Non-intrusive Load Monitoring Research Articles

A Generalizable Method for Practical Non-Intrusive Load Monitoring via Metric-Based Meta-Learning

Non-Intrusive Load Monitoring (NILM) has become a research boom due to its energy-saving benefits as the smart grid develops and the greenhouse effect intensifies. However, existing NILM models may perform poorly in the target domains collected from different sources, or those containing unseen appliance types, blocking the rollout of NILM. In contrast to the methods that up-date the trained model on the unseen data to handle domain shift or heterogeneous label space, this paper proposes a novel general-izable NILM method based on metric-based meta-learning and su-pervised pre-training. It beforehand learns the cross-domain meta-knowledge through meta-training to eliminate the need for parameter updates in the target domain. Moreover, this method is specifically designed for practical deployments at the edge where both labeled instances and computation resources are limited. The proposed method transforms the appliance current into a metric vector via lightweight 1D convolution and then uses the relation module to propagate the labels via similarity. The experimental results on three public datasets and the proposed dataset demon-strate that the proposed method can achieve high generalization performance in different datasets, users houses, appliance brands, and appliance types.

Data-Driven Carbon Footprint Management of Electric Vehicles and Emission Abatement in Electricity Networks

Electric vehicles (EVs) are believed to play an important role in mitigating carbon emissions in the transportation sector. However, EVs may still cause carbon emissions in the power sector if they are charged by electricity generated from burning fossil fuels like coal. Researchers have focused on managing emissions on the power generation side. However, the underlying driver of carbon emissions is the demand of consumers. To this end, a probabilistic carbon footprint management strategy is proposed for EVs in this paper. First, the conventional deterministic carbon emission flow model is extended to a probabilistic one (PCEF) to track the carbon footprint of EVs considering various uncertainties based on non-intrusive load monitoring (NILM) and the two-point estimation method (2PEM). Second, a stochastic chance-constrained carbon footprint management model for EV charging is proposed to address the carbon obligation allocation of EVs from the perspective of consumption and provide a technical basis for demand-driven stimulation to reduce carbon emissions. Third, a solution methodology is proposed to solve the formulated chance-constrained problem based on nonparametric Bayesian modeling and inference. The proposed model and methodologies are verified on the IEEE 39-bus system. The feasibility of the proposed PCEF model is validated. According to the simulation result, the computation speed of the proposed PCEF model is improved from 3456.8 seconds to 1341.3 seconds compared with the Monte Carlo simulation by sacrificing the accuracy within 2%. Besides, the proposed emission control strategy can realize a better emission control performance compared with the other state-of-art works. Compared with the base case, the total emission is reduced by 82 tons annually, and the emission reduction rate is increased by nearly twice.

Conditional-TimeGAN for Realistic and High-Quality Appliance Trajectories Generation and Data Augmentation in Nonintrusive Load Monitoring

Conditional-TimeGAN for Realistic and High-Quality Appliance Trajectories Generation and Data Augmentation in Nonintrusive Load Monitoring

Reconstruction-Based Supervised Contrastive Learning for Unknown Device Identification in Nonintrusive Load Monitoring

Reconstruction-Based Supervised Contrastive Learning for Unknown Device Identification in Nonintrusive Load Monitoring

Novel Fractal-Based Features for Low-Power Appliances in Non-Intrusive Load Monitoring

Novel Fractal-Based Features for Low-Power Appliances in Non-Intrusive Load Monitoring

A Holistic and Personalized Home Energy Management System With Non-Intrusive Load Monitoring

A Holistic and Personalized Home Energy Management System With Non-Intrusive Load Monitoring

