Plug Load Dataset for Office Spaces

This repository contains the office plug load dataset that was collected in the paper titled "Near-Real-Time Plug Load Identification using Low-frequency Power Data in Office Spaces: Experiments and Applications". This paper was submitted on 27th April 2020 to the Journal of Applied Energy and accepted on 9th June 2020.

Please include the following citation if you are interested in using this dataset:

Tekler ZD, Low R, Zhou Y, Yuen C, Blessing L, Spanos C. Near-real-time plug load identification using low-frequency power data in office spaces: Experiments and applications. Applied Energy 2020;275:115391. https://doi.org/10.1016/j.apenergy.2020.115391

The dataset was the result of a three-week data collection effort that was conducted in a typical office environment between February 2020 to March 2020. The dataset contains the power consumption data of several plug loads that are commonly found on the occupants' desks, including 31 laptops, 9 desktops, 35 monitors, 13 fans, and 11 task lamps. A total of 36 occupants participated in this study consisting of a mixture of researchers and administrative staff.

Each entry in the dataset contains four fields, including 1) the timestamp information, 2) the instantaneous power value of the connected plug load recorded up to two decimal places, 3) a unique ID indicating the smart power plug that recorded the information, and 4) the label of the corresponding plug load type that was provided post-data collection. The data was also collected with a sampling frequency of 1/60 Hz (equivalent to 1 sample every minute).

This dataset has also been uploaded at the following sites:
GitHub: https://github.com/zeynepduygutekler/plug-load-dataset

本仓库收录了来自论文《面向办公场景的低频电力数据实时插电负载识别：实验与应用（Near-Real-Time Plug Load Identification using Low-frequency Power Data in Office Spaces: Experiments and Applications）》的办公室插电负载（plug load）数据集。该论文于2020年4月27日提交至《应用能源（Journal of Applied Energy）》期刊，并于2020年6月9日被正式录用。

若您有意使用本数据集，请引用如下文献：
Tekler ZD, Low R, Zhou Y, Yuen C, Blessing L, Spanos C. 面向办公场景的低频电力数据实时插电负载识别：实验与应用. 应用能源（Applied Energy）2020;275:115391. https://doi.org/10.1016/j.apenergy.2020.115391

本数据集的采集工作于2020年2月至2020年3月间，在典型办公环境中开展，历时三周。数据集涵盖了办公人员桌面常见的多款插电负载的功耗数据，包含31台笔记本电脑、9台台式机、35台显示器、13台风扇以及11台任务台灯。本次研究共有36名办公人员参与，涵盖研究人员与行政职员两类群体。

数据集中的每条记录均包含四个字段：1）时间戳（timestamp）信息；2）所记录的对应插电负载的瞬时功率值，精确至小数点后两位；3）用于标识采集该数据的智能电源插头（smart power plug）的唯一ID；4）数据采集完成后标注的对应插电负载类型标签。本次数据采集的采样频率为1/60 Hz（即每分钟采集1个样本）。

本数据集同时上传至以下平台：
GitHub：https://github.com/zeynepduygutekler/plug-load-dataset