No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Load Disaggregation Based on Aided Linear Integer Programming
Abstract
Load disaggregation based on aided linear integer programming (ALIP) is proposed. We start with a conventional linear integer programming (IP) based disaggregation and enhance it in several ways. The enhancements include additional constraints, correction based on a state diagram, median filtering, and linear programming-based refinement. With the aid of these enhancements, the performance of IP-based disaggregation is significantly improved. The proposed ALIP system relies only on the instantaneous load samples instead of waveform signatures, and hence works well on low-frequency data. Experimental results show that the proposed ALIP system performs better than conventional IP-based load disaggregation.
Supplementary resource (1)
... Window-based optimization methods are often more accurate at the cost of increasing the computational complexity. In this category, integer programming (IP) solvers are a popular choice, and studies based on Aided Linear Integer Programming (AILP) [31,32], Mixed Integer Linear Programming (MILP) [33] and Mixed-Integer Nonlinear Programming [34,35] have been proposed in the literature. These approaches formulate NILM as a constrained optimization problem where the temporal linkage is preserved by imposing state transition constraints [34,35] or by a correction phase of the output of the IP solver [31]. ...
... In this category, integer programming (IP) solvers are a popular choice, and studies based on Aided Linear Integer Programming (AILP) [31,32], Mixed Integer Linear Programming (MILP) [33] and Mixed-Integer Nonlinear Programming [34,35] have been proposed in the literature. These approaches formulate NILM as a constrained optimization problem where the temporal linkage is preserved by imposing state transition constraints [34,35] or by a correction phase of the output of the IP solver [31]. The main practical limitation of these methods is their high computational complexity, especially if a resolution in the order of a few seconds is required, making its deployment difficult in an embedded system with low capabilities. ...
... The main practical limitation of these methods is their high computational complexity, especially if a resolution in the order of a few seconds is required, making its deployment difficult in an embedded system with low capabilities. Regarding unknown appliance's presence, a couple of papers [31,34] in this research group attempt to consider this issue by minimizing the number of active appliances in each time step. Although useful, this heuristic does not always hold [31], and a complement with an event detection phase could improve its effectiveness in the presence of unknown loads. ...
Real-time information on detailed power consumption can motivate users to make informed decisions to reduce their energy bills. In that sense, Non-Intrusive Load Monitoring (NILM) emerges as a cost-effective technique to achieve the previously mentioned benefits. This paper presents an online real-time robust NILM framework that only requires the aggregated active power, operates by updating the appliance’s state probabilities sequentially, and uses this information to predict the power consumption of each monitored appliance. The framework primarily focuses on the seamless integration and practical deployment of a real-time NILM algorithm, operating at frequencies around 1 Hz, on constrained edge devices. Starting with detecting edges and the base load in real-time, the appliance’s state probabilities are updated considering the possible presence of unknown loads. The power consumption of each appliance is then estimated by employing a modified Population-Based Incremental Learning algorithm (PBIL). Experiments on two publicly available datasets against state-of-the-art methods demonstrated its accuracy and robustness in the presence of unknown appliances. The real-time capabilities of the framework were verified through integration in a Home Automation framework running in a constrained edge device.
... Given the information on the number of devices (n) in the network, the operational modes, and the associated power consumption corresponding to each device, ED can be formulated as an optimization problem as a constrained/unconstrained single or multi-objective problem [16]. In the literature, most of the optimization-based ED algorithms [15,19,20] represent ED as a binary optimization problem where a device i with l i non-off modes is decomposed into l i virtual two-state (on/off (1/0)) devices. For appliance i, let P i = p 1 i , . . . ...
... In order to approximate (S(t)), the intuitive and the most commonly employed objective function in optimization-based ED is the least-square error between y(t) andŷ(t), as shown below [15,19,20]. ...
... To handle optimization-based energy disaggregation, as formulated in (3), integer programming [17], mixed integer programming [19], evolutionary algorithms [4,15,16,21], etc., have been employed. The search space associated with the binary optimization problem given by (3) increases drastically with the increase in the number of devices and their associated operational modes. ...
Recently, optimization-based energy disaggregation (ED) algorithms have been gaining significance due to their capability to perform disaggregation with minimal information compared to the pattern-based ED algorithms, which demand large amounts of data for training. However, the performances of optimization-based ED algorithms depend on the problem formulation that includes an objective function(s) and/or constraints. In the literature, ED has been formulated as a constrained single-objective problem or an unconstrained multi-objective problem considering disaggregation error, sparsity of state switching, on/off switching, etc. In this work, the ED problem is formulated as a constrained multi-objective problem (CMOP), where the constraints related to the operational characteristics of the devices are included. In addition, the formulated CMOP is solved using a constrained multi-objective evolutionary algorithm (CMOEA). The performance of the proposed formulation is compared with those of three high-performing ED formulations in the literature based on the appliance-level and overall indicators. The results show that the proposed formulation improves both appliance-level and overall ED results.
... Here, the objective of ED is to find an appropriate set of devices at each time instance (S(t)) to the best fit the measured aggregated signal (y(t)). Therefore, the least square error between the estimated and measured aggregate signals can be considered as shown below [22,25,26]. ...
... whereŷ(t) = S(t) P. A thorough enumeration of all possible combinations of S to obtain the optima to the binary optimization problem given by (3) is prohibitive due to the exponential increase in the number of possible combinations as the number of devices in the network becomes large. In literature, optimization frameworks based on integer programming [23], mixed integer programming [25], evolutionary algorithms [9,21,22,24], etc., have been explored to handle the ED problem formulation given by (3). ...
... In [25], Issue 1 was addressed by an inequality constraint that enforces the device to operate only in one of the l i states or completely switching OFF all the l i virtual devices. Issue 2 was handled with an equality constraint [25] to enforce that continuous operating devices are operating in at least one of the l i non-OFF states. ...
Energy disaggregation (ED), with minimal infrastructure, can create energy awareness and thus promote energy efficiency by providing appliance-level consumption information. However, ED is highly ill-posed and gets complicated with increase in number and type of devices, similarity between devices, measurement errors, etc. To design, test, and benchmark ED algorithms, the availability of open-access energy consumption datasets is crucial. Most datasets in the literature suit data-intensive pattern-based ED algorithms. Recently, optimization-based ED algorithms that only require information regarding the operational states of the devices are being developed. However, the lack of standard datasets and appropriate evaluation metrics is hindering the development of reproducible state-of-the-art optimization-based ED algorithms. Therefore, in this paper, we propose a dataset with multiple instances that are representative of the different challenges posed by ED in practice. Performance indicators to empirically evaluate different optimization-based ED algorithms are summarized. In addition, baseline simulation results of the state-of-the-art optimization-based ED algorithms are presented. The developed dataset, summarization of different metrics, and baseline results are expected to provide a platform for researchers to develop novel optimization-based frameworks, in general, and evolutionary computation-based frameworks in particular to solve ED.
... They concluded that it is hard to disaggregate loads with similar consumption patterns and proposed as future work a multi-objective optimization approach. Bhotto et al. [26] proposed several enhancements to the ILP model in [24], such as always-on appliances and median filtering. Most of the enhancements in [26] were included as a pre-processing step rather than constraints. ...