VMD-GRU Based Non-Intrusive Load Monitoring For Home Energy Management System

This paper proposes a solution for Non-Intrusive Load Monitoring (NILM) in residential energy systems. The presented solution offers substantial benefits for load shifting tasks and home energy management systems, enabling effective load management and optimization. The proposed approach combines Variable Mode Decomposition (VMD) and Gated Recurrent Units Network (GRU). VMD parameter selection is done through the Salp Swarm Algorithm (SSA). Correlation analysis is used to identify the most relevant Intrinsic Mode Functions (IMFs) for each appliance. A comparison is made with MILD without VMD to demonstrate the impact of VMD on disaggregation performance. The proposed strategy yielded significant results, with an average RMSE of 4.10 (W) and an average MAPE of 4.14%. In contrast, the other approach had an average RMSE of 12.79 (W) and an average MAPE of 6.40%. These findings indicate that the proposed model performs better in disaggregation tasks.

Non-Intrusive Load Monitoring for Energy Management in Smart Grids Incorporating EVs

Non-Intrusive Load Monitoring for Energy Management in Smart Grids Incorporating EVs

Integrating Graph Signal Processing and Multitask Temporal Convolutional Networks for Household Nonintrusive Load Monitoring

Integrating Graph Signal Processing and Multitask Temporal Convolutional Networks for Household Nonintrusive Load Monitoring

Household Appliance Non-Intrusive Load Monitoring using Alternating Direction Method of Multipliers Based on Relaxation Distance and Neighborhood Search

Household Appliance Non-Intrusive Load Monitoring using Alternating Direction Method of Multipliers Based on Relaxation Distance and Neighborhood Search

Attention-Enhanced Actor–Critic Learning for Household Nonintrusive Load Monitoring

Attention-Enhanced Actor–Critic Learning for Household Nonintrusive Load Monitoring

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

Adversarial Data Anomaly Detection And Calibration for Non-Intrusive Load Monitoring

Adversarial Data Anomaly Detection And Calibration for Non-Intrusive Load Monitoring

Source-Free Domain Adaptation With Self-Supervised Learning for Nonintrusive Load Monitoring

Source-Free Domain Adaptation With Self-Supervised Learning for Nonintrusive Load Monitoring

Non-Intrusive Load Identification of Residential Appliances Using Improved Dictionary Learning Technique

With the advent of time, the demand for power in the residential sector is increasing. Along with supply-side management, the demand side is also used to balance electricity and supply demand. To apply different demand-side management techniques, the energy disaggregation on metered data is used to retrieve information related to available demand. Non-intrusive load monitoring is a technique that separates the total power consumption into appliance loads with minimum invasion of privacy. Non-intrusive load monitoring covers the methods of Stochastic finite state machines, Neural Networks and Sparse Coding. Developing an efficient algorithm for NILM is a key challenge in maximizing energy conservation. Recently, a new deep learning technique called dictionary learning has been developed for energy disaggregation. Smart meters provide the whole house data, and the Dictionary technique is trained to predict an appliance's power or ON/OFF based on its power consumption. This research proposes the event-based dictionary learning technique, which can disaggregate multiple appliances through orthogonal matching pursuit (OMP) and kernel-singular value decomposition (K-SVD). The sparse matrix is predicted through OMP, and K-SVD predicts the dictionary matrix. The training, testing and validation are done on the ECO dataset. The results of this research are noticeable and show the validity of the proposed methodology for energy disaggregation.

Energy consumption prediction of a smart home using non-intrusive appliance load monitoring

Energy consumption prediction of a smart home using non-intrusive appliance load monitoring

Research on a Non-Intrusive Load Recognition Algorithm Based on High-Frequency Signal Decomposition with Improved VI Trajectory and Background Color Coding

