Image Segmentation-based Event Detection for Non-Intrusive Load Monitoring using Gramian Angular Summation Field

Tito, Shafiqur Rahman; Rehman, Attique Ur; Kim, Youngyoon; Nieuwoudt, Pieter; Aslam, Saad; Soltic, Snjezana; Lie, Tek Tjing; Pandey, Neel; Ahmed, Mahmood

doi:10.1109/ieacon51066.2021.9654789

Cited by 4 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kyrkou et al [39] used the same transformation method, where they transformed 6.4min segments into GAF images and used pretrained VGG16 DL model as feature extractor and tried to detect ON/OFF state of only fridge appliance type from REDD and UK-DALE datasets. Similar thing was done by [40], where they were detecting ON/OFF state of 3 different appliance types on AMPds dataset on GAF transformation of one hour windows. Another GAF approach in combination with CNN was shown by [41], where they used their own dataset to classify 22 different appliances in a few minute windows.…”

Section: B Dimensionality Expansion Through Time Series To Image Tran...mentioning

confidence: 90%

Dimensionality Expansion of Load Monitoring Time Series and Transfer Learning for EMS

Bertalanič¹,

Jenko²,

Fortuna³

2022

Preprint

View full text Add to dashboard Cite

Electrical management systems (EMS) are playing a central role in enabling energy savings. They can be deployed within an everyday household where they monitor and manage appliances and help residents be more energy efficient and subsequently also more economical. One of they key functionalities of EMS is to automatically detect and identify appliances within a household through the process of load monitoring. In this paper, we propose a new transfer learning approach for building EMS (BEMS) and study the trade-offs in terms of numbers of samples and target classes in adapting a backbone model during the transfer process. We also perform a first time analysis of feature expansion through video-like transformation of time series data for device classification in non intrusive load monitoring (NILM) and propose a deep learning architecture enabling accurate appliance identification. We examine the relative performance of our method on 5 different representative low-frequency datasets and show that our method performs with an average F1 score of 0.88 on these datasets.

show abstract

Section: B Dimensionality Expansion Through Time Series To Image Tran...mentioning

confidence: 90%

Dimensionality Expansion of Load Monitoring Time Series and Transfer Learning for EMS

Bertalanič¹,

Jenko²,

Fortuna³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, voltage and current signatures were used to convert raw measurements into two-dimensional signatures [ 41 , 42 ]. Moreover, in [ 43 , 44 ] time series imaging approaches for univariate time series, i.e., when only a single feature is available, were investigated. However, it is not clear which two-dimensional representations have the best disaggregation performance, since, to the best of the authors knowledge, time-series-imaging techniques have not been compared with each other before.…”

Section: Introductionmentioning

confidence: 99%

2D Transformations of Energy Signals for Energy Disaggregation

Schirmer

Mporas

2022

Sensors

View full text Add to dashboard Cite

The aim of Non-Intrusive Load Monitoring is to estimate the energy consumption of individual electrical appliances by disaggregating the overall power consumption that has been sampled from a smart meter at a house or commercial/industrial building. Last decade’s developments in deep learning and the utilization of Convolutional Neural Networks have improved disaggregation accuracy significantly, especially when utilizing two-dimensional signal representations. However, converting time series’ to two-dimensional representations is still an open challenge, and it is not clear how it influences the performance of the energy disaggregation. Therefore, in this article, six different two-dimensional representation techniques are compared in terms of performance, runtime, influence on sampling frequency, and robustness towards Gaussian white noise. The evaluation results show an advantage of two-dimensional imaging techniques over univariate and multivariate features. In detail, the evaluation results show that: first, the active and reactive power-based signatures double Fourier based signatures, as well as outperforming most of the other approaches for low levels of noise. Second, while current and voltage signatures are outperformed at low levels of noise, they perform best under high noise conditions and show the smallest decrease in performance with increasing noise levels. Third, the effect of the sampling frequency on the energy disaggregation performance for time series imaging is most prominent up to 1.2 kHz, while, above 1.2 kHz, no significant improvements in terms of performance could be observed.

show abstract

Lightweight Gramian Angular Field classification for edge internet of energy applications

Alsalemi

Amira²,

Malekmohamadi³

et al. 2022

Cluster Comput

View full text Add to dashboard Cite

With adverse industrial effects on the global landscape, climate change is imploring the global economy to adopt sustainable solutions. The ongoing evolution of energy efficiency targets massive data collection and Artificial Intelligence (AI) for big data analytics. Besides, emerging on the Internet of Energy (IoE) paradigm, edge computing is playing a rising role in liberating private data from cloud centralization. In this direction, a creative visual approach to understanding energy data is introduced. Building upon micro-moments, which are timeseries of small contextual data points, the power of pictorial representations to encapsulate rich information in a small two-dimensional (2D) space is harnessed through a novel Gramian Angular Fields (GAF) classifier for energy micro-moments. Designed with edge computing efficiency in mind, current testing results on the ODROID-XU4 can classify up to 7 million GAF-converted datapoints with ~ 90% accuracy in less than 30 s, paving the path towards industrial adoption of edge IoE. Supplementary Information The online version contains supplementary material available at 10.1007/s10586-022-03704-1.

show abstract

Image Segmentation-based Event Detection for Non-Intrusive Load Monitoring using Gramian Angular Summation Field

Cited by 4 publications

References 11 publications

Dimensionality Expansion of Load Monitoring Time Series and Transfer Learning for EMS

Dimensionality Expansion of Load Monitoring Time Series and Transfer Learning for EMS

2D Transformations of Energy Signals for Energy Disaggregation

Lightweight Gramian Angular Field classification for edge internet of energy applications

Contact Info

Product

Resources

About