Robust-mv-M-LSTM-CI: Robust Energy Consumption Forecasting in Commercial Buildings during the COVID-19 Pandemic
<p>Global direct primary energy consumption, 1800–2022. Direct primary energy consumption does not consider inefficiencies in fossil fuel production.</p> "> Figure 2
<p>Australian energy consumption, by state and territory, 2019–2021. New South Wales includes the Australian Capital Territory.</p> "> Figure 3
<p>The positioning of Hawthorn Campus and Wantirna Campus within Metropolitan Melbourne, Victoria, Australia.</p> "> Figure 4
<p>Timeline of the Australia COVID-19 pandemic.</p> "> Figure 5
<p>The overall workflow of the proposed method.</p> "> Figure 6
<p>Overview of <math display="inline"><semantics> <mrow> <mi mathvariant="monospace">mv</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">M</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">LSTM</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">CI</mi> </mrow> </semantics></math> model. <math display="inline"><semantics> <mrow> <mi mathvariant="italic">ADC</mi> <mtext mathvariant="italic">-</mtext> </mrow> </semantics></math><span class="html-italic">19</span>: Accumulated Daily COVID-19 data, <math display="inline"><semantics> <mrow> <mi mathvariant="italic">EC</mi> <mtext mathvariant="italic">-</mtext> </mrow> </semantics></math><span class="html-italic">15</span>: Energy consumption in 15-min intervals, <math display="inline"><semantics> <mrow> <mi mathvariant="italic">LBT</mi> <mtext mathvariant="italic">-</mtext> </mrow> </semantics></math><span class="html-italic">15</span>: labeled time in 15-min intervals.</p> "> Figure 7
<p>The overall training workflow of <math display="inline"><semantics> <mrow> <mi mathvariant="monospace">Robust</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">mv</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">M</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">LSTM</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">CI</mi> </mrow> </semantics></math> model. The gray block indicates that they freeze.</p> "> Figure 8
<p>The process of adversarial training in baseline models.</p> "> Figure 9
<p>Prediction for energy consumption every day of forecasting models on <math display="inline"><semantics> <mrow> <mi mathvariant="italic">DatasetS</mi> </mrow> </semantics></math><span class="html-italic">1</span>.</p> "> Figure 10
<p>Prediction for energy consumption every day of forecasting models on <math display="inline"><semantics> <mrow> <mi mathvariant="italic">DatasetS</mi> </mrow> </semantics></math><span class="html-italic">2</span>.</p> "> Figure 11
<p>Prediction for energy consumption every day of forecasting models on <math display="inline"><semantics> <mrow> <mi mathvariant="italic">DatasetS</mi> </mrow> </semantics></math><span class="html-italic">3</span>.</p> "> Figure 12
<p>The detail on prediction of <math display="inline"><semantics> <mrow> <mi mathvariant="monospace">Robust</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">mv</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">M</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">LSTM</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">CI</mi> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi mathvariant="monospace">mv</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">M</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">LSTM</mi> <mtext mathvariant="italic">-</mtext> <mi mathvariant="monospace">CI</mi> </mrow> </semantics></math> from 17 July and 18 July 2021.</p> ">
Abstract
:1. Introduction
2. Background
2.1. Energy Forecasting
2.2. Robustness of AI Model
3. Data Used in This Study
3.1. Data Collection
3.2. Daily COVID-19 Cases in Victoria
4. Methodology
4.1. Overview
4.2. Preprocessing Data
4.3. Training mv-M-LSTM-CI Model
4.4. Generating Adversarial Example
4.5. Robust-mv-M-LSTM-CI Model
5. Baseline Models
6. Experiment
6.1. Metric
6.2. Results and Discussion
7. Conclusions
- In 1, achieved an MPAE of 0.062 and an NRMSE of 0.049, slightly higher than ’s MPAE of 0.061 and NRMSE of 0.047. Despite minor increases in error rates and a slight decrease in the score, the proposed model remains competitive and outperforms other adversarially trained models.