... Bhotto et al. [26] proposed several enhancements to the ILP model in [24], such as always-on appliances and median filtering. Most of the enhancements in [26] were included as a pre-processing step rather than constraints. They also proposed a linear programming refinement to deal with the minimum and maximum transient spans of the power states. ...
... In practice, the power states of household appliances fluctuate within a range, so that approximating them with a straight line may affect the quality of the disaggregation. As observed in [26], being able to capture appliances' dynamics in addition to the steady-state ratings could be effective to increase the disaggregation accuracy. However, incorporating these dynamics as additional states in the optimization problem would dramatically increase the computational cost. ...
Energy disaggregation, known in the literature as Non-Intrusive Load Monitoring (NILM), is the task of inferring the energy consumption of each appliance given the aggregate signal recorded by a single smart meter. In this paper, we propose a novel two-stage optimization-based approach for energy disaggregation. In the first phase, a small training set consisting of disaggregated power profiles is used to estimate the parameters and the power states by solving a mixed integer programming problem. Once the model parameters are estimated, the energy disaggregation problem is formulated as a constrained binary quadratic optimization problem. We incorporate penalty terms that exploit prior knowledge on how the disaggregated traces are generated, and appliance-specific constraints characterizing the signature of different types of appliances operating simultaneously. Our approach is compared with existing optimization-based algorithms both on a synthetic dataset and on three real-world datasets. The proposed formulation is computationally efficient, able to disambiguate loads with similar consumption patterns, and successfully reconstruct the signatures of known appliances despite the presence of unmetered devices, thus overcoming the main drawbacks of the optimization-based methods available in the literature.
... Hidden Markov models (HMM) and combinatorial optimization (CO) a commonly used methods for state-based approaches. Integer programming (IP) can a be used to disaggregate without supervised learning [7,8]. ...
... Hidden Markov models (HMM) and combinatorial optimization (CO) are commonly used methods for state-based approaches. Integer programming (IP) can also be used to disaggregate without supervised learning [7,8]. ...
Nonintrusive load monitoring (NILM) is a process that disaggregates individual energy consumption based on the total energy consumption. In this study, an energy disaggregation model was developed and verified using an algorithm based on a recurrent neural network (RNN). It also aimed to evaluate the utility of the occupant location information, which is nonelectrical information. This study developed energy disaggregation models with RNN-based long short-term memory (LSTM) and gated recurrent unit (GRU). The performance of the suggested models was evaluated with a conventional method that uses the factorial hidden Markov model. As a result, when developing the GRU disaggregation model based on an RNN, the energy disaggregation performance improved in accuracy, F1-score, mean absolute error (MAE), and root mean square error (RMSE). In addition, when the location information of the occupants was used, the suggested model showed improved performance and good agreement with the real power and electricity consumption by each appliance.
... Loads can generally be divided into those that have approximately discrete steady-state levels during operation and those with variable power demand [3]. The identification of loads with discrete steady-state levels and the tracking of resulting power changes has been well-documented in literature [4]- [10]. These loads are often identified with either event-based or optimization-based techniques. ...
... Many deep learning techniques have been applied for eventbased load identification, such as convolutional neural net-works (CNNs) [11]- [13], long short-term memory (LSTM) [14], and gated recurrent units (GRUs) [15]. Optimization methods, such as hidden Markov models (HMMs) [8], [9] and mixed-integer linear programming [5], [10], attempt to find the set of energized loads that best fit the aggregate measurement. By assuming a constant power level at each load state, the energy consumption of individual loads can be computed by tracking the operating duration. ...
Nonintrusive identification of the energy consumption of individual loads from an aggregate power stream typically relies on relatively well-defined transient signatures. However, some loads have non-constant power demand that varies with loading conditions. These loads, such as computer-controlled machine tools, remain stubbornly resistant to conventional nonintrusive electrical monitoring methods. The power behavior of these loads can be modelled with stochastic processes. This paper presents statistical feature extraction techniques for identification of this fluctuating power behavior. An energy estimation procedure is presented and evaluated for two case studies: load operation on a shipboard microgrid and laboratory machine shop equipment.
... Smart meters, which are extensively employed, are generally suitable at sampling rates with large intervals [15], due to the limitations of the manufacturing cost, storage cost, and data transmission capacity. The use of low-frequency sampling signals obtained by smart meters for data analysis and mining has become a new research hotspot [16], [17]. ...
Non-intrusive load monitoring is a technique for monitoring the operating conditions of electrical appliances by collecting the aggregated electrical information at the household power inlet. Despite several studies on the mining of unique load characteristics, few studies have extensively considered the high computational burden and sample training. Based on low-frequency sampling data, a non-intrusive load monitoring algorithm utilizing the graph total variation (GTV) is proposed in this study. The algorithm can effectively depict the load state without the need for prior training. First, the combined
K
-means clustering algorithm and graph signals are used to build concise and accurate graph structures as load models. The GTV representing the internal structure of the graph signal is introduced as the optimization model and solved using the augmented Lagrangian iterative algorithm. The introduction of the difference operator decreases the computing cost and addresses the inaccurate reconstruction of the graph signal. With low-frequency sampling data, the algorithm only requires a little prior data and no training, thereby reducing the computing cost. Experiments conducted using the reference energy disaggregation dataset and almanac of minutely power dataset demonstrated the stable superiority of the algorithm and its low computational burden.
... Kolter et al. [97] developed a method for load disaggregation using discriminative sparse coding. Bhotto et al. [98] proposed an aided linear integer programming load decomposition approach for low-frequency data. Miyasawa et al. [99] employed matrix factorization for low-frequency power data disaggregation. ...
The advanced metering infrastructure (AMI) plays a crucial role in the smart
grid (SG) by collecting vast amounts of high-frequency power consumption data
from customers through smart meters. Smart meter data analytics extract valuable information and insights from this data, contributing to the improvement
of energy management and distribution within the SG. In this dissertation, we
primarily focus on developing novel deep learning-based methods for three key
applications in smart meter data analytics.
Firstly, we propose ABODE-Net, an attention-based deep learning model
that enables end-to-end non-intrusive building occupancy detection using raw
smart meter data and corresponding time information. To enhance its effectiveness in building occupancy detection, we introduce a novel lightweight Parallel
Attention (PA) block as a key component of ABODE-Net, allowing for the efficient capture of discriminating information. Experimental results demonstrate
the viability of ABODE-Net as a non-intrusive solution for building occupancy
detection using smart meter data.
Secondly, we develop a neural architecture search (NAS)-based method to
automatically discover high-performing deep neural network (DNN) architectures for data-driven customer socio-demographic information identification.
Our proposed method employs a search space based on a novel channel
attention-based fully convolutional network (CAFCN) architecture. This architecture effectively captures both temporal and spatial patterns in smart
meter data, enabling accurate identification of customer socio-demographic
information. We develop a Bayesian Optimization (BO)-based search strategy to
effectively explore the search space and discover high-performing architectures
aligned with the search objective. Performance evaluation using a widely-used
smart meter dataset demonstrates the superior performance of our proposed
method compared to state-of-the-art baselines in data-driven socio-demographic
information identification.
