A NILM dataset is a valuable tool in the development of Non-Intrusive Load Monitoring techniques, as it provides a means of evaluation of novel techniques and algorithms, as well as for benchmarking. The figure of merit of a NILM dataset includes characteristics such as the sampling frequency of the voltage, current, or power, the availability of indications (ground-truth) of load events during recording, the variety and representativeness of the loads, and the variety of situations these loads are subject to. Considering such aspects, the proposed LIT-Dataset was designed, populated, evaluated, and made publicly available to support NILM development. Among the distinct features of the LIT-Dataset is the labeling of the load events at sample level resolution and with an accuracy and precision better than 5 ms. The availability of such precise timing information, which also includes the identification of the load and the sort of power event, is an essential requirement both for the evaluation of NILM algorithms and techniques, as well as for the training of NILM systems, particularly those based on Machine Learning.

A multi-agent architecture for a Non-Intrusive Load Monitoring (NILM) solution is presented and evaluated. The underlying rationale for such an architecture is that each agent (load event detection, feature extraction, and classification) outperforms others of the same type in particular scenarios; hence, by combining the expertise of these agents, the system presents an improved performance. Known NILM algorithms, as well as new algorithms, proposed by the authors, were individually evaluated and compared. The proposed architecture considers a NILM system composed of Load Monitoring Modules (LMM) that report to a Center of Operations, required in larger facilities. For the purposed of evaluating and comparing performance, five load event detect agents, five feature extraction agents, and five classification agents were studied so that the best combinations of agents could be implemented in LMMs. To evaluate the proposed system, the COOLL and the LIT-Dataset were used. Performance improvements were detected in all scenarios, with power-ON and power-OFF detection improving up to 13%, while classification accuracy improved up to 9.4%.

No abstract

Nonintrusive Load Monitoring (NILM) uses computational methods to disaggregate and classify electrical appliances signals. The classification is usually based on the power signatures of the appliances obtained by a feature extractor. State-of-the-art results were obtained extracting NILM features with convolutional neural networks (CNN). However, it depends on the training process with large datasets or data augmentation strategies. In this paper, we propose a feature extraction strategy for NILM using the Scattering Transform (ST). The ST is a convolutional network analogous to CNN. Nevertheless, it does not need a training process in the feature extraction stage, and the filter coefficients are analytically determined (not empirically, like CNN). We perform tests with the proposed method on different publicly available datasets and compare the results with state-of-the-art deep learning-based and traditional approaches (including wavelet transform and V-I representations). The results show that ST classification accuracy is more robust in terms of waveform parameters, such as signal length, sampling frequency, and event location. Besides, ST overcame the state-of-the-art techniques for single and aggregated loads (accuracies above 99% for all evaluated datasets), in different training scenarios with single and aggregated loads, indicating its feasibility in practical NILM scenarios.

No abstract