- In 2, showed superior performance with an MPAE of 0.065, NRMSE of 0.042, and an score of 0.878, improving significantly over . These results demonstrate ’s robustness and higher accuracy, outperforming other adversarially trained models.
- In 3, excelled with an MPAE of 0.087, NRMSE of 0.024, and score of 0.945, showing substantial improvements over . This highlights the model’s robustness and enhanced accuracy, consistently outperforming other adversarially trained models.
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Sadigov, R. Rapid growth of the world population and its socioeconomic results. Sci. World J. 2022, 2022, 8110229. [Google Scholar] [CrossRef] [PubMed]
- Online Global Direct Primary Energy Consumption. Available online: https://ourworldindata.org/grapher/global-primary-energy (accessed on 8 November 2023).
- Online Buildings. Available online: https://www.iea.org/energy-system/buildings (accessed on 8 November 2023).
- Online Commercial Buildings Energy Consumption Baseline Study 2022. Available online: https://www.energy.gov.au/publications/commercial-buildings-energy-consumption-baseline-study-2022 (accessed on 8 November 2023).
- Online Electricity Prices Fall and COVID Spikes Residential Demand. Available online: https://www.accc.gov.au/media-release/electricity-prices-fall-and-covid-spikes-residential-demand (accessed on 8 November 2023).
- Online Working from Home’s Impact on Electricity Use in the Pandemic. Available online: https://www.nber.org/digest/202012/working-homes-impact-electricity-use-pandemic (accessed on 8 November 2023).
- Online Speed and Surprises: Decline and Recovery of Global Electricity Use in COVID’s First Seven Months. Available online: https://news.stanford.edu/2022/02/11/fall-rise-electricity-use-early-pandemic/ (accessed on 8 November 2023).
- IEA. Global Energy Review 2021—Analysis. 2021. Available online: https://www.iea.org/reports/global-energy-review-2021 (accessed on 18 July 2024).
- Grozinger, P.; Parsons, S. The COVID-19 Outbreak and Australia’s Education and Tourism Exports, Bulletin—December Quarter. Bulletin. 2020. Available online: https://www.rba.gov.au/publications/bulletin/2020/dec/the-covid-19-outbreak-and-australias-education-and-tourism-exports.html (accessed on 18 July 2024).
- Australian Energy Update 2021. Available online: https://www.energy.gov.au/publications/australian-energy-update-2021 (accessed on 17 May 2023).
- States and Territories. Available online: https://www.energy.gov.au/data/states-and-territories (accessed on 1 September 2023).
- Tracking Victoria’s Energy Transition 2020. Available online: https://www.sustainability.vic.gov.au/research-data-and-insights/research/research-reports/tracking-victorias-energy-transition-2020 (accessed on 1 September 2023).
