「英中对照」研究提出楼宇自动化系统数据驱动元数据标注统一架构
A Unified Architecture for Data-Driven Metadata Tagging of Building Automation Systems
研究提出融合监督分类与无监督聚类的统一架构，可自动完成楼宇自动化系统数据点的元数据标注。该架构融合两类机器学习方法的确定性与概率输出，以楼宇自动化系统可获取的时间序列数据、原始点位名称为输入，还扩展了适配实时部署的数据处理模块。在商业零售建筑、国家可再生能源实验室园区办公楼两类真实建筑数据集测试中，研究主张标签正确应用率分别为85%-90%、70%-75%。
Abstract:This article presents a Unified Architecture for automated point tagging of Building Automation System data, based on a combination of data-driven approaches. Advanced energy analytics applications-including fault detection and diagnostics and supervisory control-have emerged as a significant opportunity for improving the performance of our built environment. Effective application of these analytics depends on harnessing structured data from the various building control and monitoring systems, but typical Building Automation System implementations do not employ any standardized metadata schema. While standards such as Project Haystack and Brick Schema have been developed to address this issue, the process of structuring the data, i.e., tagging the points to apply a standard metadata schema, has, to date, been a manual process. This process is typically costly, labor-intensive, and error-prone. In this work we address this gap by proposing a UA that automates the process of point tagging by leveraging the data accessible through connection to the BAS, including time series data and the raw point names. The UA intertwines supervised classification and unsupervised clustering techniques from machine learning and leverages both their deterministic and probabilistic outputs to inform the point tagging process. Furthermore, we extend the UA to embed additional input and output data-processing modules that are designed to address the challenges associated with the real-time deployment of this automation solution. We test the UA on two datasets for real-life buildings: 1. commercial retail buildings and 2. office buildings from the National Renewable Energy Laboratory campus. The proposed methodology correctly applied 85-90 percent and 70-75 percent of the tags in each of these test scenarios, respectively. 摘要：本文提出一种基于多种数据驱动方法组合的统一架构（UA），用于楼宇自动化系统数据的自动点位标注。高级能源分析应用——包括故障检测与诊断和监督控制——已成为改善我们建筑环境性能的重要机遇。这些分析的有效应用依赖于利用来自各类楼宇控制与监测系统的结构化数据，但典型的楼宇自动化系统实施并未采用任何标准化元数据模式。尽管已制定 Project Haystack 和 Brick Schema 等标准来解决这一问题，但数据结构化过程，即为点位添加标签以应用标准元数据模式，迄今为止仍是人工过程。该过程通常成本高昂、劳动密集且容易出错。在本研究中，我们提出一种 UA 来填补这一空白；该 UA 利用通过连接 BAS 可获取的数据（包括时间序列数据和原始点位名称），实现点位标注过程自动化。该 UA 融合机器学习中的监督分类和无监督聚类技术，并利用二者的确定性输出和概率输出为点位标注过程提供依据。此外，我们扩展了 UA，以嵌入额外的输入和输出数据处理模块，这些模块旨在应对与该自动化解决方案实时部署相关的挑战。我们在两个真实建筑数据集上测试了 UA：1. 商业零售建筑；2. 来自国家可再生能源实验室园区的办公楼。所提出的方法在这些测试场景中分别正确应用了 85-90 percent 和 70-75 percent 的标签。
| Comments:评论： | 19 pages, 9 figures, accepted for publication in Automation in Construction19 pages, 9 figures，已被 Automation in Construction 接收发表 |
| Subjects:主题： | Computers and Society (cs.CY)计算机与社会 (cs.CY) |
| Cite as:引用为： | arXiv:2003.07690 [cs.CY] |
| (or arXiv:2003.07690v2 [cs.CY] for this version) (或 arXiv:2003.07690v2 [cs.CY] 对应此版本) | |
| https://doi.org/10.48550/arXiv.2003.07690 arXiv-issued DOI via DataCite 通过 DataCite 获取 arXiv 发布的 DOI |
|
| Related DOI: 相关 DOI： | https://doi.org/10.1016/j.autcon.2020.103411
DOI(s) linking to related resources 链接至相关资源的 DOI(s) |
Submission history提交历史
From: Sakshi Mishra [view email]
[v1]
Thu, 27 Feb 2020 00:35:01 UTC (1,054 KB)
[v2]
Fri, 11 Sep 2020 18:27:29 UTC (1,148 KB)
来自：Sakshi Mishra [查看邮箱]
[v1]
Thu, 27 Feb 2020 00:35:01 UTC (1,054 KB)
[v2]
Fri, 11 Sep 2020 18:27:29 UTC (1,148 KB)
来源：arXiv 建筑语义标准论文 · arxiv.org