Non-intrusive Load Monitoring Algorithms Research Articles

Transient dataset of household appliances with Intensive switching events

With the development of Non-Intrusive Load Monitoring (NILM), it has become feasible to perform device identification, energy consumption decomposition, and load switching detection using Deep Learning (DL) methods. Similar to other machine learning problems, the research and validation of NILM necessitate substantial data support. Moreover, different regions exhibit distinct characteristics in their electricity environments. Therefore, there is a need to provide open datasets tailored to different regions. In this paper, we introduce the Transient Dataset of Household Appliances with Intensive Switching Events (TDHA25). This dataset comprises switch instantaneous data from 10 typical household appliances in China. The TDHA dataset features a high sampling rate, accurate labelling, and realistic representation of actual appliance start-up waveforms. Additionally, appliance switching is achieved through precise control of relay switches, thus mitigating interference caused by mechanical switches. By furnishing such a dataset, we aim not only to enhance the recognition accuracy of existing NILM algorithms but also to facilitate the application of NILM algorithms in regions sharing similar electricity consumption characteristics to those of China.

How Can Non-Intrusive Load Monitoring Contribute to the Assessment of the Smart Readiness Indicator?

The Smart Readiness Indicator (SRI) is a framework introduced by the EU in 2018 to assess smart buildings in various aspects. However, the SRI has been criticized for several limitations, including its ambiguous service definitions. This paper proposes the application of Non-Intrusive-Load Monitoring (NILM) technology to enhance SRI evaluation on the example of SRI service E-12. NILM can be used to disaggregate energy consumption data to end use levels and allows for granular non-intrusive energy consumption measurement. The study involves a rigorous methodology using open sensor data and NILM algorithms to evaluate device-specific energy consumption We evaluate the IDEAL dataset and three different frequencies (5s, 15min, 1h), three different algorithms (CO, RNN, Seq2Point) and one data imputation strategies (forward filling). The results show that with a higher frequency, the performance metrics (F-score, normalized absolute error) increase. Regarding further considerations, we identify a trade-off between resource and energy efficiency, as well as privacy considerations with increasing measurement frequency. To achieve its aims for awareness, the SRI needs to consider interoperability and appropriate aggregations (frequency and spatial).

Perfednilm: a practical personalized federated learning-based non-intrusive load monitoring

Non-Intrusive Load Monitoring (NILM) is a valuable technique for breaking down overall power consumption into the energy usage of individual appliances. Understanding power usage patterns through NILM plays an important role in reducing energy costs and achieving carbon reduction goals. However, privacy concerns often deter consumers from sharing their electricity consumption data. To address these privacy concerns, Federated Learning (FL) has been introduced in NILM, which enables the training of NILM models while keeping power consumers’ data locally. However, FL’s reliance on a single global model leads to poor performance on clients with unique power consumption patterns. In response to this challenge, we present a Personalized Federated Learning NILM algorithm (PerFedNILM), a practical personalized FL approach for NILM. PerFedNILM limits the local update bias across clients and trains personalized models for individual clients to improve load-monitoring performance. In addition, it mitigates the negative impact of client dropout, which is a common issue in practice. Our experiments on using real-world energy data demonstrate that PerFedNILM outperforms previous FL-based NILM methods, especially in client dropout scenarios.

Non-intrusive load monitoring algorithm for coarse-grained electricity consumption data based on YOLO deep neural networks

Non-intrusive load monitoring algorithm for coarse-grained electricity consumption data based on YOLO deep neural networks

A Field Study of Nonintrusive Load Monitoring Devices and Implications for Load Disaggregation.

Evaluations of nonintrusive load monitoring (NILM) algorithms and technologies have mostly occurred in constrained, artificial environments. However, few field evaluations of NILM products have taken place in actual buildings under normal operating conditions. This paper describes a field evaluation of a state-of-the-art NILM product, tested in eight homes. The match rate metric-a technique recommended by a technical advisory group-was used to measure the NILM's success in identifying specific loads and the accuracy of the energy consumption estimates. A performance assessment protocol was also developed to address common issues with NILM mislabeling and ground-truth comparisons that have not been sufficiently addressed in past evaluations. The NILM product's estimates were compared to the submetered consumption of eight major appliances. Overall, the product had good performance in disaggregating the energy consumption of the electric water heaters, which included both electric resistance and heat-pump water heaters, but only a fair accuracy with refrigerators, dryers, and air conditioners. The performance was poor for cooking equipment, furnace fans, clothes washers, and dishwashers. Moreover, the product was often unable to detect major loads in homes. Typically, two or more appliances were not detected in a home. At least two dryers, furnace fans, and air conditioners went undetected across the eight homes. On the other hand, the dishwasher was detected in all homes where available or monitored. The key findings were qualitatively compared to those of past field evaluations. Potential areas for improvement in NILM product performance were determined along with areas where complementary technologies may be able to aid in load-disaggregation applications.

