Optimal sensor placement strategy for office buildings using clustering algorithms
Optimal sensor placement strategy for office buildings using clustering algorithms
复制标题
DOI：
10.1016/j.enbuild.2017.10.074
复制
发表时间：
2018-01-01
影响因子：
6.7
通讯作者：
Manthapuri, Sumanth
中科院分区：
文献类型：
--
作者：
Yoganathan, Duwaraka;Kondepudi, Sekhar;Manthapuri, Sumanth
嵌入在建筑环境中的传感器网络为智能建筑能源管理提供了关键信息。来自这些传感器的数据可以优化能源效率和室内环境质量，而不会影响居住者的舒适度。因此，传感器有助于以降低的运营成本实现建筑系统的高效运营。理想情况下，为了实现这些目标，应该测量和验证建筑物中所有可能的测量点。不过，这必然会耗费大量的成本和时间。可替代地，期望一种用于识别可以提供室内环境的整体画面的最佳测量点的方法。本文提出了一种新的数据驱动的方法的基础上，在办公楼的现场测量，以获得最佳的（数量和位置）的测量点。利用聚类算法、信息损失法和帕累托原理推导了传感器的最优布置策略。这项研究的结果可能对研究人员和从业人员具有重要意义。(C)2017爱思唯尔B.V.保留所有权利。
Sensor networks embedded in the built environment provide critical information for intelligent building energy management. Data from these sensors enable optimizing energy efficiency and indoor environmental quality without compromising occupant comfort. Thus sensors help achieve efficient operation of building systems at reduced operating costs. Ideally, towards these goals all possible measurement points in buildings should be measured and verified. However, this would inevitably incur tremendous cost and time. Alternatively, an approach to identify the optimal measurement points that can provide a holistic picture of the indoor environment is desirable. This paper proposes a novel data driven approach based on field measurements in an office building to derive the optimal (number and locations of) measuring points. Clustering algorithms, information loss approach and Pareto principle were used to derive the optimal sensor placement strategy. The findings of this study can have important implications for researchers and practitioners. (C) 2017 Elsevier B.V. All rights reserved.