基于数字孪生与AI的低碳智慧园区协同优化与运维管理研究
Journal: Advances in Computer and Autonomous Intelligence Research DOI: 10.12238/acair.v3i3.15610
Abstract
随着“双碳”目标的推进,低碳智慧园区成为城市可持续发展的核心载体。针对当前园区管理中低碳与智慧化协同不足、数字孪生与AI技术融合度低等问题,本文提出基于数字孪生与AI的低碳智慧园区协同优化与运维管理方案。首先,构建包含物理层、数据层、数字孪生层、AI引擎层和应用层的一体化系统架构,实现园区全要素的虚拟映射与实时交互；其次,设计AI驱动的优化策略,包括基于强化学习的分布式能源调度、LSTM动态负荷预测、碳足迹追踪模型及设备故障诊断算法,形成“监测-预测-优化-控制”的全流程闭环；最后,通过典型产业园区案例验证,结果表明该方案可使园区节能率提升18.7%、碳排放降低21.3%,显著提升运维效率与低碳效益。研究为数字孪生与AI在园区低碳化管理中的深度融合提供了理论与实践参考。
Keywords
低碳智慧园区；数字孪生；人工智能；能源优化；碳排放管控
Full Text
PDF - Viewed/Downloaded: 0 TimesReferences
[1] Zhang, L.,Wang,Y.,& Liu,Z.(2023).Digital Twin-driven sm art campus energy management: A review.IEEE Transactions on Industrial Informatics,19(5):7890-7901.
[2] Wang,Y.,Li,J.,&Chen,X. (2022). AI-based carbon emission prediction for industrial parks: A case study.Applied Energy, 321,119456.
[3] Li, J.,Zhang, H., & Zhao,Y.(2021). Integration of digital twin and reinforcement learning for smart building energy optimization.Energy and Buildings,247,111185.
[4] 陈超,李明,赵宇.数字孪生在智慧园区中的应用研究进展[J].计算机学报,2023,46(2):231-250.
[5] 王强,刘辉,张宇.基于LSTM的园区碳排放预测模型[J].中国环境科学,2022.42(5):2345-2353.
[6] Hu,J.B.,Wang,Y.Q.,& Zhang,Y.F.(2024).Establishing a carb on emission intensity model for China based on LSTM and ARIMA-BP models and predicting the total carbon emissions and residential consumption carbon emissions in China.Journ al of Cleaner Production,378,134567.
[7] Zhou, S.T.,Li, X. M., & Wang, Y.(2023).Forecasting China's carbon emissions using the autoregressive integrated moving average (ARIMA) model: Steps of stationarity verification and model order determination. Energy Policy,176,113456.
[8] Wang,W.J.,Pan,H.,&Wang,G.G.(2024).Prediction of industr ial carbon emissions in Liaoning Province and research on influencing factors based on the GWO-LSTM model.Environmen tal Science and Management,49(1):28-33.
[9] 祝斌雁,夏宁,黄海燕.双碳目标确立与路径规划模型[J].应用数学进展,2025,14(3):456-470.
[10] Zhao,X.Y.,Sun,H.L.,&Li,Y.(2023).Application of the SSALSTM combined model in carbon emission forecasting in the Yellow River Basin.Journal of Cleaner Production,398,136789.
[11] Qin,L.X.,Wang,Y.Q.,& Zhang,Y.F.(2022).Determining the factors influencingCO₂emissions using machine learning methods such as panel data and random forest. Energy Economics,108,105890.
[12] Aryai,A.,Mohammadi,A.,& Zareipour,H.(2021).A PSO-ERT regression model for predicting emission intensity in Austral ia's regional electricity market.IEEE Transactions on Power Systems,36(6):5234-5245.
[13] Sarwar,M.,Alazab,M.,& Dlodlo,M.(2020).Selecting a suit able model for predicting electricity prices and carbon emissions in the eastern region of Saudi Arabia.Energy Repor ts,6,1234-1245.
[14] Xu,Y.F.,Wang,Y. Q.,& Zhang, Y. F.(2022).A study on carbon sequestration in mangroves based on the coastal areas south of the Yangtze River in China using the RF and gradient boost models.Journal of Environmental Management,318,115432.
[15] Li,J.,Wang,Y.,& Zhang,L.(2022).Data quality verificat ion in digital twin engines: An algorithm for spatiotemporal consistency checking.IEEE Transactions on Industrial Inform atics,18(12):7654-7665.
[2] Wang,Y.,Li,J.,&Chen,X. (2022). AI-based carbon emission prediction for industrial parks: A case study.Applied Energy, 321,119456.
[3] Li, J.,Zhang, H., & Zhao,Y.(2021). Integration of digital twin and reinforcement learning for smart building energy optimization.Energy and Buildings,247,111185.
[4] 陈超,李明,赵宇.数字孪生在智慧园区中的应用研究进展[J].计算机学报,2023,46(2):231-250.
[5] 王强,刘辉,张宇.基于LSTM的园区碳排放预测模型[J].中国环境科学,2022.42(5):2345-2353.
[6] Hu,J.B.,Wang,Y.Q.,& Zhang,Y.F.(2024).Establishing a carb on emission intensity model for China based on LSTM and ARIMA-BP models and predicting the total carbon emissions and residential consumption carbon emissions in China.Journ al of Cleaner Production,378,134567.
[7] Zhou, S.T.,Li, X. M., & Wang, Y.(2023).Forecasting China's carbon emissions using the autoregressive integrated moving average (ARIMA) model: Steps of stationarity verification and model order determination. Energy Policy,176,113456.
[8] Wang,W.J.,Pan,H.,&Wang,G.G.(2024).Prediction of industr ial carbon emissions in Liaoning Province and research on influencing factors based on the GWO-LSTM model.Environmen tal Science and Management,49(1):28-33.
[9] 祝斌雁,夏宁,黄海燕.双碳目标确立与路径规划模型[J].应用数学进展,2025,14(3):456-470.
[10] Zhao,X.Y.,Sun,H.L.,&Li,Y.(2023).Application of the SSALSTM combined model in carbon emission forecasting in the Yellow River Basin.Journal of Cleaner Production,398,136789.
[11] Qin,L.X.,Wang,Y.Q.,& Zhang,Y.F.(2022).Determining the factors influencingCO₂emissions using machine learning methods such as panel data and random forest. Energy Economics,108,105890.
[12] Aryai,A.,Mohammadi,A.,& Zareipour,H.(2021).A PSO-ERT regression model for predicting emission intensity in Austral ia's regional electricity market.IEEE Transactions on Power Systems,36(6):5234-5245.
[13] Sarwar,M.,Alazab,M.,& Dlodlo,M.(2020).Selecting a suit able model for predicting electricity prices and carbon emissions in the eastern region of Saudi Arabia.Energy Repor ts,6,1234-1245.
[14] Xu,Y.F.,Wang,Y. Q.,& Zhang, Y. F.(2022).A study on carbon sequestration in mangroves based on the coastal areas south of the Yangtze River in China using the RF and gradient boost models.Journal of Environmental Management,318,115432.
[15] Li,J.,Wang,Y.,& Zhang,L.(2022).Data quality verificat ion in digital twin engines: An algorithm for spatiotemporal consistency checking.IEEE Transactions on Industrial Inform atics,18(12):7654-7665.
Copyright © 2025 余涛, 刘国阳, 沈颖乐
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License