A real-time carbon emission estimation framework for industrial parks with non-intrusive load monitoring

Accurate carbon emission accounting is critical for combating climate change. Real-time estimation of emissions can further facilitate precise carbon emission control, emission reduction planning, and post-analysis. Industrial parks are significant carbon emitters and require more attention to their carbon emissions and environmental impact. However, current carbon emission estimation methods for industrial parks have limitations in terms of time lag, inconsistent standards, and low reliability. In response, an innovative real-time carbon emission estimation framework for industrial parks is proposed to improve the estimation accuracy while taking into account both Scope 1 and Scope 2 emissions. It is a data-driven method based on non-intrusive load monitoring (NILM) algorithms and reliable real-time meter data. It estimates the emissions of the factories based on factory-level data and then aggregates it. Scope 1 emissions are calculated based on the direct carbon emission factors of the devices (DEFs) and the corresponding device states obtained by a novel NILM algorithm named Adaptive Weighted Recurrence Graph-Convolutional Neural Network-Bidirectional Long Short-Term Memory (AWRG-CNN-BLSTM), achieving an average identification accuracy of 93.4%. Scope 2 emissions from imported electricity or heat are calculated using more accurate electricity-related or heat-related emission factors, locational marginal carbon emission factors of electricity (MEFs), and locational carbon emission factors of heat consumption (LEFs), respectively. The experiments on an industrial park with four factories demonstrate the effectiveness of our proposed method. The carbon emission estimation error is only 0.44% in one year, which is 2.32% lower than that of the IPCC method. Moreover, real-time estimation enables emissions tracking.

Customer-Centered Pricing Strategy Based on Privacy-Preserving Load Disaggregation

Demand response (DR) is a demand reduction or shift of electricity use by customers to make electricity systems flexible and reliable, which is beneficial under increasing shares of intermittent renewable energy. For residential loads, thermostatically controlled loads (TCLs) are considered as major DR resources. In a price-based DR program, an aggregation agent, such as a retailer, formulates price signals to stimulate the customers to change electricity usage patterns. The conventional DR management methods fully rely on mathematical models to describe the customer’s price responsiveness. However, it is difficult to fully master the customers’ detailed demand elasticities, cost functions, and utility functions in practice. Hence, in this paper, we proposed a data-driven non-intrusive load monitoring (NILM) approach to study the customers’ power consumption behaviors and usage characteristics. Based on NILM, the DR potential of the TCLs can be properly estimated, which assists the retailer in formulating a proper pricing strategy. To realize privacy protection, a privacy-preserving NILM algorithm is proposed. The proposed methodologies are verified in case studies. It is concluded that the proposed NILM algorithm not only reaches a better prediction performance than state-of-art works but also can protect privacy by slightly sacrificing accuracy. The DR pricing strategy with NILM integrated brings more profit and lower risks to the retailer, whose results are close to the fully model-based method with strong assumptions. Furthermore, a NILM algorithm with higher performance can help the retailer earn more benefits and help the grids better realize DR 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.

DeepEdge-NILM: A case study of non-intrusive load monitoring edge device in commercial building

Non-intrusive load monitoring (NILM) has become an emerging technology in the energy sector for its effectiveness in the energy disaggregation of individual loads from the measured aggregated energy in main power supply of building. NILM has also evolved gradually over the past few decades, due to the significant advancements in artificial intelligence (AI), embedded/edge devices, and internet of things (IoT). From the review of literature, we observed that most of the NILM systems were only capable of monitoring different loads and it fails to detect and disaggregate the energy consumption of similar loads effectively. Further, most of works in the literature were focused on improving the performance of energy disaggregation of different loads in residential buildings and very limited works have been carried out for the similar loads in NILM system. Therefore, there is a necessity to study and evaluate the state of the art NILM algorithms for similar loads in real case scenarios for effective implementation in the commercial and industrial buildings. In this paper, we present a case study of the developed DeepEdge-NILM device which has been installed in a commercial building (office environment), where multiple similar and identical air conditioners (AC's) are used in the monitoring environment. A deep learning framework, Long Short-Term Memory (LSTM) has been implemented on the new set of features which comprises of electrical features and intrinsic features derived from the basic electric features using geometric mean for capturing the signature of similar loads effectively. From our case study, we observed that the performance of the deployed NILM system outperforms the baseline systems that have been developed using basic electrical features significantly. Finally, the limitations and future scope of the NILM edge device in the commercial buildings are also discussed.

