Non-intrusive load monitoring (NILM) techniques are central techniques to achieve the energy sustainability goals through the identification of operating appliances in the residential and industrial sectors, potentially leading to increased rates of energy savings. NILM received significant attention in the last decade, reflected by the number of contributions and systematic reviews published yearly. In this regard, the current paper provides a meta-analysis summarising existing NILM reviews to identify widely acknowledged findings concerning NILM scholarship in general and neural NILM algorithms in particular. In addition, this paper emphasizes federated neural NILM, receiving increasing attention due to its ability to preserve end-users’ privacy. Typically, by combining several locally trained models, federated learning has excellent potential to train NILM models locally without communicating sensitive data with cloud servers. Thus, the second part of the current paper provides a summary of recent federated NILM frameworks with a focus on the main contributions of each framework and the achieved performance. Furthermore, we identify the non-availability of proper toolkits enabling easy experimentation with federated neural NILM as a primary barrier in the field. Thus, we extend existing toolkits with a federated component, made publicly available and conduct experiments on the REFIT energy dataset considering four different scenarios.
13 September 2022 in IEEE Transactions on Industrial Informatics
Non-Intrusive Load Monitoring (nilm) techniques are increasingly becoming a key instrument for identifying the power consumption of individual appliances based on a single metering point. Particularly Deep learning models are gaining interest in this regard. However, the challenges brought by NILM datasets and the non-availability of common experimental guidelines tend to compromise comparison, research transparency and replicability. The limited adoption of efficient research instruments and a lack of best practices guidelines contribute in huge part to this problem, where no features, encouraging standardised formats for benchmarking and results sharing, are offered. To address these issues, we first present a brief overview of recent best practices for Deep Learning (DL) and highlight how deep NILM research can benefit from these practices. Furthermore, we suggest a novel open-source toolkit leveraging these practices: Deep-NILMTK. The proposed toolkit offers a common testing bed for NILM algorithms independently of the underlying deep learning framework with a modular NILM pipeline that can easily be customised. Furthermore, Deep-NILMTK introduces the concept of Experiment Templating to offer pre-designed experiments allowing to enhance research efficiency. Leveraging this concept and DL best practices, we present a case-study of creating an online NILM benchmark repository 11https://github.com/BHafsa/DNN-NILM-benchmark considering eight of the most popular deep NILM algorithms. All sources relative to the tool are made publicly available on Github 22https://github.com/BHafsa/deep-nilmtk-v1 along with the corresponding documentation.
Supporting the elderly to maintain their independence, safety, and well-being through Active Assisted Living (AAL) technologies, is gaining increasing momentum. Recently, Non-intrusive Load Monitoring (NILM) approaches have become the focus of these technologies due to their non-intrusiveness and reduced price. Whilst some research has been carried out in this respect; it still is challenging to design systems considering the heterogeneity and complexity of daily routines. Furthermore, scholars gave little attention to evaluating recent deep NILM models in AAL applications. We suggest a new interactive framework for activity monitoring based on custom user-profiles and deep NILM models to address these gaps. During evaluation, we consider four different deep NILM models. The proposed contribution is further assessed on two households from the REFIT dataset for a period of one year, including the influence of NILM on activity monitoring. To the best of our knowledge, the current study is the first to quantify the error propagated by a NILM model on the performance of an AAL solution. The results achieved are promising, particularly when considering the UNET-NILM model, a multi-task convolutional neural network for load disaggregation, that revealed a deterioration of only 10% in the f1-measure of the framework’s overall performance.
To support the increasing number of older people, new (assistive) technologies are constantly being developed. For these technologies to be used successfully, future users need to be trained. Due to demographic change, this will become difficult in the future, as the resources for training will no longer be available. In this respect, coaching robots could have great potential to support younger seniors in particular. However, there is little evidence in the literature about the perceptions and potential impact of this technology on the well-being of older people. This paper provides insights into the use of a robot coach (robo-coach) to train younger seniors in the use of a new technology. The study was carried out in Austria in autumn 2020, involving 34 participants equally distributed among employees in their last three years of service and retirees in their first three years of retirement (23 female; 11 male). The aim was to assess participants’ expectations and perceptions by examining the perceived ease of use and user experience of the robot in providing assistance during a learning session. The findings reveal a positive impression of the participants and promising results for using the robot as a coaching assistant in daily tasks.
Consumer's privacy is a main concern in Smart Grids (SGs) due to the sensitivity of energy data, particularly when used to train machine learning models for different services. These data-driven models often require huge amounts of data to achieve acceptable performance leading in most cases to risks of privacy leakage. By pushing the training to the edge, Federated Learning (FL) offers a good compromise between privacy preservation and the predictive performance of these models. The current paper presents an overview of FL applications in SGs while discussing their advantages and drawbacks, mainly in load forecasting, electric vehicles, fault diagnoses, load disaggregation and renewable energies. In addition, an analysis of main design trends and possible taxonomies is provided considering data partitioning, the communication topology, and security mechanisms. Towards the end, an overview of main challenges facing this technology and potential future directions is presented.
The need for reducing our energy consumption footprint and the increasing number of electric devices in today's homes is calling for new solutions that allow users to efficiently manage their energy consumption. Real-time feedback at device level would be of a significant benefit for this application. In addition, the aging population and their wish to be more autonomous have motivated the use of this same real-time data to indirectly monitor the household's occupants for their safety. By breaking down aggregate power consumption into its components, Non-Intrusive Load Monitoring provides information on individual appliances and their current state of operation. Since no additional metering equipment is required, residents are not confronted with intrusion into their familiar environment. Our work aims to depict an architecture supporting non-intrusive measurement with a smart electricity meter and the handling of these data using an open-source platform that allows to visualize and process real-time data about the total energy consumed. As a case study, we describe a series of measurements from common household devices and show how abnormal behavior can be detected.