Lastly, we propose MOGSM-Net, a multi-objective genetically searched
multi-label deep model for non-intrusive load monitoring (NILM). MOGSM-Net
is designed using the NAS technique, employing a search space based on a hybrid architecture that leverages the strengths of convolutional neural networks
(CNNs), long short-term memory (LSTM) networks, and attention mechanisms
to effectively capture spatial and temporal information in smart meter data. To
enhance the model’s ability to handle the inherent highly imbalanced data in the
NILM problem, we design a multi-label classification head for MOGSM-Net, optimized with an Asymmetric loss function. We employ a multi-objective genetic
search strategy that considers both performance and computational complexityobjectives to effectively explore the search space. Experimental results on two
popular NILM datasets demonstrate the superiority of MOGSM-Net over all
baselines in terms of both performance and computational complexity.
... The UK-DALE (UK Domestic Appliance-Level Electricity) dataset is the second most popular, used three times alone and 11 times with other datasets. The AMPds (Almanac of Minutely Power dataset) ranks third, appearing in seven articles [31][32][33][34]. The least popular datasets-LIFTED, GreenD [35], Eco [35], Low Carbon London (LCL) [36], and Rainforest Automation Energy Dataset (RAE) [37]-each appeared only once in the reviewed articles. ...
Energy load disaggregation can contribute to balancing power grids by enhancing the effectiveness of demand-side management and promoting electricity-saving behavior through increased consumer awareness. However, the field currently lacks a comprehensive overview. To address this gap, this paper conducts a scoping review of load disaggregation domains, data types, and methods, by assessing 72 full-text journal articles. The findings reveal that domestic electricity consumption is the most researched area, while others, such as industrial load disaggregation, are rarely discussed. The majority of research uses relatively low-frequency data, sampled between 1 and 60 s. A wide variety of methods are used, and artificial neural networks are the most common, followed by optimization strategies, Hidden Markov Models, and Graph Signal Processing approaches.
... Load signatures are the main basis for load identification, which are divided into steady-state signatures and transient signatures [3][4][5][6][7]. Steady-state signatures are simpler to measure and process, and easy to extract. ...
Non-intrusive load monitoring is a novel and cost-effective technology for monitoring details of electricity consumption and identifying the operating status of appliances. It supports the construction of the energy internet and big data on electricity consumption in smart cities. However, one of the most challenging problems in this area is that machine learning algorithms often require large amounts of labeled data. In this paper, a non-intrusive load monitoring model based on the Self-supervised Regularization is proposed. The model reduces the pre-processing stage compared to the traditional methods. We make full use of the unlabeled data by using them to generate proxy labels to participate in the model training together with the true labels. We performed experiments on the common data set PLAID to compare performance with the existing method Mean Teacher and CoMatch. The experimental results show that: (1) when using all labeled data, the model with self-supervised regularization significantly improves the traditional supervised classifier with a recognition accuracy of 0.965. (2) when coupled with unlabeled data, our model produces good semi-supervised performance. It is highly competitive with current state-of-the-art Mean Teacher and Contrastive Learning.
... However, concerns with transient signals and circumstances where certain states are binary combinations of other states have been demonstrated to cause problems with IP-based disaggregation on actual data. To overcome this problem, Bhotto et al. [132] presented an ALIP disaggregation method that addresses many of the drawbacks of the IP-based approach. The experimental findings show that ALIP has a considerable accuracy improvement over the IP-based disaggregation approach. ...
The implementation of non-intrusive load monitoring has gained significant attention as a promising solution for disaggregating and identifying individual appliances' energy consumption within households and commercial buildings. The issue at the core of non-intrusive load monitoring is the accurate separation of the aggregated energy consumption signal into the energy contributions of distinct appliances without requiring additional sensors or invasive measurements. The erroneous implementation of non-intrusive load monitoring model leads to poor results and a waste of time where the criteria for the implementation are necessary. This paper presents an overview of non-intrusive load monitoring's implementation, addressing the stages of non-intrusive load monitoring and advancements in the field. The overview explores the various techniques, methodologies, and algorithms employed in non-intrusive load monitoring systems to achieve load identification and energy disaggregation. By addressing the problem statement and reviewing the state of the art, this paper sets the stage for understanding the significance of non-intrusive load monitoring in commercial buildings and modern energy-efficient households. Moreover, the limitations, applications, and recommended criteria for the implementation of non-intrusive load monitoring are highlighted as a contribution to the field of load disaggregation. Future research should address scalability, privacy concerns, and the integration of non-intrusive load monitoring with smart grids to propel energy disaggregation towards more efficient, accessible, and sustainable energy management systems. We envision that this endeavour can serve as a guiding beacon, offering a roadmap for researchers and professionals.
... Variants of Hidden Markov Models (HMMs) are the most famous former NILM method that can provide an efficient representation of appliances [6]. Linear and non-linear mixed integer optimization are other popular means of load identification in the literature [25,26]. Recently, DL, mostly Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNN), has drawn significant attention for performing NILM [27]. ...
For many years, energy monitoring at the most disaggregate level has been mainly sought through the idea of Non-Intrusive Load Monitoring (NILM). Developing a practical application of this concept in the residential sector can be impeded by the technical characteristics of case studies. Accordingly, several databases, mainly from Europe and the US, have been publicly released to enable basic research to address NILM issues raised by their challenging features. Nevertheless, the resultant enhancements are limited to the properties of these datasets. Such a restriction has caused NILM studies to overlook residential scenarios related to geographically-specific regions and existent practices to face unexplored situations. This paper presents applied research on NILM in Quebec residences to reveal its barriers to feasible implementations. It commences with a concise discussion about a successful NILM idea to highlight its essential requirements. Afterward, it provides a comparative statistical analysis to represent the specificity of the case study by exploiting real data. Subsequently, this study proposes a combinatory approach to load identification that utilizes the promise of sub-meter smart technologies and integrates the intrusive aspect of load monitoring with the non-intrusive one to alleviate NILM difficulties in Quebec residences. A load disaggregation technique is suggested to manifest these complications based on supervised and unsupervised machine learning designs. The former is aimed at extracting overall heating demand from the aggregate one while the latter is designed for disaggregating the residual load. The results demonstrate that geographically-dependent cases create electricity consumption scenarios that can deteriorate the performance of existing NILM methods. From a realistic standpoint, this research elaborates on critical remarks to realize viable NILM systems, particularly in Quebec houses.
... To solve this problem, integer programming (IP) approaches can be utilized. Bhotto et al. propose a load disaggregation approach based on aided linear IP in [14], with improvement on feature extraction, constraint selection, pre-and post-processing. However, when disaggregating a large number of appliances, global optimization becomes hard. ...
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.
... The authors of a previous paper [14] defined the NILM problem as a sparse optimization problem and solved it using a dictionary learning approach. In previous studies [15,16], the scholars solved the NILM problem by an integer programming-based approach. ...