- Jogunola, O.; Morley, C.; Akpan, I.J.; Tsado, Y.; Adebisi, B.; Yao, L. En-ergy Consumption in Commercial Buildings in a Post-COVID-19 World. IEEE Eng. Manag. Rev. 2022, 50, 54–64. [Google Scholar] [CrossRef]
- Abulibdeh, A.; Zaidan, E.; Jabbar, R. The impact of COVID-19 pandemic on electricity consumption and electricity demand forecasting accuracy: Empirical evidence from the state of Qatar. Energy Strategy Rev. 2022, 44, 100980. [Google Scholar] [CrossRef]
- Hosseini, S.M.; Carli, R.; Dotoli, M. Robust Optimal Demand Response of Energy-efficient Commercial Buildings. In Proceedings of the 2022 European Control Conference (ECC), London, UK, 12–15 July 2022; pp. 1–6. [Google Scholar] [CrossRef]
- Shin, S.-Y.; Woo, H.-G. Energy Consumption Forecasting in Korea Using Machine Learning Algorithms. Energies 2022, 15, 4880. [Google Scholar] [CrossRef]
- Di Piazza, M.C.; La Tona, G.; Luna, M.; Di Piazza, A. A two-stage Energy Management System for smart buildings reducing the impact of demand uncertainty. Energy Build. 2017, 139, 1–9. [Google Scholar] [CrossRef]
- Aldahdooh, A.; Hamidouche, W.; Fezza, S.A.; Déforges, O. Adversarial example detection for DNN models: A review and experimental comparison. Artif. Intell. Rev. 2022, 55, 4403–4462. [Google Scholar] [CrossRef]
- Bai, T.; Luo, J.; Zhao, J.; Wen, B.; Wang, Q. Recent Advances in Adversarial Training for Adversarial Robustness. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence (IJCAI-21), Montreal, QC, Canda, 19–27 August 2021; pp. 4312–4321. [Google Scholar] [CrossRef]
- Dinh, T.; Thirunavukkarasu, G.; Seyedmahmoudian, M.; Mekhilef, S.; Stojcevski, A. Energy Consumption Forecasting in Commercial Buildings during the COVID-19 Pandemic: A Multivariate Multilayered Long-Short Term Memory Time-series Model with Knowledge Injection. Sustainability 2023, 15, 12951. [Google Scholar] [CrossRef]
- Bourdeau, M.; Zhai, X.; Nefzaoui, E.; Guo, X.; Chatellier, P. Modeling and forecasting building energy consumption: A review of data-driven techniques. Sustain. Cities Soc. 2019, 48, 101533. [Google Scholar] [CrossRef]
- Runge, J.; Zmeureanu, R. Forecasting Energy Use in Buildings Using Artificial Neural Networks: A Review. Energies 2019, 12, 3254. [Google Scholar] [CrossRef]
- Fathi, S.; Srinivasan, R.; Fenner, A.; Fathi, S. Machine learning applications in urban building energy performance forecasting: A systematic review. Renew. Sustain. Energy Rev. 2020, 133, 110287. [Google Scholar] [CrossRef]
- Liu, T.; Tan, Z.; Xu, C.; Chen, H.; Li, Z. Study on deep reinforcement learning techniques for building energy consumption forecasting. Energy Build. 2020, 208, 109675. [Google Scholar] [CrossRef]
- Somu, N.; Raman, M.G.; Ramamritham, K. A hybrid model for building energy consumption forecasting using long short term memory networks. Appl. Energy 2020, 261, 114131. [Google Scholar] [CrossRef]
- Gassar, A.A.A.; Cha, S.H. Energy prediction techniques for large-scale buildings towards a sustainable built environment: A review. Energy Build. 2020, 224, 110238. [Google Scholar] [CrossRef]
- Dong, Z.; Liu, J.; Liu, B.; Li, K.; Li, X. Hourly energy consumption prediction of an office building based on ensemble learning and energy consumption pattern classification. Energy Build. 2021, 241, 110929. [Google Scholar] [CrossRef]
- AEMO. Quarterly Energy Dynamics Q2 2020 Market Insights and WA Market Operations. July 2020. Available online: https://aemo.com.au/-/media/files/major-publications/qed/2020/qed-q2-2020.pdf?la=en (accessed on 18 July 2024).
- Statistical Review of World Energy, BP. 2021. Available online: https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy-economics/statistical-review/bp-stats-review-2021-full-report.pdf (accessed on 18 July 2024).
- Irena, Renewable Capacity Statistics. 2021. Available online: https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2021/Apr/IRENA_RE_Capacity_Statistics_2021.pdf (accessed on 18 July 2024).
