家庭电力载荷实时感知数据集

能源和可持续性问题提出了大量的问题，可以使用数据挖掘和机器学习的方法来解决，但由于缺乏公开的数据，这些问题的牵引力一直很缓慢。Reference Energy Disaggregation Data Set (REDD)是一个公共数据库，其中包含来自几个家庭的详细的用电量信息，旨在进一步研究能量分解（从一个聚合的电信号中确定组件设备贡献的任务）。数据包含来自真实家庭的整个房子以及房子中每个单独电路的功耗（由该电路上的主要设备类型标记）。REDD数据监控10个家庭，总共有119天的数据（综合了所有的家庭），268个独特的监视器，和超过1tb的原始数据。

Numerous challenges in the domains of energy and sustainability can be addressed using data mining and machine learning techniques, yet progress in research on these issues has been sluggish due to the scarcity of publicly available datasets. The Reference Energy Disaggregation Data Set (REDD) is a public database containing detailed electricity consumption information collected from multiple households, aimed at advancing research on energy disaggregation—the task of determining the contributions of individual connected devices from an aggregated electrical signal. The dataset includes power consumption data for entire homes and each individual circuit within real-world households, with each circuit labeled by the primary device type connected thereto. REDD monitors 10 households, encompassing a total of 119 days of aggregated data across all participating households, 268 unique meters, and over 1 terabyte of raw data.

搜集汇总

背景与挑战

背景概述

该数据集是Reference Energy Disaggregation Data Set (REDD)的一部分，专注于家庭电力载荷的实时感知，旨在支持能量分解研究，即从聚合电信号中识别单个电器设备的用电贡献。数据包含来自10个家庭的用电信息，总数据量2.8GB，覆盖119天，涉及268个监视器，适用于非侵入性负荷监测和电器识别等计算机应用领域。