Non-intrusive Load Monitoring Approach Research Articles

Conditioned fully convolutional denoising autoencoder for multi-target NILM

AbstractEnergy management requires reliable tools to support decisions aimed at optimising consumption. Advances in data-driven models provide techniques like Non-Intrusive Load Monitoring (NILM), which estimates the energy demand of appliances from total consumption. Common single-target NILM approaches perform energy disaggregation by using separate learned models for each device. However, the use of single-target systems in real scenarios is computationally expensive and can obscure the interpretation of the resulting feedback. This study assesses a conditioned deep neural network built upon a Fully Convolutional Denoising AutoEncoder (FCNdAE) as multi-target NILM model. The network performs multiple disaggregations using a conditioning input that allows the specification of the target appliance. Experiments compare this approach with several single-target and multi-target models using public residential data from households and non-residential data from a hospital facility. Results show that the multi-target FCNdAE model enhances the disaggregation accuracy compared to previous models, particularly in non-residential data, and improves computational efficiency by reducing the number of trainable weights below 2 million and inference time below 0.25 s for several sequence lengths. Furthermore, the conditioning input helps the user to interpret the model and gain insight into its internal behaviour when predicting the energy demand of different appliances.

Occupancy detection for enhanced energy disaggregation

Non-Intrusive Load Monitoring (NILM) attempts to break down the aggregated electrical consumption signal into the power consumption of each individual appliance, which can provide helpful understanding on energy consumption patterns and helps reduce overall energy usage and costs. This paper proposes an occupancy-aided energy disaggregation approach to address the NILM problem. Our methodology encompasses three key steps: firstly, features extraction from environmental sensors through the training of a DAE model; secondly, inference of occupancy information using the K-means algorithm; and finally, the disaggregation process using a Recurrent Neural Network (RNN) model, incorporating the detected occupancy status alongside power data. Experiments conducted on our real-world dataset demonstrate that our method significantly outperforms the state-of-the-art models while having good generalization capacity, achieving roughly 40% Mean Absolute Error (MAE) gain and 30% Root Mean Squared Error (RMSE) gain on a specific appliances disaggregation compared to the conventional NILM approach where only the power data is used.

Performance-Aware NILM Model Optimization for Edge Deployment

Non-Intrusive Load Monitoring (NILM) describes the extraction of the individual consumption pattern of a domestic appliance from the aggregated household consumption. Nowadays, the NILM research focus is shifted towards practical NILM applications, such as edge deployment, to accelerate the transition towards a greener energy future. NILM applications at the edge eliminate privacy concerns and data transmission-related problems. However, edge resource restrictions pose additional challenges to NILM. NILM approaches are usually not designed to run on edge devices with limited computational capacity, and therefore model optimization is required for better resource management. Recent works have started investigating NILM model optimization, but they utilize compression approaches arbitrarily without considering the trade-off between model performance and computational cost. In this work, we present a NILM model optimization framework for edge deployment. The proposed edge optimization engine optimizes a NILM model for edge deployment depending on the edge device’s limitations and includes a novel performance-aware algorithm to reduce the model’s computational complexity. We validate our methodology on three edge application scenarios for four domestic appliances and four model architectures. Experimental results demonstrate that the proposed optimization approach can lead up to a 36.3% average reduction of model computational complexity and a 75% reduction of storage requirements.

Detection of Electric Bicycle Indoor Charging for Electrical Safety: A NILM Approach

The increasing fire accidents caused by electric bicycle (EB) indoor charging have raised great concerns. To ensure residential electrical safety, EB is prohibited from charging indoors by regulations, but efficient monitoring of EB indoor charging behaviour in pratice still remains as a challenging problem. From the perspective of non-intrusive load monitoring (NILM), this paper proposes a holistic approach to detect the EB indoor charging behaviour. Our approach utilizes four key elements: (i) a multi-window based cumulative sum algorithm to capture the turning-on event from aggregate power; (ii) a twostage feature selection method to determine the optimal features to form an effective appliance signature for load identification; (iii) a one-class support vector machine classifier to identify whether the turned-on appliance is EB or other appliances in a semi-supervised way; (iv) a Raspberry-Pi based edge device to implement the overall NILM algorithm. The experiment results verify the effectiveness of our NILM based approach, which can obtain a high detection performance with a lowdimensional feature space, thus providing a cost-effective and high-performance solution for EB indoor charging detection.

