfToWmxmOUYs4EGqwBR2vnZ0ucQWmNmsbHEceg2_VMuH21c8l2F_tqjj76_9Nt787npxvRAh-nWo5ORTkcng__eu-XwbTmadn7so-mn8WQNiQmWuXeZsTQTs2UXKmawSguiViGwlV_cgRlqYVzYzTGxMN1xKhDwHyCaZ75uoUS19zWh7Pdubjc1S_Occy6ukk0migq2683d13uTN-MJXRY-EB_vZVXZ9fyXP_Y7n2Xt9PYuyeEtcjPEEnQXOXCbXDH9HXLjB4XJu-TTOhvoig10sDSwgTo20K6nkQ0U2UCRDdSzgTo20MgGimy4Rz683Dt88SoJTTUSnedikYC73cpKNkalXGUta1PLVcOVMIyXKQQDnDnVNWYqqVKp20w1umq0aYvCClGw_D7Z6IfePCDUZkXDSm21EpqXSkrLuHWJSfC7LDjaW4THZat1UJx3jU9O6_hq4bQOq1271a5xtbfIzmg2Q8mVywxkxKQOfiP6gzUQ6TLT7YhhHZ7hec14lVUiAzI9_PeZH5Hrq4dim2wszpbmMbmmvy66-dmTwMjvN5ygiA |
| 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=Multi-step-ahead+prediction+of+thermal+load+in+regional+energy+system+using+deep+learning+method&rft.jtitle=Energy+and+buildings&rft.au=Lu%2C+Yakai&rft.au=Tian%2C+Zhe&rft.au=Zhou%2C+Ruoyu&rft.au=Liu%2C+Wenjing&rft.date=2021-02-15&rft.pub=Elsevier+B.V&rft.issn=0378-7788&rft.volume=233&rft_id=info:doi/10.1016%2Fj.enbuild.2020.110658&rft.externalDocID=S0378778820334447 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0378-7788&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0378-7788&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0378-7788&client=summon |