Event-driven non-intrusive load monitoring algorithm based on targeted mining multidimensional load characteristics

Nowadays, the advancement of nonintrusive load monitoring (NILM) has been hastened by the ever-increasing requirements for the reasonable use of electricity by users and demand side management. Although existing researches have tried their best to extract a wide variety of load features based on transient or steady state of electrical appliances, it is still very difficult for their algorithm to model the load decomposition problem of different electrical appliance types in a targeted manner to jointly mine their proposed features. This paper presents a very effective event-driven NILM solution, which aims to separately model different appliance types to mine the unique characteristics of appliances from multi-dimensional features, so that all electrical appliances can achieve the best classification performance. First, we convert the multi-classification problem into a serial multiple binary classification problem through a pre-sort model to simplify the original problem. Then, ConTrastive Loss K-Nearest Neighbour (CTLKNN) model with trainable weights is proposed to targeted mine appliance load characteristics. The simulation results show the effectiveness and stability of the proposed algorithm. Compared with existing algorithms, the proposed algorithm has improved the identification performance of all electrical appliance types.

An active learning framework for the low-frequency Non-Intrusive Load Monitoring problem

With the widespread deployment of smart meters worldwide, quantification of energy used by individual appliances via Non-Intrusive Load Monitoring (NILM), i.e., virtual submetering, is an emerging application to inform energy management within buildings. Low-frequency NILM refers to NILM algorithms designed to perform load disaggregation at sampling rates in the order of seconds and minutes, as per smart meter data availability. Recently, many deep learning solutions for NILM have appeared in the literature, with promising results. However, besides requiring large, labelled datasets, the proposed deep learning models are not flexible and usually under-perform when tested in a new environment, affecting scalability. The dynamic nature of appliance ownership and usage inhibits the performance of the developed supervised NILM models and requires large amounts of training data. Transfer learning approaches are commonly used to overcome this issue, but they often assume availability of good quality labelled data from the new environment. We propose an active learning framework, that is able to learn and update the deep learning NILM model from small amounts of data, for transfer to a new environment. We explore the suitability of different types of acquisition functions, which determines which function inputs are most valuable. Finally, we perform a sensitivity analysis of the hyperparameters on model performance. In addition, we propose a modification to the state-of-the-art BatchBALD acquisition function, to address its high computational complexity. Our proposed framework achieves optimal accuracy-labelling effort trade-off with only 5%–15% of the query pool labelled. The results on the REFIT dataset, demonstrate the potential of the proposed active learning to improve transferability and reduce the cost of labelling. Unlike the common approach of retraining the entire model once a new set of labels is provided, we demonstrate that full re-training is not necessary, since a fine-tuning approach can offer a good trade-off between performance achieved and computational resources needed.

A non-intrusive load decomposition method of resident by multi-scale attention mechanism

Non-intrusive load monitoring (NILM) is one of the important technologies in home energy management and power demand response scenario. However, the presence of multi-mode appliances and appliances with close power values have affected in diminishing the accuracy of identification based NILM algorithms. To tackle these challenges, the work proposes a resident load decomposition method combining multi-scale attention mechanism and convolutional neural network. At the first stage, the attention scores of the normal load data at the previous few moments of the attention model are smoothed dynamically against the abnormal scores at the current moment. The load identification attention model is optimized by constraint factors. Then, on this basis, convolution filters of different sizes are used to model the mixed load data of different electrical equipment, to mine more abundant characteristic information. Finally, to illustrate the proposed processes and validate its effectiveness, taking the PLAID data set as an example, the method proposed in the article is compared with respect to the existing NILM techniques. The experimental results show that the method based on the multi-scale attention mechanism in this paper can greatly improve the effect of load decomposition. Moreover, it reduces the confusion problem of electrical appliance identification with similar load characteristics.

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.

Very low-resolution residential load disaggregation with unsupervised graph signal processing

An unsupervised graph signal processing-based non-intrusive load monitoring algorithm is used to disaggregate very low-resolution 15-min real power signals from advanced metering infrastructure smart meters. This study presents the application of load disaggregation to new datasets that are more representative of single-family houses within the U.S. than those previously presented in the graph signal processing-based load disaggregation literature. Filtering is performed on signals within the signal database used for labeling reconstructed signals resulting from disaggregation. Results demonstrate improved accuracy due to filtering target device signals. Level 2 electric vehicle chargers are demonstrated to be the most promising candidate for very low-resolution load disaggregation using graph signal processing, with most houses demonstrating FM > 0.8 and absolute percent error in energy estimation less than 25%. The implications of these results on demand response assessment are presented, with an estimated 1350 MWh of energy available during peak demand periods from EV chargers for demand response applications in regions with acceptable disaggregation accuracy.

Neural Load Disaggregation: Meta-Analysis, Federated Learning and Beyond

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.