Non-intrusive condition monitoring based on event detection and functional data clustering

Implementing monitoring electricity consumption strategies in industrial environments provides improvements in both the maintenance process and energy efficiency. The contribution of this work is an industry-oriented non-intrusive load monitoring approach based on an unsupervised algorithm, encompassing a method of event detection, functional data clustering, and condition monitoring. With this method, multiple devices can be monitored by only one electric meter in industrial environments, enhancing the early detection of anomalies and energy inefficiencies. The proposed approach presents a robust event detection to deal with different industrial contexts due to its intuitive parameters, which allows adapting the detection to be more sensitive or to filter out higher noise variations. Unlike other feature-based clusterings, the proposed functional data clustering enables high-precision identification of transient state patterns, characterizing specific shapes for each pattern to properly cluster them, and provides higher reliability during load detection. Thus, this load detection segments the power consumption to extract transient states and identify which load acts on each event based on functional data clusters. By detecting when loads start to consume, the proposed energy disaggregation extracts the frequency spectrum of each device from the aggregate current consumption, which is used in a condition monitoring strategy to track the spectrum behavior of each device. In this way, the condition of multiple loads can be monitored using a single electric meter, whose information can be relevant to accurately schedule maintenance interventions and detect anomalies or inefficiencies early. The proposed approach was validated in three industrial contexts: The load detection accuracy was verified in an industrial testbed, giving a precision higher than 99% for monitoring five devices. The second and third industrial scenarios validate the accuracy of the proposed condition monitoring method. The last scenario was carried out at Bilbao airport to track the condition of multiple conveyor belts of a baggage handling system located at check-in. As a result, the degradation trend of three sets of conveyor belts was monitored.

A Novel Schematic Design of Enhanced State Gated Transformers for Energy Disaggregation of Nonintrusive Load Monitoring

Non-intrusive load monitoring (NILM) is a technique for estimating the energy consumption of individual appliances without the need for direct metering. In this letter, we examine the attention mechanism of transformers and present two transformer-based NILM approaches, named State Gated Transformer (SGT) and enhanced State Gated Transformer (eSGT) to learn complex non-linear relationships between aggregate and individual power consumption of appliances. The proposed SGT employs parallel self-attention networks, whereas the proposed eSGT enhances the former with a cross-coupled pair structure. We present the schematic design of transformers which allows for concise characterization of the attention mechanism, inspiring the exploration of novel designs at the circuit level. Our experimental results demonstrate the effectiveness and versatility of the proposed approach for real-world NILM datasets.

Explainability-Informed Feature Selection and Performance Prediction for Nonintrusive Load Monitoring

With the massive, worldwide, smart metering roll-out, both energy suppliers and users are starting to tap into the potential of higher resolution energy readings for accurate billing, improved demand response, improved tariffs better tuned to users and the grid, and empowering end-users to know how much their individual appliances contribute to their electricity bills via nonintrusive load monitoring (NILM). A number of NILM approaches, based on machine learning (ML), have been proposed over the years, focusing on improving the NILM model performance. However, the trustworthiness of the NILM model itself has hardly been addressed. It is important to explain the underlying model and its reasoning to understand why the model underperforms in order to satisfy user curiosity and to enable model improvement. This can be done by leveraging naturally interpretable or explainable models as well as explainability tools. This paper adopts a naturally interpretable decision tree (DT)-based approach for a NILM multiclass classifier. Furthermore, this paper leverages explainability tools to determine local and global feature importance, and design a methodology that informs feature selection for each appliance class, which can determine how well a trained model will predict an appliance on any unseen test data, minimising testing time on target datasets. We explain how one or more appliances can negatively impact classification of other appliances and predict appliance and model performance of the REFIT-data trained models on unseen data of the same house and on unseen houses on the UK-DALE dataset. Experimental results confirm that models trained with the explainability-informed local feature importance can improve toaster classification performance from 65% to 80%. Additionally, instead of one five-classifier approach incorporating all five appliances, a three-classifier approach comprising a kettle, microwave, and dishwasher and a two-classifier comprising a toaster and washing machine improves classification performance for the dishwasher from 72% to 94% and the washing machine from 56% to 80%.