- Madhukumar, M.; Sebastian, A.; Liang, X.; Jamil, M.; Shabbir, M. Regression model-based short-term load forecasting for university campus load. IEEE Access 2022, 10, 8891–8905. [Google Scholar] [CrossRef]
- Zhang, C.; Li, J.; Zhao, Y.; Li, T.; Chen, Q.; Zhang, X. A hybrid deep learn-ing-based method for short-term building energy load prediction combined with an interpretation process. Energy Build. 2020, 225, 110301. [Google Scholar] [CrossRef]
- Zeng, A.; Ho, H.; Yu, Y. Prediction of building electricity usage using Gaussian Process Regression. J. Build. Eng. 2020, 28, 101054. [Google Scholar] [CrossRef]
- Massaoudi, M.; Refaat, S.; Chihi, I.; Trabelsi, M.; Oueslati, F.; Abu-Rub, H. A novel stacked generalization ensemble-based hybrid LGBM-XGB-MLP model for Short-Term Load Forecasting. Energy 2021, 214, 118874. [Google Scholar] [CrossRef]
- Zor, K.; Çelik, Ö.; Timur, O.; Teke, A. Short-Term Building Electrical Energy Consumption Forecasting by Employing Gene Expression Programming and GMDH Networks. Energies 2020, 13, 1102. [Google Scholar] [CrossRef]
- Zhu, Y.-R.; Wang, J.-G.; Sun, Y.-Q.; Wu, J.-J.; Zhao, G.-Q.; Yao, Y.; Liu, J.-L.; Chen, H.-L. Short-term Load Forecasting of CCHP System Based on PSO-LSTM. In Proceedings of the 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS), Xiangtan, China, 12–14 May 2023; pp. 639–644. [Google Scholar] [CrossRef]
- Zhang, L.; Wen, J. Active learning strategy for high fidelity short-term da-ta-driven building energy forecasting. Energy Build. 2021, 244, 111026. [Google Scholar] [CrossRef]
- Yu, B.; Li, J.; Liu, C.; Sun, B. A novel short-term electrical load forecasting framework with intelligent feature engineering. Appl. Energy 2022, 327, 120089. [Google Scholar] [CrossRef]
- Zhang, J.M.; Harman, M.; Ma, L.; Liu, Y. Machine Learning Testing: Survey, Landscapes and Horizons. IEEE Trans. Softw. Eng. 2022, 48, 1–36. [Google Scholar] [CrossRef]
- Silva, S.; Najafirad, P. Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey. arXiv 2020, arXiv:2007.00753. [Google Scholar]
- Carlini, N.; Wagner, D. Towards Evaluating the Robustness of Neural Networks. In Proceedings of the 2017 IEEE Symposium on Security and Privacy (SP), San Jose, CA, USA, 22–26 May 2017; pp. 39–57. [Google Scholar] [CrossRef]
- Szegedy, C.; Zaremba, W.; Sutskever, I.; Bruna, J.; Erhan, D.; Goodfellow, I.; Fergus, R. Intriguing properties of neural networks. arXiv 2014, arXiv:1312.6199. [Google Scholar]
- Zhang, W.E.; Sheng, Q.Z.; Alhazmi, A.; Li, C. Adversarial Attacks on Deep-learning Models in Natural Language Processing: A Survey. ACM Trans. Intell. Syst. Technol. 2020, 11, 1–41. [Google Scholar] [CrossRef]
- Wu, S.; Xiao, X.; Ding, Q.; Zhao, P.; Wei, Y.; Huang, J. Adversarial Sparse Transformer for Time Series Forecasting. In Advances in Neural Information Processing Systems; Curran Associates, Inc.: Red Hook, NY, USA, 2020; Volume 33, pp. 17105–17115. [Google Scholar]
- Carlini, N.; Wagner, D. Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods. In Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security, Dallas, TX, USA, 3 November 2017; pp. 3–14. [Google Scholar] [CrossRef]
- Dinh, T.N.; Thirunavukkarasu, G.S.; Seyedmahmoudian, M.; Mekhilef, S.; Stojcevski, A. Predicting Commercial Building Energy Consumption Using a Multivariate Multilayered Long-Short Term Memory Time-Series Model. Appl. Sci. 2023, 13, 7775. [Google Scholar] [CrossRef]
- Australia Bucket List Timeline of Every Victoria Lockdown (Dates & Restrictions). 2021. Available online: https://bigaustraliabucketlist.com/victoria-lockdowns-dates-restrictions/ (accessed on 18 July 2024).
