使用者行为影响下大型公共建筑能耗动态分析与智能优化策略研究

Chinese large-scale public buildings consume tremendous amounts of energy, which have become a key factor restricting the green development of the construction industry. To make energy conservation of the large-scale public buildings relies on not only the optimization of building performances, but also reducing the negative impact of occupant behavior. This study will firstly classify the large-scale public buildings based on the characteristics of their energy consumptions, and then identify the critical factors of building performance and occupant behavior which have significant impact on energy consumption. An Internet of Things and BIM based visual monitoring network is then able to be built referring to these critical factors. This visual monitoring network will be used to measure and save the data of energy consumption, space efficiency and occupant behavior, etc. The data will be divided into three energy intensity categories by means of clustering analysis. In each category, the inherent relevancies between occupant behavior and the energy consumption of large-scale public building will be dug. Based on above, the visual monitoring network can be used for daily energy control. Meanwhile, optimization measures which can fulfill the requirements of occupants and leading them to environmental-friendly habits will be designed specifically. These measures are then be selected and integrated by dint of simulation and intelligent optimization algorithms. A win-win energy optimization proposal between environment benefit and economic benefit will consequently generated. Taking occupant behavior influences into consideration, the reliability of this energy optimization proposal can be proved by energy performance prediction with Artificial Neural Network. Finally, this study will provide an effective intelligent energy consumption optimization management mechanism of large-scale public buildings to satisfy the requirements of project owners, occupants and the environment.

我国大型公建能耗总量巨大，是制约建筑业绿色发展的重要因素。对大型公建的节能改造不仅依赖于对建筑物理性能上的优化，还要降低使用者行为带来的消极影响。本研究将基于建筑能耗表现特征将大型公建进行分类，分别识别各类中影响能耗表现的建筑物理性能和使用者行为关键因素。以关键因素为导向构建物联网与BIM耦合的可视监测网络，对能耗、空间利用率和使用者行为等进行测量与存储。运用聚类分析将测量数据按照能耗强度分为高、中、低三个层次，在每个层次内挖掘使用者行为与大型公建能耗表现的内在关联性。基于以上工作，一方面运用可视监测网络对建筑进行日常能耗管理；另一方面，有针对性地制定能够满足使用者需求并引导其减少对能耗消极影响的优化措施，结合仿真模拟与智能优化算法对优化措施进行筛选和整合，得到满足环境与经济效益双赢的优化方案，并运用人工神经网络预测该方案的可靠性，最终提出满足业主、用户、环境需求的智能能耗优化管理机制。

大型公共建筑由于具备体量巨大、寿命周期长、运维设备多、使用者数量大等特点，所消耗的能源在众多类型建筑中具有领先地位。在建筑运营期内进行合理的运营管理和优化能够节约大量能源。当前国内大型公建的运营管理体系并不完善，管理架构简单，且没有长期监测数据作为管理参考依据，因此节能效果甚微。考虑到建筑使用者在室内的活动行为对建筑能耗产生巨大影响，本项目通过文献分析、实地调研、现场实验和模拟研究等方法，对大型公建的建筑室内环境信息和使用者信息进行监测，并深入分析所测数据，获取大型公建的用能模式、关键影响因素、使用者行为特征，从而探寻建筑节能潜力较大的环节，有针对性地进行运维管理、节能设计和优化。本项目大量调查大型公建多年能耗数据，通过聚类分析得到了公共建筑的用能模式及其影响因素；基于DNAS理论挖掘了使用者行为对建筑能耗影响的关键因素及其影响力；基于以上内容，采用Wi-Fi定位技术和物联网技术在室内构建涵盖建筑物理信息和使用者行为信息的多维信息监测网络，并结合BIM建立了可视化数据监测系统，帮助管理者进行实时能耗优化管理和长期节能策略分析。最后，运用参数化建模技术结合人工智能技术提出了高效率、高精度的智能建筑能耗预测方法，以及能够充分考虑项目多方参与者目标重要度的建筑节能优化方法，帮助大型公建提出更加合理、更实用的建筑节能优化方案。本项目研究成果对大型公建的高效率、低成本、智能化运营管理和有效节能提供了理论参考和技术支持。