Apply Graph Signal Processing on NILM: An Unsupervised Approach Featuring Power Sequences.

As a low-cost demand-side management application, non-intrusive load monitoring (NILM) offers feedback on appliance-level electricity usage without extra sensors. NILM is defined as disaggregating loads only from aggregate power measurements through analytical tools. Although low-rate NILM tasks have been conducted by unsupervised approaches based on graph signal processing (GSP) concepts, enhancing feature selection can still contribute to performance improvement. Therefore, a novel unsupervised GSP-based NILM approach with power sequence feature (STS-UGSP) is proposed in this paper. First, state transition sequences (STS) are extracted from power readings and featured in clustering and matching, instead of power changes and steady-state power sequences featured in other GSP-based NILM works. When generating graph in clustering, dynamic time warping distances between STSs are calculated for similarity quantification. After clustering, a forward-backward power STS matching algorithm is proposed for searching each STS pair of an operational cycle, utilizing both power and time information. Finally, load disaggregation results are obtained based on STS clustering and matching results. STS-UGSP is validated on three publicly accessible datasets from various regions, generally outperforming four benchmarks in two evaluation metrics. Besides, STS-UGSP estimates closer energy consumption of appliances to the ground truth than benchmarks.

A Non-Intrusive Load Monitoring Model for Electric Vehicles Based on Multi-Kernel Conventional Neural Network

With the widespread use of electric vehicles (EVs), the charging behavior of these resources has brought a large amount of load growth to the grid, leading to a series of problems such as increased peak valley load difference and line flow violation. Non-intrusive load monitoring (NILM) is a key technology that can be employed to monitor the multi-source load data information in the power grid and support the high-proportion access of electric vehicles. However, traditional NILM approaches are designed to identify the operation of household appliances and cannot be applied at the substation level directly due to frequent and intricate switching events of electrical equipment at this stage. In this paper, a NILM algorithm that can be applied for the monitoring of the charging behavior of electric vehicles at the substation level is proposed to support the high-proportion injection of distributed energy resources. The proposed approach employs a deep learning framework and a multi-kernel convolutional neural network (multi-kernel CNN) framework is used. The performance of the proposed method is verified on the self-organized datasets based on Pecan Street data and results showed that the obtained f1 score is over 90% for both the training sets and testing sets.

Real-Time Non-Intrusive Load Monitoring for Low-Power Appliances Using Odd Current Harmonics

Non-intrusive load monitoring (NILM) systems have gained immense popularity in the last two decades. Although many solutions were developed using the NILM approach, the majority of solutions failed to identify low-power-consuming appliances. However, a NILM system deployed in a domestic installation should be capable of identifying low-power-consuming appliances precisely. Therefore, we propose two real-time NILM solutions that can identify low-power residential appliances. These two solutions rely on active power and harmonic information extracted from the measured current waveform. We analyzed the performances of the two solutions using simulations and real-world experiments. Results indicate that both solutions can correctly identify appliances with an average accuracy above 90 percent.