Event detection is the foundation of event-based non-intrusive load detection solutions. Conventional event detection methods require a comprehensive consideration of the rated power levels of all devices within the detection scenario to set an appropriate threshold value. However, it cannot accurately detect both high- and low-power load events because of their fixed thresholds when loads with widely varying power change amplitudes are present simultaneously. Thus, in this study, an adaptive threshold event detection method based on standard deviation is proposed. First, the aggregated power data is intercepted by a sliding window for a short period of time, and the standard deviation is calculated for the aggregated power data within the window, and the event ends when the standard deviation reaches its maximum value. Then calculate the threshold for event detection based on the standard deviation, and then perform event detection based on the calculated threshold and then based on the bilateral sliding window cumulative sum (CUSUM) method. Finally, various load tests are performed with Electricity Consumption & Occupancy (ECO) data sets and private data sets , and the F1 values exceeded 90% in all three scenes: office, factory and laboratory, indicating that the proposed method in this study has high event detection performance.
... To escape the need for bulk training data and advanced computing machines, many researchers have employed optimization algorithms. Some optimization-based NILM approaches include integer linear programming [34], integer non-linear programming (INLP) [11], mixed ILNP [35], quadratic programming [36], sparse optimization [37], etc. Egarter and Elmenreich [38] solved NILM using six metaheuristic ealgorithms and found it difficult to identify appliances with overlapping features. This problem can be tackled either by using additional information about appliance usage or by expanding the feature set. ...
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.
... This consent binds the experiment organizers to strictly use the data of subject within the limits of experiment. The data includes response to questionnaire, measured load curve, appliance commitments and contact details. 21 General data protection regulation (EU) : https://gdpr.eu/what-is-gdpr/ ...
The energy transition towards renewable energy sources and distributed generation is essential to mitigate climate change. However, the renewable energy sources are intermittent in nature, and therefore it may cause network congestion during peak consumption in future. Therefore, demand side energy flexibility is inevitable in complement to the supply side management and energy storage. The residential sector is the largest energy consumer and is therefore essential to implement demand-side energy flexibility. In addition to the existing approaches of energy flexibility, we present in this thesis a non-monetary “indirect" flexibility through the implementation of a behavioral science experiment with 175 French households. The set of nudges used in this experiment does not oblige the participating households to implement energy flexibility for either monetary gain or loss aversion. Nudge alerts are generated by predictive algorithms and sent by SMS to a group of households with the aim of carrying out either load shifting or load shedding. The nudge alerts are complemented with the suggestions about using (or not using) committed appliances to implement energy flexibility. After each alert, the group receives graphical feedback in which the load curve measured by LINKY smart meter is superposed by a theoretical reference curve specific to each household. In comparison to a control group, we can quantify the impact of these nudge signals on the energy consumption of the treated group. Finally, an energy disaggregation algorithm is presented that will allow us to detect the equipment used during the flexibility and thus will do an analysis of the commitment of households to look for potential appliances for energy flexibility.
... To this purpose, energy research community explored different automated analysis tools, such as Aided Integer Linear Programming [13] and Factorial Hidden Markov model [14], to assess electrical consumption of individual appliances in residential field. Yan et al. [15] exploited a model driven simplified approach, considering energy balance of buildings and linear regression models, tuned accordingly to an optimization algorithm. ...
As the telecommunication industry becomes more and more energy intensive, energy efficiency actions are crucial and urgent measures to achieve energy savings. The main contribution to the energy demand of buildings devoted to the operation of the telecommunication network is cooling. The main issue in order to assess the impact of cooling equipment energy consumption to support energy managers with awareness over the buildings energy outlook is the lack of monitoring devices providing disaggregated load measurements. This work proposes a Non-Intrusive Load Disaggregation (NILD) tool that exploits a literature-based decomposition with an innovative LSTM Neural Network-based decomposition algorithm to assess cooling demand. The proposed methodology has been employed to analyze a real-case dataset containing aggregated load profiles from around sixty telecommunication buildings, resulting in accurate, compliant, and meaningful outcomes.
... Ref. [17] described a spectrum-smoothing-based load disaggregation strategy for dealing well with many appliances turning on and off at the same time. There have also been proposals for NILM algorithms based on integer programming [18] and mixed-integer linear programming [19]. ...
From a single meter that measures the entire home’s electrical demand, energy disaggregation calculates appliance-by-appliance electricity consumption. Non-intrusive load monitoring (NILM), also known as energy disaggregation, tries to decompose aggregated energy consumption data and estimate each appliance’s contribution. Recently, methodologies based on Artificial Intelligence (AI) have been proposed commonly used in these models, which can be expensive to run on a server or prohibitive when the target device has limited capabilities. AI-based models are typically computationally expensive and require a lot of storage. It is not easy to reduce the computing cost and size of a neural network without sacrificing performance. This study proposed an efficient non-parametric supervised machine learning network (ENSML) architecture with a smaller size, and a quick inference time without sacrificing performance. The proposed architecture can maximise energy disaggregation performance and predict new observations based on past ones. The results showed that employing the ENSML model considerably increased the accuracy of energy prediction in 99 percent of cases.
... Optimization-based techniques for NILM require less prior information as compared to MLbased models [15]. Some optimization-based NILM formulations are quadratic programming [37], sparse optimization [38], integer programming [39], mixedinteger linear programming [40], integer nonlinear programming [12], etc. D. Egarter and W. Elmenreich applied six metaheuristic techniques to solve NILM and found it difficult to discern appliances with overlapping features [41]. This issue can be resolved by increasing the number of features and including the appliance usage information. ...
... The proposed method has also shown its applicability, while the same type of devices is operating simultaneously. Load disaggregation based on Aided Linear Integer Programming (ALIP) has recently been proposed [102]. In addition to the LIP-based disaggregation approach, additional constraints, state diagram-based correction, and median filtering have been added. ...
... NILM problem is formulated as a mixed-integer programming (MIP) in Piga et al. (2016), and the results are compared with a learning-based one. Authors in Bhotto et al. (2017) modified the formulation of IP by defining novel boundaries for the appliances. To reduce the computation burden in MIP-based NILM, a window-based algorithm is utilized in Wittmann et al. (2018). ...
The power network is becoming increasingly intermittent as the contribution from renewable energy generation rises. To maintain stability and functionality of the power network, storage of renewable energies and demand-side control techniques are required. Smart grids provide the communication infrastructure to accomplish this goal. Smart grid control originated from the idea that the demand-side of the power grid can shift or shed load to reduce the strain on the network, while also maintaining consumer satisfaction and other specialist requirements.
... NILM problem is formulated as a mixed-integer programming (MIP) in Piga et al. (2016), and the results are compared with a learning-based one. Authors in Bhotto et al. (2017) modified the formulation of IP by defining novel boundaries for the appliances. To reduce the computation burden in MIP-based NILM, a window-based algorithm is utilized in Wittmann et al. (2018). ...
Smart grid control and management is a growing research area that affects the global goals of net-zero emissions, increased renewable energy generation, and efficient energy management. Active buildings are the building blocks of the smart grid of the future. Their passive role as energy consumer buildings is replaced in the new paradigm as active components, which are not only the energy consumers but also provide energy services to the net- work or neighbouring areas in time of need. This chapter puts a spotlight on the control of a community of active buildings and considers high-level smart grid control approaches that use active buildings; aggregation, frequency and voltage regulation as well as security considerations. The chapter also provides a detailed case study that considers demand-side frequency regulation, power tracking and formal control synthesis methods using energy storage systems (ESSs) and thermostatically controlled loads (TCLs).