Against the backdrop of the current Chinese national carbon peak and carbon neutrality policies, higher requirements have been put forward for the construction and upgrading of smart grids. Non-intrusive Load Monitoring (NILM) technology is a key technology for advanced measurement systems at the end of the power grid. This technology obtains detailed power information about the load without the need for traditional hardware deployment. The key step to solve this problem is load decomposition and identification. This study first utilized the Long Short-Term Memory Denoising Autoencoder (LSTM-DAE) to decompose the mixed current signal of a household busbar and obtain the current signals of the multiple independent loads that constituted the mixed current. Then, the obtained independent current signals were combined with the voltage signals to generate multicycle colored Voltage–Current (VI) trajectories, which were color-coded according to the background. These color-coded VI trajectories formed a feature library. When the Convolutional Neural Network (CNN) was used for load recognition, in light of the influence of the hyperparameters on the recognition results, the Bayesian Optimization Algorithm (BOA) was used for optimization, and the optimized CNN network was employed for VI trajectory recognition. Finally, the proposed method was validated using the PLAID dataset. The experimental results show that the proposed method exhibited better performance in load decomposition and identification than current methods.

Distributionally robust deep network based on [formula omitted]-divergence for non-intrusive load monitoring

Non-Intrusive Load Monitoring (NILM) is a crucial technique for energy conservation as it provides detailed energy consumption data for individual household appliances, leading to targeted energy saving strategies. In practice, the startup phase of appliances is more significant than the shutdown phase, especially when predicting appliance-level consumption. When implementing Sequence-to-Point (S2P) based network models, where each appliance is trained separately, the primary source of training error comes from state transitions during startup, particularly those involving other appliances. These transitions can be mistakenly learned by the network as spurious correlations. This paper presents a method for effectively filtering such samples, and extracting features that are more resistant to spurious correlation confusion. We propose a Distributionally Robust Deep Network (DRDN) model grounded in ϕ-divergence, to address the NILM disaggregation problem. Our model incorporates an adaptive loss-scaled robust parameter to enable sample selection and improve decision-making. Our experimental results on the REDD and UK-DALE datasets demonstrate that DRDN models exhibit superior transferability and effectiveness in solving domain generalization problem compared to traditional deep learning approaches across different buildings.

Enhancing non-intrusive load monitoring with channel attention guided bi-directional temporal convolutional network for sequence-to-point learning

This study addresses the crucial aspect of identifying individual appliance power consumption without extensive sensor deployment, a cornerstone of modern smart grid planning and demand response. Non-Intrusive Load Monitoring (NILM) offers a solution by estimating appliance energy usage from aggregated meter data, increasingly powered by deep learning techniques. However, the depth-dependent effectiveness of load feature extraction can lead to gradient-related issues. To mitigate this, novel approach using a Bi-directional Temporal Convolutional Network (BiTCN) as a residual block foundation, enhanced by a Squeeze-and-Excitation Network (SENet) attention mechanism for channel-wise feature extraction. A departure from conventional methods involves substituting bidirectional non-causal convolution with channel attention for improved feature extraction was introduced. Evaluation on REDD and UK-DALE datasets demonstrates our model’s superiority in load disaggregation compared to existing approaches, emphasizing its potential in advancing NILM through effective deep learning strategies

Enhancing smart home appliance recognition with wavelet and scalogram analysis using data augmentation

The development of smart homes, equipped with devices connected to the Internet of Things (IoT), has opened up new possibilities to monitor and control energy consumption. In this context, non-intrusive load monitoring (NILM) techniques have emerged as a promising solution for the disaggregation of total energy consumption into the consumption of individual appliances. The classification of electrical appliances in a smart home remains a challenging task for machine learning algorithms. In the present study, we propose comparing and evaluating the performance of two different algorithms, namely Multi-Label K-Nearest Neighbors (MLkNN) and Convolutional Neural Networks (CNN), for NILM in two different scenarios: without and with data augmentation (DAUG). Our results show how the classification results can be better interpreted by generating a scalogram image from the power consumption signal data and processing it with CNNs. The results indicate that the CNN model with the proposed data augmentation performed significantly higher, obtaining a mean F1-score of 0.484 (an improvement of +0.234), better than the other methods. Additionally, after performing the Friedman statistical test, it indicates that it is significantly different from the other methods compared. Our proposed system can potentially reduce energy waste and promote more sustainable energy use in homes and buildings by providing personalized feedback and energy savings tips.