Non-Intrusive Load Monitoring: A Review

The rapid development of technology in the electrical energy sector within the last 20 years has led to growing electric power needs through the increased number of electrical appliances and automation of tasks. In parallel the global climate protection goals, energy conservation and efficient energy management arise interest for reduction of the overall energy consumption. These requirements have led to the recent adoption of smart-meters and smart-grids, as well as to the rise of Load Monitoring (LM) using energy disaggregation, also referred to as Non-Intrusive Load Monitoring (NILM), which enables appliance-specific energy monitoring by only observing the aggregated energy consumption of a household. The real-time information on appliance level can be used to get deeper insights in the origin of energy consumption and to make optimization, strategic load scheduling and demand management feasible. The three main contributions are as follows: First, a generalized up-to-date review of NILM approaches including a high-level taxonomy of NILM methodologies is provided. Second, previously published results are grouped based on the experimental setup which allows direct comparison. Third, the article is accompanied by a software implementation of the described NILM approaches.

An Efficient Lightweight Event Detection Algorithm for On-Site Non-Intrusive Load Monitoring

Non-intrusive load monitoring (NILM) aims to determine individual-appliance energy consumption with minimum cost by decomposing aggregated electricity measurements. Although important for achieving energy conservation and cost minimization, NILM requires high-frequency sampling rates to provide accurate results. This requirement significantly increases the need for storage and computational resources in the electric utility’s fog/cloud infrastructure and for bandwidth on the customer’s side. To resolve these issues, on- site disaggregation, i.e., on the monitoring device, can be employed. However, to keep device-cost low, lightweight NILM algorithms are needed. To this end, a lightweight event-detection algorithm designed to ease on- site implementation, on either software or hardware, is proposed. Event detection is the first, critical half of the well-established event-based NILM approach; it identifies appliance state changes (events). Although a few lightweight event-detection techniques, utilizing high-frequency data, have been presented in the literature, their performance is relatively low in complex-load cases. The proposed algorithm utilizes simple-to-compute features and employs multiple simple criteria to declare an event as detected and slope-coefficient inspection to identify steady states. Moreover, it can detect events with very small time difference between them. Comparisons show that its performance is superior even against more complex event-detection approaches, while its low computational cost is also verified.

Energy Disaggregation Using Multi-Objective Genetic Algorithm Designed Neural Networks

Energy-saving schemes are nowadays a major worldwide concern. As the building sector is a major energy consumer, and hence greenhouse gas emitter, research in home energy management systems (HEMS) has increased substantially during the last years. One of the primary purposes of HEMS is monitoring electric consumption and disaggregating this consumption across different electric appliances. Non-intrusive load monitoring (NILM) enables this disaggregation without having to resort in the profusion of specific meters associated with each device. This paper proposes a low-complexity and low-cost NILM framework based on radial basis function neural networks designed by a multi-objective genetic algorithm (MOGA), with design data selected by an approximate convex hull algorithm. Results of the proposed framework on residential house data demonstrate the designed models’ ability to disaggregate the house devices with excellent performance, which was consistently better than using other machine learning algorithms, obtaining F1 values between 68% and 100% and estimation accuracy values ranging from 75% to 99%. The proposed NILM approach enabled us to identify the operation of electric appliances accounting for 66% of the total consumption and to recognize that 60% of the total consumption could be schedulable, allowing additional flexibility for the HEMS operation. Despite reducing the data sampling from one second to one minute, to allow for low-cost meters and the employment of low complexity models and to enable its real-time implementation without having to resort to specific hardware, the proposed technique presented an excellent ability to disaggregate the usage of devices.