... where Pre is defined as the precision classification in all positive estimations; Re is expressed as the percentage of correctly recognized appliances. In addition, the method mentioned in [20] is also used to evaluate the accuracy of power decomposition. For T samples, the total power decomposition accuracy is defined as PD acc ; the power decomposition accuracy for each load is defined as PD (m)acc : ...
Appliance-level data is a prerequisite for establishing friendly two-way interactions between customers and the power company, and this data is now mainly obtained by non-intrusive load monitoring. However, as the number of loads increases, the number of possible appliances state combinations tends to grow exponentially, leading to a significant increase in the time of load identification. In order to reduce the search range of the load state combinations and shorten the algorithm response time, a non-intrusive load monitoring method based on the time-segmented state probability is proposed in this paper. Firstly, the affinity propagation (AP) clustering algorithm is introduced to obtain the power templates of the load, and then the power templates are used to count the time-segmented state probabilities. Secondly, a number of appliance state matrices are generated using the probabilities, and the optimal matrix is selected by the function as the identification result of the appliance state. Finally, the performance of the algorithm is tested on the public NILM dataset and compared to several state-of-the-art techniques. The results illustrate that the proposed method achieves an accuracy of 96% for load state identification and 89% for power decomposition of the load, and is able to meet the real-time application requirements.
... R. Machlev ve arkadaşları Modifiye Cross-Entropy yöntemi ismini verdikleri bir yöntemle aktif gücün düşük oranlı örneklemesini kullanarak gerçek zamanlı karar vermeye dayalı bir algoritma sunmuşlardır (Machlev et al., 2018). M. Z. A. Bhotto ve arkadaşları doğrusal destekli tamsayı programlamaya dayalı yük ayrıştırma önerisi ile IP tabanlı ayrıştırmanın performansı önemli ölçüde arttırdığını önermiştir (Bhotto et al., 2016). F. M. Wittmann ve arkadaşları, doğrusal programlamaya dayalı yaptığı müdahaleye gerek olmayan yük ayrıştırma üzerine yaptığı çalışmada her bir cihazın tüketimlerini toplam güç ölçümünden ayırırken ortaya çıkan çoklu anahtarlama sorunuyla ilgilenmiştir (Wittmann et al., 2018). ...
Non-intrusive load monitoring (NILM) is pivotal for extracting individual appliance power consumption from aggregated signals, enabling cost-effective energy management. While deep learning has gained prominence in NILM, existing approaches often assume uniform distribution in input (mains) and output spaces (appliances), leading to suboptimal performance across datasets. In this paper, we introduce DistribuNet, a novel neural paradigm designed to alleviate distributional shifts in NILM. At its core, Dual-CoefNet captures distinct distributions in both input and output spaces, effectively addressing challenges associated with intra-space and inter-space shifts. Additionally, the integration of the Hierarchical Attention Fusion Network (HAFN) further enhances DistribuNet’s robustness, enabling adaptation to variations in data distribution. We also introduce knowledge-induced training strategy to improve the efficiency and effectiveness of the training process. Comprehensive evaluations across diverse datasets demonstrate DistribuNet’s superior performance, particularly in handling different datasets with varied temporal patterns. The results showcase its potential for real-world applications in NILM.
Esta obra traz os conceitos e aplicações básicas que possibilitam a transição das redes de energia elétricas convencionais para redes de energia elétricas inteligentes. Apesar de não ser um elemento que integra uma smart grid, os conceitos básicos sobre geração distribuída e suas especificações no Brasil são apresentados. Não menos importante, uma discussão sobre as políticas públicas necessárias para a facilitação desse processo de transição é feita. Expõe as áreas de maior interesse para essa transição nos setores de geração, transmissão e distribuição de energia elétrica, incluindo a importância da segurança cibernética nestas. Sistemas de armazenamento de energia, suas aplicações ao redor do mundo, os conceitos básicos envolvidos nas principais tecnologias utilizadas, uma introdução sobre baterias de segunda vida e os custos envolvidos, são temas presentes na obra. As possibilidades de aplicação de ferramentas de inteligência artificial nas smart grids são exemplificadas por meio de resultados de pesquisas onde foram utilizadas redes neurais, redes neurais profundas e algoritmos de otimização. Por fim, cinco códigos de algoritmos de otimização são disponibilizados para que o leitor possa ter um melhor entendimento de como os procedimentos de cada um podem ser traduzidos para uma linguagem de programação, em específica a do software MatLab®.
Non-intrusive load monitoring (NILM) technology aims to infer the operation information of electrical appliances from the total household load signals, which is of great significance for energy conservation and planning. However, existing methods are difficult to effectively capture the complex nonlinear features of the power consumption flow, which affects the energy disaggregation accuracy. To this end, this paper designs a method based on temporal convolutional network (TCN), efficient channel attention (ECA), and long short-term memory (LSTM). The method first creatively proposes a two-stage improved TCN (TSTCN), which overcomes its problems of extracting discontinuous information and poor correlation of long-distance information while enhancing the ability to extract high-level load features. Then a novel improved ECA attention mechanism (IECA) is embedded, which is also combined with the skip connection technique to pay channel-weighted attention to important feature maps and promote information fusion. Finally, the LSTM with strong temporal memory capability is introduced to learn the dependencies in the load power sequence and realize load disaggregation. Experiments on two real-world datasets, REDD and UK-DALE, show that the proposed model significantly outperforms other comparative NILM algorithms and achieves satisfactory tracking with the actual appliance operating power. The results show that the mean absolute error (MAE) of all appliances decreases by 18.67% on average, and the F1 score improves by 38.70%.
Non-intrusive load monitoring (NILM), a sophisticated load monitoring technology, has garnered considerable interest for its potential to assist consumers in lowering their energy expenditures. In this paper, we present a continuous non-convex optimization model for NILM that employs the norm-box constraint to convert the discrete integer variables in the model into continuous ones. Subsequently, we apply the alternating direction method of multipliers (ADMM) algorithm to tackle the non-convex problem. To enhance the sluggish convergence of the ADMM algorithm, we introduce a linear penalty term based on relaxation distance (RD) to supplant the conventional quadratic penalty term. Furthermore, we devise a heuristic refinement method based on neighborhood search (NS) to augment the solution quality of our algorithm. Simultaneously, by utilizing a dynamic window partitioning technique, the NILM task can be split into multiple small subtasks. These subtasks can be allocated to multiple consumer electronics with computing capabilities to achieve distributed computing. Ultimately, we validate our proposed algorithm on the AMPds dataset, and the experimental results demonstrate that it has faster convergence and yields better solutions compared to a state-of-the-art solver and traditional ADMM algorithms. Using our algorithm, the NILM system can offer consumers efficient, convenient, and economical services.
