Towards reproducible state-of-the-art energy disaggregation

Batra, Nipun; Kukunuri, Rithwik; Pandey, Ayush; Malakar, Raktim; Kumar, Rajat; Krystalakos, Odysseas; Zhong, Mingjun; Meira, Paulo C. M.; Parson, Oliver

doi:10.1145/3360322.3360844

Cited by 96 publications

(76 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…• We see a great potential in defining a standard evaluation protocol that defines training and testing folds for cross-validation of models per dataset. Of course it should respect particularities of the NILM setting such as the evaluation scenarios and it would ideally be in a machine readable form such as proposed for the ExperimentAPI of NILMTK [23]. Publication Figure 4.…”

Section: Performance Comparisonmentioning

confidence: 99%

Review on Deep Neural Networks Applied to Low-Frequency NILM

Huber¹,

Calatroni²,

Rumsch³

et al. 2021

Preprint

View full text Add to dashboard Cite

This paper reviews non-intrusive load monitoring (NILM) approaches that employ deep neural networks to disaggregate appliances from low frequency data, i.e. data with sampling rates lower than the AC base frequency. We first review the many degrees of freedom of these approaches, what has already been done in literature, and compile the main characteristics of the reviewed publications in an extensive overview table. The second part of the paper discusses selected aspects of the literature and corresponding research gaps. In particular, we do a performance comparison with respect to reported MAE and F$_1$-scores and observe different recurring elements in the best performing approaches, namely data sampling intervals below 10\,s, a large field of view, the usage of GAN losses, multi-task learning, and post-processing. Subsequently, multiple input features, multi-task learning and related research gaps are discussed, the need for comparative studies is highlighted, and finally, missing elements for a successful deployment of NILM approaches based on deep neural networks are pointed out. We conclude the review with an outlook on possible future scenarios.

show abstract

Section: Performance Comparisonmentioning

confidence: 99%

Review on Deep Neural Networks Applied to Low-Frequency NILM

Huber¹,

Calatroni²,

Rumsch³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…As intended by the authors of (Makonin and Popowich 2015), we consider all submeter signals recorded during the measurement campaign to compute the NAR. These datasets were selected because of their compatibility to NILMTK, a toolkit that enables reproducible NILM experiments (Batra et al 2014b;Batra et al 2019). We excluded from consideration the dataset BLUED (Anderson et al 2012) due to the lack of sub-metered power data, Tracebase (Reinhardt et al 2012) and GREEND (Monacchi et al 2014) due to the lack of household aggregate power data.…”

Section: Assessing Signal Noise Levelsmentioning

confidence: 99%

“…• The Combinatorial Optimization (CO) algorithm, introduced in Hart (1992), has been used repeatedly in literature to serve as baseline (Batra et al 2019). The CO algorithm estimates the power demand of appliances and their operational mode.…”

Section: Evaluation Setupmentioning

confidence: 99%

See 1 more Smart Citation

Investigating the performance gap between testing on real and denoised aggregates in non-intrusive load monitoring

2021

View full text Add to dashboard Cite

Prudent and meaningful performance evaluation of algorithms is essential for the progression of any research field. In the field of Non-Intrusive Load Monitoring (NILM), performance evaluation can be conducted on real-world aggregate signals, provided by smart energy meters or artificial superpositions of individual load signals (i.e., denoised aggregates). It has long been suspected that testing on these denoised aggregates provides better evaluation results mainly due to the fact that the signal is less complex. Complexity in real-world aggregate signals increases with the number of unknown/untracked loads. Although this is a known performance reporting problem, an investigation into the actual performance gap between real and denoised testing is still pending. In this paper, we examine the performance gap between testing on real-world and denoised aggregates with the aim of bringing clarity into this matter. Starting with an assessment of noise levels in datasets, we find significant differences in test cases. We give broad insights into our evaluation setup comprising three load disaggregation algorithms, two of them relying on neural network architectures. The results presented in this paper, based on studies covering three scenarios with ascending noise levels, show a strong tendency towards load disaggregation algorithms providing significantly better performance on denoised aggregate signals. A closer look at the outcome of our studies reveals that all appliance types could be subject to this phenomenon. We conclude the paper by discussing aspects that could be causing these considerable gaps between real and denoised testing in NILM.

show abstract

“…Introduced in 11 , it provides functionalities to perform dataset analysis and aims to enable benchmarking of load disaggregation algorithms. Recent contributions, presented in 12 , extend the toolkit by introducing www.nature.com/scientificdata www.nature.com/scientificdata/ new APIs for disaggregation and experiments. To lower the entry barrier for NILMTK users, we provide a NILMTK-compatible version of our synthetic dataset.…”

Section: Synd_csvzipmentioning

confidence: 99%

A synthetic energy dataset for non-intrusive load monitoring in households

et al. 2020

View full text Add to dashboard Cite

Research on smart grid technologies is expected to result in effective climate change mitigation. Non-Intrusive Load Monitoring (NILM) is seen as a key technique for enabling innovative smart-grid services. By breaking down the energy consumption of households and industrial facilities into its components, NILM techniques provide information on present appliances and can be applied to perform diagnostics. As with related Machine Learning problems, research and development requires a sufficient amount of data to train and validate new approaches. As a viable alternative to collecting datasets in buildings during expensive and time-consuming measurement campaigns, the idea of generating synthetic datasets for NILM gain momentum recently. With SynD, we present a synthetic energy dataset with focus on residential buildings. We release 180 days of synthetic power data on aggregate level (i.e. mains) and individual appliances. SynD is the result of a custom simulation process that relies on power traces of real household appliances. In addition, we present several case studies that demonstrate similarity of our dataset and four real-world energy datasets. Background & Summary Load monitoring is vital for effective and accurate energy monitoring in buildings. Detailed insights can empower further research, help streamlining processes, and improve a building's energy efficiency 1. Introduced in 2 , Non-Intrusive Load Monitoring (NILM) techniques serve to break down a building's aggregate energy consumption to identify active appliances and also to provide diagnostic information. Extensive reviews can be obtained from 3 and 4. NILM can be considered as Machine Learning problem. As such, it requires datasets to train models, to conduct performance evaluation, to evaluate the benefit in real scenarios, and also to perform benchmarking on a common basis. In case of NILM, ground-truth data on aggregate and appliance-level energy consumption are crucial 4. Traditionally, NILM scholarship relies on energy consumption datasets. Such datasets usually contain information on energy consumption on aggregate level (monitored at the mains) and individual loads, which is provided by plug-level meters. Energy consumption datasets are the outcome of measurement campaigns in buildings or industrial facilities, which require expensive measurement equipment, bring bureaucratic burdens, and are time-consuming activities 5. As a viable alternative, the idea of generating synthetic data gain momentum recently. The main motivation behind generating synthetic datasets is to reduce costs for measurement campaigns and save valuable work hours. Instead, custom simulators provide energy consumption datasets on-demand and in contrast to real datasets, without limitations on measurement periods. Furthermore, real datasets suffer from missing readings (gaps), misaligned timestamps, and corrupted data as a result of sensor miscalculation or malfunction 6,7. Synthetic data does not show such issues. With SynD, we present a synthetic energy consumption da...

show abstract

Towards reproducible state-of-the-art energy disaggregation

Cited by 96 publications

References 13 publications

Review on Deep Neural Networks Applied to Low-Frequency NILM

Review on Deep Neural Networks Applied to Low-Frequency NILM

Investigating the performance gap between testing on real and denoised aggregates in non-intrusive load monitoring

A synthetic energy dataset for non-intrusive load monitoring in households

Contact Info

Product

Resources

About