A mixed-integer programming approach for industrial non-intrusive load monitoring

With the development of the smart grid, more load data can be collected and utilized to facilitate bidirectional communications between the supply-side and demand-side. More detailed data can achieve better results in applications, but it is not feasible due to sensor placement and data transmission limitations. Non-intrusive load monitoring (NILM) is an exceptionally low-cost solution for providing appliance-level load information of a building without installing extra sensors. While most current studies focus on NILM in residential buildings, the potential benefit of industrial NILM is preferable due to the grander power consumption scale and more professional management. This paper considers the challenges and strengths of industrial NILM, and a mixed-integer programming NILM approach is proposed for flowline industries. This approach separately models equipment with different load features, resulting in better performance in industrial scenarios containing various loads. Temporal dependencies between equipment are exploited by recognizing the statistical regularity of load fluctuation and are introduced as a novel flowline constraint in the model. A pulse width constraint that restricts the state duration of equipment is also introduced to improve performance. Several cases are designed and tested on two public industrial datasets to evaluate the effectiveness and transferability of the model. The classical factorial hidden Markov model (FHMM), a state-of-the-art matrix factorization method, and a deep learning model are used as benchmarks for comparison.

An Event Matching Energy Disaggregation Algorithm Using Smart Meter Data

Energy disaggregation algorithms disintegrate aggregate demand into appliance-level demands. Among various energy disaggregation approaches, non-intrusive load monitoring (NILM) algorithms requiring a single sensor have gained much attention in recent years. Various machine learning and optimization-based NILM approaches are available in the literature, but bulk training data and high computational time are their respective drawbacks. Considering these drawbacks, we devised an event matching energy disaggregation algorithm (EMEDA) for NILM of multistate household appliances using smart meter data. Having limited training data, K-means clustering was employed to estimate appliance power states. These power states were accumulated to generate an event database (EVD) containing all combinations of appliance operations in their various states. Prior to matching, the test samples of aggregate demand events were decreased by event-driven data compression for computational effectiveness. The compressed test events were matched in the sorted EVD to assess the contribution of each appliance in the aggregate demand. To counter the effects of transient spikes and/or dips that occurred during the state transition of appliances, a post-processing algorithm was also developed. The proposed approach was validated using the low-rate data of the Reference Energy Disaggregation Dataset (REDD). With better energy disaggregation performance, the proposed EMEDA exhibited reductions of 97.5 and 61.7% in computational time compared with the recent smart event-based optimization and optimization-based load disaggregation approaches, respectively.

Transferable Tree-Based Ensemble Model for Non-Intrusive Load Monitoring

Sustainable energy management systems have been increasingly studied in recent years. Non-intrusive load monitoring (NILM), as a key component, estimates the power consumption of individual appliances from the main readings only. However, most NILM approaches are computationally expensive, and their generality is negatively affected by the data drift occurred when the models are used across domains. Besides, the threats of privacy violation will rise in the model transfer due to the possible leakage of the personal information of the users from the source domain. To address all these challenges, we designed a cost-efficient learning method using LightGBM for energy disaggregation. We also proposed a model-based transfer learning algorithm using feature importance analysis, which enhances the generalisation capability of tree-based ensemble models applied in different domains while protecting privacy. We conducted experiments with real-world data sets. The performance of our approach is superior to the state-of-the-art solutions.

Quantifying Compressed Air Leakage through Non-Intrusive Load Monitoring Techniques in the Context of Energy Audits

Following the recent European directives highlighting the need to increase energy efficiency in the European Union, this work aims to show the possibility of using Non-Intrusive Load Monitoring (NILM) techniques to improve energy audits by estimating the compressed air leakage from a dataset of a tertiary building. The first step towards the reduction of energy consumption is performing an energy audit, in which a detailed analysis of the energy performance is executed. This analysis usually uses on-site measured data by the auditors. However, the time available for these measurements is limited and may not include some modes of operation. One example of that is the quantification of compressed air leaks. This task can be performed by estimating the flow rate during a no compressed air consumption period. However, these periods may not coincide with the auditors’ original schedule. This problem could be addressed by using historical data. Nevertheless, historical data from energy management systems usually are only available for global consumption, and rarely for individual appliances. In this context, a NILM approach would be helpful to enhance energy audits carrying analysis of modes of operation not included in the on-site measurements. In this paper, the leaks are firstly quantified using measurements mostly for benchmarking purposes. The results suggested 62% of leaks in the study case. In a second step, the Factorial Hidden Markov Model (FHMM) was applied to the data. Five typical working days, simulating the context of an energy audit, were used as training data, while one week during vacation time, with no compressed air consumption, was used to quantify the leaks. The results show that it was possible, in the context of an energy audit, to estimate the compressed air leakage using NILM techniques in this dataset with less than a 1% difference when compared to the estimation made with actual measurement. Finally, savings estimations considering the elimination of the leaks were performed, varying between 10% and 100% of the leakage repair. Considering the ideal scenario of complete leaks elimination, the savings would represent around 44% in the compressed air system and 4.75% of the current annual global consumption.