The rising demand for energy conservation in residential buildings has increased interest in load monitoring techniques by exploiting energy consumption data. In recent years, hundreds of research articles have been published that have mainly focused on data-driven, non-intrusive load monitoring (NILM) approaches. Due to the high volume of research articles published in this domain, it has become necessary to provide a review of the up-to-date research in NILM and highlight the current challenges associated with its application. This paper reviews the state-of-the-art of NILM by following a structured assessment process to consider relevant and most recent documents in the literature. It presents the pros and cons of data-driven NILM methods, available datasets, and performance evaluation mechanisms. Even though research in NILM solutions has matured in recent years thanks to the use of deep learning models, there are still gaps in their effective deployment related to data requirements, real-time performance, and interpretability. Therefore, the paper also addresses the NILM development and implementation challenges and includes promising improvement measures that can be utilized to solve them.
For the nonintrusive load monitoring problem, we propose a new efficient mixed-integer linear programming model. Compared to other optimization-based models that are only capable of distinguishing between the
on/off
states of appliances, our model takes it a step further by incorporating continuous power variables and constructing power feature constraints, allowing for a more accurate fitting of the power consumption of each appliance. To improve the computation efficiency of our model, we present new state constraints, new linear penalty terms, new state transition constraints, and new minimum active time constraints. To extract features from load data containing noise efficiently, we propose automatic feature extraction algorithms based on distributionally robust optimization theory and linear regression. These algorithms can extract power boundary features, power fluctuation features, and minimum active time features. Our proposed method and six state-of-the-art optimization-based methods are tested on the almanac of minutely power dataset (AMPds) and REFIT to verify the performance of the proposed method. The results show that our method outperforms other methods in terms of disaggregation accuracy and computational efficiency. Moreover, our load feature extraction algorithms can effectively reduce the noise interference in the data. Our modeling approach can improve the computational efficiency of other optimization-based methods.
The big data era has raised public concern regarding private information leakage. Therefore, in the electricity sector, many classical privacy preserving mechanisms based on noise injection have been designed and implemented for meter data. However, injected noise of large magnitudes can affect the statistical structure of these data. Therefore, in this study, we identify the inherent randomness embedded in time series data to mitigate this issue. To this end, we study the potential of using this inherent randomness to protect the privacy for both high and low resolution time series data. We propose a privacy preserving mechanism using stochastic differential equation modeling. We theoretically prove the effectiveness of our proposed framework and design several methods to implement our mechanism to aid various data-driven consumer behavior analysis tasks in the electricity sector. The numerical results indicate that our framework can simultaneously maintain the desired level of privacy preservation and value of data in practice.
Non‐intrusive load monitoring (NILM) is an elegant solution for monitoring energy consumption. However, the main challenge of NILM is to accurately analyze the aggregate load data and determine the electrical consumption of each appliance. Recently, there have been some deep learning (DL) techniques proposed for NILM. Generally, these DL methods can outperform some of the existing NILM models. However, some of these DL methods cannot handle well on multi‐state appliances, appliances with sparse patterns, and appliances with rapid changing patterns. This chapter presents a new energy disaggregation method based on a parallel bidirectional long short‐term memory model. Moreover, a feature extractor is proposed to unmask useful statistical features from the aggregate signals to improve the learning capability of the network. The benchmark dataset REDD was used for testing the proposed method and the state of the arts. The results indicate that the proposed method can successfully outperform those methods.
Appliance energy consumption tracking in a building is one of the vital enablers of energy and cost saving. An economical and viable solution would be to estimate individual appliance consumption from a single-point measurement without using dedicated appliance-wise sensors or sub-meters. Such solutions are called non-intrusive load monitoring or energy disaggregation. Though the technology was introduced around three decades ago, the recent deployments of smart meters and machine learning-based algorithms attract the researcher to develop many novel solutions to the technology. To this end, this paper aims to provide a comprehensive, unbiased, and systematic state-of-the-art review of the energy disaggregation technology by evaluating two research questions. These are, first, how did the research community approach to solve the NILM problem and how has the technology evolved, and second, how do various NILM solutions stand compared to others in terms of a few key attributes. Further, some key challenges and future research directions of the technology are highlighted.
The rapid development of the smart energy system promotes bidirectional communications between the supply-side and demand-side. End users can handily receive real-time prices and adjust their electric energy consumption behaviors. Acquiring the time-varying price elasticity of demand (PED) of electricity can help utility companies to understand the time-varying electricity consumption behavior affected by price, thereby facilitating demand-side management. However, estimating time-varying PED is rarely considered in existing studies. This paper bridges the gap, proposing a time-varying PED estimation algorithm. To analyze PEDs more precisely, the proposed algorithm is applied to each appliance based on the advanced non-intrusive load monitoring (NILM) technology. Moreover, a demand-side smart dynamic pricing mechanism is also proposed to provide decision support of individually optimal dynamic pricing for utility companies to encourage end users to participate in the demand response (DR) program. Comprehensive experiments have been conducted to validate the practical feasibility of the proposed mechanism. Numerical simulations show that the proposed mechanism can facilitate the DR program by reducing the peak-to-average ratio (PAR) in electricity consumption without suffering from the price risk.
Non-intrusive load monitoring (NILM) is an elegant solution for monitoring energy consumption. Essentially, it only requires a set of voltage and current sensors to be installed at the electrical entry point for load disaggregation. However, the main challenge of NILM is to accurately analyze the aggregate load data and determine the electrical consumption of each appliance. Recently, there have been some deep learning (DL) techniques proposed for NILM. These include deep convolutional neural networks (DCNNs), gated linear unit and residual network (GLU-Res), bidirectional long short-term memory (BLSTM), and autoencoder (AE). Generally, they can outperform some of the existing NILM models such as factorial hidden Markov model. Nevertheless, some of these DL methods cannot handle well on multi-state appliances, appliances with sparse patterns, and appliances with rapid changing patterns. This article proposes a new NILM model, which involves parallel convolution neural networks and BLSTM layers. Moreover, a feature extractor is proposed to unmask useful statistical features from aggregate signals to improve the learning capability of the network. The benchmark dataset REDD was used for testing the proposed method and the state-of-the-arts such as DCNN, GLU-Res, BLSTM, and AE. The results indicate that the proposed method can successfully outperform those methods.
Energy management has become the main focus of researchers to propose efficient strategies to use available resources and reduce CO2 emissions, thus preserving the environment and creating smarter cities. One of the most efficient energy management tools in this sector is non-intrusive load monitoring (NILM). NILM aims to extract the power consumption of each appliance from the given total consumption through purely analytical methods. This information about the appliances can be sent as feedback to the consumers, which increases their knowledge about their consumption behavior and helps them make the right decisions to reduce their consumption and cost while maintaining their comfort. In this chapter, we present the basic concepts about NILM, various algorithms that have been explored to develop more practical and accurate NILM techniques and their challenges. Finally, we illustrate the application of NILM in determining the energy flexibility potential of each consumer.KeywordsDemand-side managementEnergy managementEnergy flexibilityResidential sectorLoad monitoringIntrusive load monitoringSemi-intrusive load monitoringNon-intrusive load monitoringOptimizationSupervised methodsUnsupervised methodsClassificationClusteringSample-based methodsEvent-based methodsAMPds dataset
The knowledge of the users' electricity consumption pattern is an important coordinating mechanism between the utility company and the electricity consumers in terms of key decision makings. The load decomposition is therefore crucial to reveal the underlying relationship between the load consumption and its characteristics. However, load decomposition is conventionally performed on the residential and commercial loads, and adequate consideration has not been given to the high-energy-consuming industrial loads leading to inefficient results. This paper thus focuses on the load decomposition of the industrial park loads (IPL). The commonly used parameters in a conventional method are however inapplicable in high-energy-consuming industrial loads. Therefore, a more robust approach is developed comprising a three-algorithm model to achieve this goal on the IPL. First, the improved variational mode decomposition (IVMD) algorithm is introduced to denoise the training data of the IPL and improve its stability. Secondly, the convolutional neural network (CNN) and simple recurrent units (SRU) joint algorithms are used to achieve a non-intrusive and non-invasive decomposition process of the IPL using a double-layer deep learning network based on the IPL characteristics. Specifically, CNN is used to extract the IPL data characteristics while the improved long and short-term memory (LSTM) network, SRU, is adopted to develop the decomposition model and further train the load data. Through the robust decomposition process, the underlying relationship in the load consumption is extracted. The results obtained from the numerical examples show that this approach outperforms the state-of-the-art in the conventional decomposition process.