- Victoria, P. State of Emergency Declared in Victoria over COVID-19. 2020. Available online: https://www.premier.vic.gov.au/state-emergency-declared-victoria-over-covid-19 (accessed on 18 July 2024).
- Association, N. COVID-19 Update|7 July 2020. Available online: https://www.nra.net.au/covid-19-update-7-july-2020 (accessed on 18 July 2024).
- 9NEWS Coronavirus: What Changes under Melbourne’s Stage 4 Restrictions. 2020. Available online: https://www.9news.com.au/national/coronavirus-melbourne-stage-4-restrictions-explained-curfews-lockdown-what-is-open-closed-changes-covid19/652ee0f9-cc22-41df-b465-9a765a45c496 (accessed on 18 July 2024).
- Australian Energy Update 2020. Department of Industry, Science, Energy and Resources, September 2020. Available online: https://www.energy.gov.au/sites/default/files/Australian%20Energy%20Statistics%202020%20Energy%20Update%20Report_0.pdf (accessed on 18 July 2024).
- Victoria, P. Statement from the Premier. 2021. Available online: https://www.premier.vic.gov.au/statement-premier-92 (accessed on 18 July 2024).
- 7News COVID Lockdown: Victoria Announces Major Restrictions to Combat Coronavirus Outbreak. 2021. Available online: https://7news.com.au/lifestyle/health-wellbeing/covid-lockdown-victoria-announces-major-restrictions-to-combat-coronavirus-outbreak-c-2148043 (accessed on 17 May 2023).
- Victoria, P. Extended Lockdown and Stronger Borders to Keep Us Safe. 2021. Available online: https://www.premier.vic.gov.au/extended-lockdown-and-stronger-borders-keep-us-safe (accessed on 18 July 2024).
- Victoria, P. Seven Day Lockdown to Keep Victorians Safe. 2021. Available online: https://www.premier.vic.gov.au/seven-day-lockdown-keep-victorians-safe (accessed on 18 July 2024).
- Zhou, L.; Zhao, C.; Liu, N.; Yao, X.; Cheng, Z. Improved LSTM-based deep learning model for COVID-19 prediction using optimized approach. Eng. Appl. Artif. Intell. 2023, 122, 106157. [Google Scholar] [CrossRef] [PubMed]
- Charfeddine, L.; Zaidan, E.; Alban, A.Q.; Bennasr, H.; Abulibdeh, A. Modeling and forecasting electricity consumption amid the COVID-19 pandemic: Machine learning vs. nonlinear econometric time series mod-els. Sustain. Cities Soc. 2023, 98, 104860. [Google Scholar] [CrossRef]
- Eykholt; Kevin; Evtimov, I.; Fernandes, E.; Li, B.; Rahmati, A.; Xiao, C.; Prakash, A.; Kohno, T.; Song, D. Robust Physical-World Attacks on Deep Learning Visual Classification. In Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 19–21 June 2018; pp. 1625–1634. [Google Scholar] [CrossRef]
- Goodfellow, I.; Shlens, J.; Szegedy, C. Explaining and Harnessing Adversarial Examples. In Proceedings of the 3rd International Conference on Learning Representations, ICLR, San Diego, CA, USA, 7–9 May 2015. [Google Scholar]
ID | Forecasting Model | Year | Country | Forecast Horizon | Ref | RMSE | MAE | MAPE | |