Impact of NILM-based energy efficiency on environmental degradation and kuznets hypothesis analysis

Nonintrusive load monitoring (NILM) breaks down the aggregated electrical consumption data into individual appliances. The feedback of disaggregated data to the consumers enables awareness and behaviour change to conserve electricity, consequently reducing CO2 emissions to the environment. However, the limited literature regarding the impact of NILM and Kuznets hypothesis (EKC) analysis on CO2 emissions reduction has restricted policymakers in developing effective mitigation measures. This work aims to assess the impact of NILM-based based energy efficiency (EE) on environmental improvement. The combined approach of scenario simulation and EKC analysis was adopted to gauge the effectiveness of NILM that leads to sustainable development. The monotonically increase relationship between environmental degradation and economic growth in Malaysia without peaking beyond 2030 implies that the current mitigation measures and policies imposed may not effectively cope with the future power demands for sustainable development. NILM-based EE measures could be a great potential for reducing CO2 emissions by 10.2%. The inverted-U curves and reduced turning points of environmental degradation from the income level of USD 20,063.36 to USD 16,305.19. Therefore, NILM approach can accelerate sustainable development with lower environmental deterioration. The work may beneficial to policymakers to analyse the impact and effectiveness of mitigation measures quantitatively.

Trainingless Energy Disaggregation Without Plug-Level Sensing

The majority of studies in non-intrusive load monitoring (NILM) are intrusive and supervised in nature since they require appliance-wise energy consumption data for training. A “true” NILM approach should not require appliance-wise data even for training; this automatically demands the approach to be unsupervised in nature. Four prior studies fall under this category; two of them use integer programming; the third employs graph signal processing (GSP); the fourth one assumes a pre-trained model that generalizes without the requirement of sub-metered training data. This work proposes an alternative approach based on the sparse coding (SC) formulation. Experiments on benchmarks datasets [reference energy disaggregation dataset (REDD) and Pecan Street] show that the proposed method improves by approximately 10%–20% on REDD and about 15% on Pecan Street, over existing true/unsupervised non-intrusive techniques; however, it is worse than the intrusive/supervised approaches on REDD (3%–10%) and Pecan Street (10%).

Exploiting Smart Meter Power Consumption Measurements for Human Activity Recognition (HAR) with a Motif-Detection-Based Non-Intrusive Load Monitoring (NILM) Approach.

Numerous approaches exist for disaggregating power consumption data, referred to as non-intrusive load monitoring (NILM). Whereas NILM is primarily used for energy monitoring, we intend to disaggregate a household’s power consumption to detect human activity in the residence. Therefore, this paper presents a novel approach for NILM, which uses pattern recognition on the raw power waveform of the smart meter measurements to recognize individual household appliance actions. The presented NILM approach is capable of (near) real-time appliance action detection in a streaming setting, using edge computing. It is unique in our approach that we quantify the disaggregating uncertainty using continuous pattern correlation instead of binary device activity states. Further, we outline using the disaggregated appliance activity data for human activity recognition (HAR). To evaluate our approach, we use a dataset collected from actual households. We show that the developed NILM approach works, and the disaggregation quality depends on the pattern selection and the appliance type. In summary, we demonstrate that it is possible to detect human activity within the residence using a motif-detection-based NILM approach applied to smart meter measurements.