The knowledge of the users’ electricity consumption pattern is an important coordinating mechanism between the utility company and the electricity consumers in terms of key decision makings. The load decomposition is therefore crucial to reveal the underlying relationship between the load consumption and its characteristics. However, load decomposition is conventionally performed on the residential and commercial loads, and adequate consideration has not been given to the high-energy-consuming industrial loads leading to inefficient results. This paper thus focuses on the load decomposition of the industrial park loads (IPL). The commonly used parameters in a conventional method are however inapplicable in high-energy-consuming industrial loads. Therefore, a more robust approach is developed comprising a three-algorithm model to achieve this goal on the IPL. First, the improved variational mode decomposition (IVMD) algorithm is introduced to denoise the training data of the IPL and improve its stability. Secondly, the convolutional neural network (CNN) and simple recurrent units (SRU) joint algorithms are used to achieve a non-intrusive and non-invasive decomposition process of the IPL using a double-layer deep learning network based on the IPL characteristics. Specifically, CNN is used to extract the IPL data characteristics while the improved long and short-term memory (LSTM) network, SRU, is adopted to develop the decomposition model and further train the load data. Through the robust decomposition process, the underlying relationship in the load consumption is extracted. The results obtained from the numerical examples show that this approach outperforms the state-of-the-art in the conventional decomposition process.
Appliance-level data is important for developing flexible two-way interactions between users and smart grids. Non-intrusive load monitoring (NILM) is a better way to obtain appliance power consumption information. Algorithms are used to decompose customers’ total electricity consumption data into electricity consumption data of various appliances. In order to realize real-time load identification, a load identification method is proposed based on the operating probability of load in different periods. During the training phase, historical data is used to count the probability of the device being in various states at various time periods. Then, in the load decomposition stage, several appliances state estimation matrices are generated using the time-segmented state probability, and the performance function selects the optimal matrix as the identification result of the appliance state. Finally, the proposed algorithm is tested on the low-frequency dataset, and the test results verified that the load status recognition accuracy is more than 96%, which meets the application requirements of NILM.
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%).
This paper presents a critical approach to the non-intrusive load monitoring (NILM) problem, by thoroughly reviewing the experimental framework of both legacy and state-of-the-art studies. Some of the most widely used NILM datasets are presented and their characteristics, such as sampling rate and measurements availability are presented and correlated with the performance of NILM algorithms. Feature engineering approaches are analyzed, comparing the hand-made with the automatic feature extraction process, in terms of complexity and efficiency. The evolution of the learning approaches through time is presented, making an effort to assess the contribution of the latest state-of-the-art deep learning models to the problem. Performance evaluation methods and evaluation metrics are demonstrated and it is attempted to define the necessary requirements for the conduction of fair evaluation across different methods and datasets. NILM limitations are highlighted and future research directions are suggested.
With the cost of consuming resources increasing (both economically and ecologically), homeowners need to find ways to curb consumption. The Almanac of Minutely Power dataset Version 2 (AMPds2) has been released to help computational sustainability researchers, power and energy engineers, building scientists and technologists, utility companies, and eco-feedback researchers test their models, systems, algorithms, or prototypes on real house data. In the vast majority of cases, real-world datasets lead to more accurate models and algorithms. AMPds2 is the first dataset to capture all three main types of consumption (electricity, water, and natural gas) over a long period of time (2 years) and provide 11 measurement characteristics for electricity. No other such datasets from Canada exist. Each meter has 730 days of captured data. We also include environmental and utility billing data for cost analysis. AMPds2 data has been pre-cleaned to provide for consistent and comparable accuracy results amongst different researchers and machine learning algorithms.
Understanding how appliances in a house consume power is important when making intelligent and informed decisions about conserving energy. Appliances can turn ON and OFF either by the actions of occupants or by automatic sensing and actuation (e.g., thermostat). It is, also, difficult to understand how much a load consumes at any given operational state. Occupants could buy sensors that would help, but this comes at a high financial cost. Power utility companies around the world are now replacing old electro-mechanical meters with digital meters (smart meters) that have enhanced communication capabilities. These smart meters are essentially free sensors that offer an opportunity to use computation to infer what loads are running and how much each load is consuming (i.e., load disaggregation). We present a new load disaggregation algorithm that uses a super-state hidden Markov model and a new Viterbi algorithm variant which preserves dependencies between loads and can disaggregate multi-state loads, all while performing computationally efficient exact inference. Our sparse Viterbi algorithm can efficiently compute sparse matrices with a large number of super-states. Additionally, our disaggregator can run in real-time on an inexpensive embedded processor using low sampling rates.
With recent developments in the infrastructure of smart meters and smart grid, more electric power data is available and allows real-time easy data access. Modeling individual home appliance loads is important for tasks such as non-intrusive load disaggregation, load forecasting, and demand response support. Previous methods usually require sub-metering individual appliances in a home separately to determine the appliance models, which may not be practical, since we may only be able to observe aggregated real power signals for the entire-home through smart meters deployed in the field. In this paper, we propose a model, named Explicit-Duration Hidden Markov Model with differential observations (EDHMM-diff), for detecting and estimating individual home appliance loads from aggregated power signals collected by ordinary smart meters. Experiments on synthetic data and real data demonstrate that the EDHMM-diff model and the specialized forward-backward algorithm can effectively model major home appliance loads.
With the help of smart metering valuable information of the appliance usage
can be retrieved. In detail, non-intrusive load monitoring (NILM), also called
load disaggregation, tries to identify appliances in the power draw of an
household. In this paper an unsupervised load disaggregation approach is
proposed that works without a priori knowledge about appliances. The proposed
algorithm works autonomously in real time. The number of used appliances and
the corresponding appliance models are learned in operation and are
progressively updated. The proposed algorithm is considering each useful and
suitable detected power state. The algorithm tries to detect power states
corresponding to on/off appliances as well as to multi-state appliances based
on active power measurements in 1s resolution. We evaluated the novel
introduced load disaggregation approach on real world data by testing the
possibility to disaggregate energy demand on appliance level.