---|---|---|---|---|---|---|---|---|---|
1 | Rational Quadratic Gaussian Process Regression (GPR) | 2022 | Canada | day ahead | [31] | 301.76 kW | 203.67 kW | 0.97 | |
2 | Gated Recurrent Unit—Artificial Neural Network (GRU-ANN) | 2020 | China | 1 h ahead | [32] | 484.29 kW | 311.43 kW | 0.925 | |
3 | Gaussian Process Regression (GPR) | 2020 | China | 15 min ahead | [33] | 23.78 kWh | 0.88 | ||
4 | Stacked XGB-LGBM-MLP | 2021 | Qatar | day ahead | [34] | 1509.74 MW | 1070.67 MW | 2.69% | 0.99 |
5 | Gene Expression Programming (GEP) | 2020 | Turkey | 1 h ahead | [35] | 26.668 kWh | 0.641% | 0.99955 | |
6 | LSTM Network Prediction Model Optimized By Particle Swarm Optimization (PSO-LSTM) | 2023 | China | 1 h ahead | [36] | 543 kWh | 0.923 | ||
7 | MARS + Active Learning | 2021 | USA | 15 min ahead | [37] | 0.77 | |||
8 | Deep STL-CLSTM | 2022 | China | 15 min ahead | [38] | 60.74 kW | 45.97 kW | 8.95% | 0.77 |
9 | Improved Sine Cosine Optimization Algorithm based LSTM (ISCOA-LSTM) | 2020 | India | 30 min ahead | [25] | 0.0559 kWh | 0.0369 kWh | 3.3159% | |
10 | Deep Deterministic Policy Gradient (DDPG) | 2020 | China | 1 h ahead | [24] | 30.440 kW | 13.175 kW | 0.978 |
Dataset | Model | MAPE | NRMSE | |
---|---|---|---|---|
1 | 0.062 | 0.049 | 0.926 | |
[20] | 0.061 | 0.047 | 0.931 | |
0.119 | 0.082 | 0.791 | ||
Adv-M-LSTM | 0.128 | 0.086 | 0.772 | |
Adv-LSTM | 0.305 | 0.136 | 0.426 | |
Adv-BiLSTM | 0.353 | 0.099 | 0.696 | |
Adv-LR | 0.369 | 0.091 | 0.745 | |
2 | 0.065 | 0.042 | 0.878 | |
[20] | 0.093 | 0.062 | 0.729 | |
0.094 | 0.064 | 0.715 | ||
Adv-M-LSTM | 0.121 | 0.078 | 0.570 | |
Adv-LSTM | 0.172 | 0.097 | 0.336 | |
Adv-BiLSTM | 0.219 | 0.070 | 0.658 | |
Adv-LR | 0.235 | 0.084 | 0.498 | |
3 | 0.087 | 0.024 | 0.945 | |
[20] | 0.158 | 0.033 | 0.895 | |
0.173 | 0.04 | 0.847 | ||
adv-M-LSTM | 0.314 | 0.051 | 0.752 | |
adv-LSTM | 0.841 | 0.083 | 0.338 | |
adv-BiLSTM | 0.914 | 0.052 | 0.735 | |
adv-LR | 0.972 | 0.053 | 0.728 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Dinh, T.N.; Thirunavukkarasu, G.S.; Seyedmahmoudian, M.; Mekhilef, S.; Stojcevski, A. Robust-mv-M-LSTM-CI: Robust Energy Consumption Forecasting in Commercial Buildings during the COVID-19 Pandemic. Sustainability 2024, 16, 6699. https://doi.org/10.3390/su16156699
Dinh TN, Thirunavukkarasu GS, Seyedmahmoudian M, Mekhilef S, Stojcevski A. Robust-mv-M-LSTM-CI: Robust Energy Consumption Forecasting in Commercial Buildings during the COVID-19 Pandemic. Sustainability. 2024; 16(15):6699. https://doi.org/10.3390/su16156699
Chicago/Turabian StyleDinh, Tan Ngoc, Gokul Sidarth Thirunavukkarasu, Mehdi Seyedmahmoudian, Saad Mekhilef, and Alex Stojcevski. 2024. "Robust-mv-M-LSTM-CI: Robust Energy Consumption Forecasting in Commercial Buildings during the COVID-19 Pandemic" Sustainability 16, no. 15: 6699. https://doi.org/10.3390/su16156699