Leveraging sequence-to-sequence learning for online non-intrusive load monitoring in edge device

Non-intrusive load monitoring (NILM), extracting the appliances’ usage profiles by decomposing a household’s aggregate electricity consumption, has become increasingly important for home energy management. Studies show that users are more likely to adjust their usage behaviors if the real-time feedback on their energy consumption details is provided. Recent state-of-the-art deep learning models have delivered a good performance on load disaggregation; however, these models require tremendous computation and memory to tune huge numbers of parameters, which makes them difficult to be implemented on edge devices to derive real-time appliance usage information for users while keep user’s privacy. In this paper, a lightweight sequence-to-sequence neural network model consisting of only two convolutional layers and dense layers is proposed, which can provide users with both real-time appliance power usage and improved energy consumption profiles. The online results are produced by picking up every end-point of output sequences of the network, whereas the past energy profiles or offline results of appliance can be obtained by averaging all points of the output sequence. By implementing the model in edge device, the user’s privacy is maintained. The lightweight NILM algorithm is evaluated with low-frequency sampling data in two public available datasets: UKDALE and REFIT. Assessment results prove that the proposed model exhibits comparable or even better disaggregation performance with dramatically reduced model size and computation requirements, and thus demonstrates its application potential on edge devices.

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.

Reserve Evaluation and Energy Management of Micro-grids in Joint Electricity Markets Based on Non-Intrusive Load Monitoring

The heating, ventilation, and air-conditioning (HVAC) units are regarded as major demand response (DR) resources in micro-grids. However, due to privacy concerns and technical constraints, it is difficult for the system operator to obtain the complete information on each individual appliance. In this paper, we present a non-intrusive load monitoring (NILM) based framework for the operation strategy of the micro-grid in the joint energy, reserve, and regulation markets. The NILM technologies enable the operator to disaggregate the power of the HVAC units from the reading of the smart meters. Hence, the operation state of the appliances and the behavior of consumers can be studied without obtaining detailed data of each individual appliance. An advanced NILM algorithm is proposed, and the Coupled Generative Adversarial Networks (C-GAN) are utilized to enhance the generalizability of the trained model. Based on the NILM result, a novel method based on Hidden Markov Model (HMM) is proposed to evaluate the upward and downward reserve capacity of the HVAC units. The evaluated reserve capacity can help the operator better bid in the joint market based on the proposed optimization model. The proposed framework and methodology are verified through case studies. The simulation result reveals that with NILM, the market operator can save more energy consumption costs and load curtailment costs and earn more revenues in the joint market by selling excessive energy and providing ancillary services.

Incorporating coincidental water data into non-intrusive load monitoring

Non-intrusive load monitoring (NILM) is among successful approaches aiding residential energy management. However, the presence of multi-mode appliances and appliances with close power values and lack of a proper volume of training dataset have remained influential in worsening the computational complexity and diminishing the accuracy of classification-based NILM algorithms. To tackle these challenges, we propose a novel classification process, which considers the correlation of water and electricity consumption of some appliances as a novel signature in the network to improve the accuracy of disaggregation process in overlapping modes and tackle the lack of proper volume of training dataset. In the first phase of the proposed method, the K -nearest neighbors method, as a fast classification technique, is employed to extract power signals of appliances with exclusive non-close power values. Then, two different deep learning-based methods are proposed to disaggregate the consumption of appliances with close consumption values considering the correlation of electricity and water consumption of some appliances. Throughout these methods the water consumption of these appliances are also disaggregated. The main objectives of the proposed methods are increasing accuracy in the close modes of power consumption, and informing consumers about the water consumption pattern of some appliances. To illustrate the proposed processes and validate its effectiveness, the Almanac Minutely Power Dataset as a real dataset with a sampling rate of 1-minute is considered. The numerical results show marked improvement with respect to the existing classification-based NILM techniques. Moreover, it shows the applicability of proposed methods in dealing with low-frequency readings carried out by existing smart meters.

A multi-head Convolutional Neural Network based non-intrusive load monitoring algorithm under dynamic grid voltage conditions

In recent times, non-intrusive load monitoring (NILM) has emerged as an important tool for distribution-level energy management systems owing to its potential for energy conservation and management. However, load monitoring in smart building environments is challenging due to high variability of real-time load and varied load composition. Furthermore, as the volume and dimensionality of smart meters data increases, accuracy and computational time are key concerning factors. In view of these challenges, this paper proposes an improved NILM technique using multi-head (Mh-Net) convolutional neural network (CNN) under dynamic grid voltage conditions. An attention layer is introduced into the proposed CNN model, which helps in improving estimation accuracy of appliance power consumption. The performance of the developed model has been verified on an experimental laboratory setup for multiple appliance sets with varied power consumption levels, under dynamic grid voltages. Moreover, the effectiveness of the proposed model has been verified on widely used UK-DALE data, and its performance has been compared with existing NILM techniques. Results depict that the proposed model accurately identifies appliances, power consumptions and their time-of-use even during practical dynamic grid voltage conditions.