Plug Load Dataset for Office Spaces

Published: 30-06-2020| Version 3 | DOI: 10.17632/dnx6bc59rj.3
Zeynep Tekler


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. 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:

Download All