miscellaneous-10

Non-Intrusive Load Monitoring

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in non-intrusive-load-monitoring-30

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find non-intrusive-load-monitoring-30 models and implementations

Datasets

SynD

Subtasks

No subtasks available.

Most implemented papers

Improving Non-Intrusive Load Disaggregation through an Attention-Based Deep Neural Network

V. Piccialli, A. M. Sudoso•Thu Feb 04 2021

A deep neural network that combines a regression subnetwork with a classification subnetwork for solving the NILM problem is proposed and it is shown that modeling attention translates into the network’s ability to correctly detect the turning on or off an appliance and to locate signal sections with high power consumption.

86

Content

0

Paper Graph

Transfer Learning for Non-Intrusive Load Monitoring

M. D'Incecco, S. Squartini, Mingjun Zhong•Fri Feb 22 2019

Two transfer learning schemes are proposed, i.e., appliance transfer learning (ATL) and cross-domain transferLearning (CTL), and the conclusion is that the seq2point learning is transferable.

264 0

Paper Graph

Wavenilm: A Causal Neural Network for Power Disaggregation from the Complex Power Signal

I. Bajić, S. Makonin, Alon Harell•Thu Feb 21 2019

This work presents a causal 1-D convolutional neural network inspired by WaveNet for NILM on low-frequency data and demonstrates that using all four components available in a popular NilM dataset achieves faster convergence and higher performance than state-of-the-art results for the same dataset.

103 0

Paper Graph

On Metrics to Assess the Transferability of Machine Learning Models in Non-Intrusive Load Monitoring

S. Makonin, Christoph Klemenjak, W. Elmenreich, A. Faustine•Wed Dec 11 2019

This paper discusses several metrics to assess the generalisation ability of NILM algorithms and demonstrates how these metrics can be utilised to evaluate NilM algorithms by means of two case studies.

18 0

Paper Graph

Towards Comparability in Non-Intrusive Load Monitoring: On Data and Performance Evaluation

S. Makonin, Christoph Klemenjak, W. Elmenreich•Sun Jan 19 2020

Comparability in NILM is drawn attention with a focus on highlighting the considerable differences amongst common energy datasets used to test the performance of algorithms, and a close view on evaluation processes is given.

55 0

Paper Graph

On time series representations for multi-label NILM

Christoforos Nalmpantis, D. Vrakas•Fri May 01 2020

This research investigates in-depth multi-label NILM systems and suggests a novel framework which enables a cost-effective solution and can be deployed on an embedded device, and thus, privacy can be preserved.

35 0

Paper Graph

Energy Disaggregation using Variational Autoencoders

M. Carbonneau, A. Langevin, M. Cheriet, G. Gagnon•Sun Mar 21 2021

An energy disaggregation approach based on the variational autoencoders framework that accurately generates more complex load profiles, thus improving the power signal reconstruction of multi-state appliances and improves the generalization capabilities of the model across different houses is proposed.

67 0

Paper Graph

ELECTRIcity: An Efficient Transformer for Non-Intrusive Load Monitoring

N. Doulamis, A. Doulamis, S. Sykiotis, Maria Kaselimi•Thu Mar 31 2022

The proposed approach, called ELECTRIcity, utilizes transformer layers to accurately estimate the power signal of domestic appliances by relying entirely on attention mechanisms to extract global dependencies between the aggregate and the domestic appliance signals.

55 0

Paper Graph

COLD: Concurrent Loads Disaggregator for Non-Intrusive Load Monitoring

E. Gryazina, Ilia Kamyshev, Dmitrii Kriukov•Thu Jun 03 2021

COLD, a transformer-based model specifically designed to address the challenges of disaggregating high-frequency data with multiple simultaneously working devices is proposed, achieving 95% load identification accuracy and 82% disaggregation performance on the test data.

12 0

Paper Graph

Adding a benchmark result helps the community track progress.