绿色建筑大数据关联分析与节能优化研究

The Internet of Things provides us a powerful tool for the green running control of buildings. To realize the green objectives of comfort, safety and low energy consumption, it is a realistic but difficult problem in the research domain of intelligent buildings to carry out association analysis of the buildings’ big data collected through comprehensive perception and further to provide a novel effective tool for the global optimization of building energy. To solve such problems, this project will carry out association analysis of the buildings’ big data, and based on the association analysis results, a total energy control scheme will be presented to realize the energy-saving and comfortable operation of green buildings. The detailed contents of this project are as follows. Firstly, the service model for the buildings’ big data will be constructed, and the association analysis will be studied. Then, the data driven models for the building energy consumption and the comfort will be constructed using the results from association analysis. Secondly, under the energy consumption constraints, the dynamic allocation problem on each room’s energy index will be studied. The demands driven dynamic energy index allocation model will be built and solved. Thirdly, under the energy index constraint of each room, the distributed multi-objective optimization problem of the operation control of the building’s electrical equipments will be explored. The optimization model on the operation of the electrical equipments in each room will be derived to accomplish energy-saving and comfortable indoor environment, and mixing-variable multi-objective optimization methods will be presented to effectively obtain the solution of the constructed model. At last, the proposed buildings’ big data association analysis methods and energy-saving optimization strategies will be applied to Internet of Things for the building electrical equipments. The results obtained from this project can provide novel theoretical and technical supports for the energy-saving and comfortable operation of the green buildings.

物联网为建筑物绿色运行控制提供了有效手段，为达到舒适、安全、低能耗的绿色目标，对全面感知获得的大数据进行关联分析，进而给出楼宇全局优化节能的控制策略，是建筑智能化控制领域目前面对的具有现实意义的难题。针对该问题，本项目拟展开建筑大数据关联分析研究，并在此基础上通过能耗总量控制实现建筑物节能舒适绿色运行。主要研究内容包括：建立建筑大数据服务模型，进行建筑大数据的关联分析，在关联分析基础上构建出能耗及舒适性数据驱动预测模型；研究建筑能耗总额限制下各房间能耗指标的动态分配问题，构建面向需求的能耗指标分配模型并求解；研究能耗指标约束下建筑物用电设备运行分布式多目标优化问题，构建建筑物各房间用电设备节能舒适运行优化模型，提出有效的混合变量多目标优化方法实现该模型的求解；最后，在建筑设备物联网系统上实现绿色建筑大数据分析与节能优化策略的验证及应用。本项目将为绿色建筑节能舒适运行提供新的理论与技术支撑。

建筑物联网为全面感知建筑运行信息提供了技术支撑，但如何处理全面感知获得的大数据，如何得到楼宇全局优化节能的控制策略，是急需解决的问题。该课题针对建筑大数据的深度挖掘利用进行了深入研究，取得了系列成果，为建筑节能优化运行提供了新的理论与技术支撑。本项目的主要成果包括：1）搭建了建筑大数据云平台，提出了一套物联网环境下绿色建筑大数据关联分析方法； 2）给出了基于建筑大数据和深度学习的建筑能耗精准预测方法；3）给出了数据驱动单输入规则模块模糊神经系统设计方法，实现了在舒适性预测中的应用；4）提出了基于数据驱动模型面向需求的能耗指标动态分配方法；5）给出了建筑物联网中多射频节点多信道数据传输优化控制方案；6）提出了能耗指标约束下建筑物用电设备运行分布式多目标优化方案，推动了所提出的绿色建筑大数据分析及节能优化管控方法在建筑物联网系统中的应用。本项目的研究申请国家发明专利30余项，其中授权密切相关发明专利9项；主持编写国家标准 1 项，参编国家、行业与地方标准 3 项；申请并获得软件著作权4 项；出版专著1部，发表高质量论文31篇，其中 SCI、EI检索25篇，ESI高被引论文2篇；培养博士生4名，其中毕业1名，在读3名，培养硕士研究生20余名；获山东省高等学校科学技术一等奖、山东省机械工业科技进步奖一等奖各1项。项目研究成果有助于推动智能建筑及相关领域科技进步，有助于节能降耗、保护资源与环境。