AMPds: A public dataset for load disaggregation and eco-feedback research

Makonin, Stephen; Popowich, Fred; Bartram, Lyn; Gill, Bob; Bajić, Ivan V.

doi:10.1109/epec.2013.6802949

Cited by 263 publications

(153 citation statements)

References 12 publications

Supporting

Mentioning

143

Contrasting

Unclassified

Order By: Relevance

“…In this work, we adopt current, measured in amperes (A), as the stream to measure and predict. This follows Makonin et al's [7] analysis showing that current is the most reliable measure for data analysis because it fluctuates to a much lesser degree than other measures such as real power (which is frequently used in other research works in this field).…”

Section: Introductionmentioning

confidence: 99%

“…The Almanac of Minutely Power dataset ("AMPds") was introduced by Makonin et al [7] in 2013, and a revision (version 2) released sometime afterwards. This dataset is a record of one house containing both total and submetered power data from 25 submeters at one minute intervals for two years (from 1 April 2012 until 31 March 2014).…”

Section: Ampds V2 Datasetmentioning

confidence: 99%

“…If a good regression-based solution to the problem is found, the approach can be adapted to the classification setting by discretising current using the approach described by Makonin et al [7]. Another reason for adopting a regression-based approach to the problem is that the finite state machine abstraction is not true for all appliances: some devices clearly vary continuously in their energy usage, and even for appliances that do appear to operate in discrete states, the number of and boundaries between each state are not always clear (as Figure 1 attests).…”

Section: Ampds V2 Datasetmentioning

confidence: 99%

“…Then, in Section 4, a random search-based algorithm is presented to select a regression model that optimizes TW-RMSE. Section 5 evaluates the presented approaches on the AMPds dataset [7] and makes a complete comparison between RMSE and TW-RMSE. Lastly, Section 6 concludes the paper.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Towards a New Evolutionary Subsampling Technique for Heuristic Optimisation of Load Disaggregators

Mayo

Omranian

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper we present some preliminary work towards the development of a new evolutionary subsampling technique for solving the non-intrusive load monitoring (NILM) problem. The NILM problem concerns using predictive algorithms to analyse whole-house energy usage measurements, so that individual appliance energy usages can be disaggregated. The motivation is to educate home owners about their energy usage. However, by their very nature, the datasets used in this research are are massively imbalanced in their target value distributions. Consequently standard machine learning techniques, which often rely on optimising for root mean squared error (RMSE), typically fail. We therefore propose the target-weighted RMSE (TW-RMSE) metric as an alternative fitness function for optimising load disaggregators, and show in a simple initial study in which random search is utilised that TW-RMSE is a metric that can be optimised, and therefore has the potential to be included in a larger evolutionary subsampling-based solution to this problem.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Ampds V2 Datasetmentioning

confidence: 99%

Section: Ampds V2 Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards a New Evolutionary Subsampling Technique for Heuristic Optimisation of Load Disaggregators

Mayo

Omranian

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…To date there are, to the best of our knowledge, five public datasets specifically created for NILM research, namely, the Reference Energy Disaggregation Dataset (REDD) [10], the Building-Level fUlly-labeled dataset for Electricity Disaggregation (BLUED) [11], the Almanac of Minutely Power dataset (AMPds) [12] the Pecan Street Research Institute (PSRI) [13] dataset and lastly the "UK-Dale" [14]. All these datasets provide whole house consumption data and ground-truth information of individual appliances either by labeling the power changes in the whole house consumption signal (the case of BLUED) or by providing the aggregated consumption of individual appliances or circuits in the house (REDD, AMPds, PSRI and "UK-Dale").…”

Section: A Related Workmentioning

confidence: 99%

SustData: A Public Dataset for ICT4S Electric Energy Research

Pereira¹,

Quintal²,

Gonçalves³

et al. 2014

Proceedings of the 2014 Conference ICT for Sustainability

View full text Add to dashboard Cite

Abstract-Energy and environmental sustainability can benefit a lot from advances in data mining and machine learning techniques. However, those advances rely on the availability of relevant datasets required to develop, improve and validate new techniques. Only recently the first datasets were made publicly available for the energy and sustainability research community. In this paper we present a freely available dataset containing power usage and related information from 50 homes. Here we describe our dataset, the hardware and software setups used when collecting the data and how others can access it. We then discuss potential uses of this data in the future of energy ecofeedback and demand side management research.

show abstract

Recent trends of smart nonintrusive load monitoring in buildings: A review, open challenges, and future directions

Himeur

Alsalemi

Bensaali

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Smart nonintrusive load monitoring (NILM) represents a cost-efficient technology for observing power usage in buildings. It tackles several challenges in transitioning into a more effective, sustainable, and digital energy efficiency environment. This paper presents a comprehensive review of recent trends in the NILM field, in which we propose a multiperspective classification of existing smart NILM techniques. More attention is devoted to describing the contributions of deep learning, feature extraction, computing platforms, and application scenarios for NILM development. Accordingly, NILM technical aspects are first Int J Intell Syst. 2022;37:7124-7179. wileyonlinelibrary.com/journal/int investigated, including data collection devices and public data sets. Next, event-based and non-eventbased NILM algorithms are overviewed. Furthermore, potential limitations of existing solutions are identified, highlighting their technical challenges, especially those related to security and privacy preservation, data scarcity, results reproduction, and implementation and business difficulties. Lastly, future directions are explored to overcome the identified limitations.

show abstract

AMPds: A public dataset for load disaggregation and eco-feedback research

Cited by 263 publications

References 12 publications

Towards a New Evolutionary Subsampling Technique for Heuristic Optimisation of Load Disaggregators

Towards a New Evolutionary Subsampling Technique for Heuristic Optimisation of Load Disaggregators

SustData: A Public Dataset for ICT4S Electric Energy Research

Recent trends of smart nonintrusive load monitoring in buildings: A review, open challenges, and future directions

Contact Info

Product

Resources

About