A Comprehensive Feature Study for Appliance Recognition on High Frequency Energy Data

Kahl, Matthias; Haq, Anwar Ul; Kriechbaumer, Thomas; Jacobsen, Hans‐Arno

doi:10.1145/3077839.3077845

Cited by 36 publications

(23 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers make use of public datasets to study the characteristics of appliances and to build models representing load profiles and per-appliance usage. This can be beneficial for energy reduction 3 , 4 , pattern recognition 5–8 , energy demand forecasting 9 , and similar fields of study.…”

Section: Background and Summarymentioning

confidence: 99%

“…The amount of information contained in electricity signals increases steadily with sampling rates ranging up to 1MHz. Higher sampling rates can capture subtle changes (high frequency ripples), which are useful for appliance identification 5 , 23–25 . Capturing the voltage and current waveforms allows energy disaggregation algorithms such as BOLT 6 to extract patterns directly from the raw measurement data.…”

Section: Background and Summarymentioning

confidence: 99%

See 1 more Smart Citation

BLOND, a building-level office environment dataset of typical electrical appliances

Kriechbaumer

Jacobsen

2018

Sci Data

Self Cite

View full text Add to dashboard Cite

Energy metering has gained popularity as conventional meters are replaced by electronic smart meters that promise energy savings and higher comfort levels for occupants. Achieving these goals requires a deeper understanding of consumption patterns to reduce the energy footprint: load profile forecasting, power disaggregation, appliance identification, startup event detection, etc. Publicly available datasets are used to test, verify, and benchmark possible solutions to these problems. For this purpose, we present the BLOND dataset: continuous energy measurements of a typical office environment at high sampling rates with common appliances and load profiles. We provide voltage and current readings for aggregated circuits and matching fully-labeled ground truth data (individual appliance measurements). The dataset contains 53 appliances (16 classes) in a 3-phase power grid. BLOND-50 contains 213 days of measurements sampled at 50kSps (aggregate) and 6.4kSps (individual appliances). BLOND-250 consists of the same setup: 50 days, 250kSps (aggregate), 50kSps (individual appliances). These are the longest continuous measurements at such high sampling rates and fully-labeled ground truth we are aware of.

show abstract

Section: Background and Summarymentioning

confidence: 99%

Section: Background and Summarymentioning

confidence: 99%

BLOND, a building-level office environment dataset of typical electrical appliances

Kriechbaumer

Jacobsen

2018

Sci Data

Self Cite

View full text Add to dashboard Cite

show abstract

“…These features or parameters are utilized to detect appliance switching from the aggregate load using machine learning algorithms. Khal et al [59] have identified 36 such features, including wavelet analysis, voltage-current (V-I) trajectory, inrush current ratio, waveform approximation, and log attack time, along with other spectral and temporal features. When considering load disaggregation, it is always better to incorporate more parameters, as certain parameters work better for particular load types [33,60].…”

Section: Appliance Identification Parametersmentioning

confidence: 99%

Prospects of Appliance-Level Load Monitoring in Off-the-Shelf Energy Monitors: A Technical Review

Haq

Jacobsen

2018

Energies

Self Cite

View full text Add to dashboard Cite

Abstract:The smart grid initiative has encouraged utility companies worldwide to roll-out new and smarter versions of energy meters. Before an extensive roll-out, which is both labor-intensive and incurs high capital costs, consumers need to be incentivised to reap the long-term benefits of such smart meters. Off-the-shelf energy monitors (e-monitors) can provide consumers with an insight into such potential benefits. As e-monitors are owned by the consumer, the consumer has greater control over the data, which significantly reduces the privacy and data confidentiality concerns. Because only limited online technical information is available about e-monitors, we evaluate several existing e-monitors using an online technical survey directly from the vendors. Besides automated e-monitoring, the use of different off-the-shelf e-monitors can also help to demonstrate state-of-the-art techniques such as non-intrusive load monitoring (NILM), data analytics, and the predictive maintenance of appliances. Our survey indicates a trend towards the incorporation of such state-of-the-art capabilities, particularly the appliance-level e-monitoring and load disaggregation. We have also discussed some essential requirements to implement load disaggregation in the next generation e-monitors. In future, these intelligent e-monitoring techniques will encourage effective consumer participation in the demand-side management (DSM) programs.

show abstract

“…When an event is detected, load signatures can be extracted by analyzing the difference of electrical signal before and after the event [6]. In the machine learning context, these signatures are called features [7]. Then a classifier is used to identify which appliance caused the event.…”

Section: Introductionmentioning

confidence: 99%

A Load Identification Method Based on Active Deep Learning and Discrete Wavelet Transform

et al. 2020

View full text Add to dashboard Cite

Non-Intrusive Load Monitoring (NILM) makes it possible for users and energy providers to track the fine-grained energy consumption information of residential and commercial buildings. The load identification methods in NILM usually require labeling many samples for training and evaluation, which is always expensive and time-consuming. In order to reduce the labeling cost, this paper proposed a load identification method based on Active Deep Learning (ADL). In this method, Discrete Wavelet Transform (DWT) was applied to extract high-dimensional appliance features from original current signals. Then a pool-based or stream-based active deep learning model was built to learn the features and select high-value samples that worthy of labeling. A mixed dataset based on three public datasets was formed to evaluate the proposed method and three sampling approaches of active learning. The results showed that the proposed method could significantly reduce labeling cost on large datasets, and the number of samples required is 33% lower than the state-of-the-art method when the F1 score is equal. Compared with pool-based sampling approaches, the stream-based approach's benefits are that the classifier improved and the query frequency decreased with continuous input of samples. INDEX TERMS NILM, load identification, active deep learning, semi-supervised learning, CNN, pool-based sampling, stream-based sampling.

show abstract

A Comprehensive Feature Study for Appliance Recognition on High Frequency Energy Data

Cited by 36 publications

References 26 publications

BLOND, a building-level office environment dataset of typical electrical appliances

BLOND, a building-level office environment dataset of typical electrical appliances

Prospects of Appliance-Level Load Monitoring in Off-the-Shelf Energy Monitors: A Technical Review

A Load Identification Method Based on Active Deep Learning and Discrete Wavelet Transform

Contact Info

Product

Resources

About