Nonintrusive load monitoring (NILM), sometimes referred to as load disaggregation, is the process of determining what loads or appliances are running in a house from analysis of the power signal of the whole-house power meter. As the popularity of NILM grows, we find there is no consistent way researchers are measuring and reporting accuracies. In this short communication, we present a unified approach that would allow for consistent accuracy testing.
Smart metering and fine-grained energy data are one of the major enablers for the future smart grid and improved energy efficiency in smart homes. By using the information provided by smart meter power draw, valuable information can be extracted as disaggregated appliance power draws by non-intrusive load monitoring (NILM). NILM allows to identify appliances according to their power characteristics in the total power consumption of a household, measured by one sensor, the smart meter. In this paper we present a NILM approach, where the appliance states are estimated by particle filtering (PF). PF is used for non-linear and non-Gaussian disturbed problems and is suitable to estimate the appliance state. On/off appliances, multi-state appliances, or combinations of them are modeled by hidden Markov models (HMM) and their combinations result in a factorial hidden Markov model (FHMM) modeling the household power demand. We evaluate the PF-based NILM approach on synthetic and on real data from a well-known dataset to show that our approach achieves an accuracy of 90% on real household power draws.
A home-based intelligent energy conservation system needs to know what appliances (or loads) are being used in the home and when they are being used in order to provide intelligent feedback or to make intelligent decisions. This analysis task is known as load disaggregation or non-intrusive load monitoring (NILM). The datasets used for NILM research generally contain real power readings, with the data often being too coarse for more sophisticated analysis algorithms, and often covering too short a time period. We present the Almanac of Minutely Power dataset (AMPds) for load disaggregation research; it contains one year of data that includes 11 measurements at one minute intervals for 21 sub-meters. AMPds also includes natural gas and water consumption data. Finally, we use AMPds to present findings from our own load disaggregation algorithm to show that current, rather than real power, is a more effective measure for NILM.
Non-intrusive load monitoring (NILM) identifies used appliances in a total power load according to their individual load characteristics. In this paper we propose an evolutionary optimization algorithm to identify appliances, which are modeled as on/off appliances. We evaluate our proposed evolutionary optimization by simulation with Matlab, where we use a random total load and randomly generated power profiles to make a statement of the applicability of the evolutionary algorithm as optimization technique for NILM. Our results shows that the evolutionary approach is feasible to be used in NILM systems and can reach satisfying detection probabilities.
The smart meter is often heralded as the key component supporting energy displays that can notify home occupants of their energy usage. But, a smart meter is only a digital power meter with enhanced communications capabilities -- it is not actually smart. We need to look beyond the smart meter and define what intelligence is needed to actually make a meter smart. One area with promise is load disaggregation. Load disaggregation can be used to determine what loads contributing to the consumption reading at the smart meter. A smart meter incorporating load disaggregation intelligence can be seen as going beyond the traditional smart meter -- what we call a cognitive power meter (c-meter). However, using load disaggregation, in its current form, is not feasible. We critically review the requirements for a c-meter and provide insights as to how load disaggregation research needs to change to make the c-meters a reality.
This paper presents a new nonintrusive appliance load monitoring technique based on the integer programming. Nonintrusive appliance load monitoring is a problem to identify the operating conditions of the electric appliances in a house only by observing the overall load current and voltage. Since the overall load current is expressed as a superposition of each current of the operating appliance, the monitoring problem can be formulated as an integer quadratic programming problem by expressing the operating conditions as integer variables. This problem is solvable with sufficiently small computational burden thanks to the recent development of the commercial software. The proposed method does not require the relearning even when a new appliance is installed in the house. Furthermore, the proposed formulation is applicable to cases that some appliance has multiple modes, and cases that some same-type appliances operate simultaneously. Usefulness of the proposed technique is verified through some experimental results.
We propose a dynamic intelligent load balancing and reactive power compensation switch plugin for power meters and a smart scalable system architecture applicable as addon on top of the existing power distribution networks. The load balancing module will improve the load balance in a network and thus decrease the power losses, whereas the distributed reactive power compensator will decrease the total reactive power. The novel approach is presented from its IT/networks perspective as well as from the physical layer with oscilloscope measurements and time analysis of the phase swapping event.
With ongoing large-scale smart energy metering deployments worldwide, disaggregation of a household’s total energy consumption down to individual appliances using analytical tools, also known as non-intrusive appliance load monitoring (NALM), has generated increased research interest lately. NALM can deepen energy feedback, support appliance retrofit advice, and support home automation. However, despite the fact that NALM was proposed over 30 years ago, there are still many open challenges with respect to its practicality and effectiveness at low sampling rates. Indeed, the majority of NALM approaches, supervised or unsupervised, require training to build appliance models, and are sensitive to appliance changes in the house, thus requiring regular re-training. In this paper, we tackle this challenge by proposing an NALM approach that does not require any training. The main idea is to build upon the emerging field of graph signal processing to perform adaptive thresholding, signal clustering, and pattern matching. We determine the performance limits of our approach and demonstrate its usefulness in practice. Using two open access datasets—the US REDD data set with active power measurements downsampled to 1 min resolution and the UK REFIT data set with 8-s resolution, we demonstrate the effectiveness of the proposed method for typical smart meter sampling rate, with the state-of-the-art supervised and unsupervised NALM approaches as benchmarks.11Part of this work was presented at IEEE GlobalSIP-2015 [1]. The REFIT dataset used to generate the results can be accessed via DOI 10.15129/31da3ece-f902-4e95-a093-e0a9536983c4.
This paper proposes a novel load disaggregation algorithm, based on non-intrusive appliance monitoring that provides accurate results without the need of the reactive power component. The proposed technique requires the minimum possible number of sensor meters within the smart meter hardware and reduces the data traffic of the advanced metering infrastructure network, since it processes the only low-frequency smart meter data of the active power component of a residential unit. We propose a three-stage process for energy usage analysis, based on pulse extraction, pulse clustering and classification, and pulse to appliance association. The smart meter data are decomposed into a discrete set of pulses, and each pulse is associated with the operation of the appliances. The outcome of the algorithm is a pulse to appliance association that creates a daily load disaggregation map for the residential unit. In addition, this paper presents an overview and comparison of existing solutions that will help the reader obtain a good overview of the landscape of load disaggregation from smart meter sensor readings.
Energy and sustainability issues raise a large number of problems that can be tackled using approaches from data mining and machine learning, but traction of such problems has been slow due to the lack of publicly available data. In this paper we present the Reference Energy Disaggregation Data Set (REDD), a freely available data set containing de-tailed power usage information from several homes, which is aimed at furthering research on energy disaggregation (the task of determining the component appliance contributions from an aggregated electricity signal). We discuss past ap-proaches to disaggregation and how they have influenced our design choices in collecting data, we describe the hardware and software setups for the data collection, and we present initial benchmark disaggregation results using a well-known Factorial Hidden Markov Model (FHMM) technique.
Recent studies have highlighted that a significant part of the electrical energy consumption in residential buildings is caused by an improper use of home appliances. The development of low-cost systems for profiling the consumption of electric appliances can play a key role in stimulating the users to adopt adequate policies for energy saving. In this paper, we describe a novel methodology for extracting the power consumption of each appliance deployed in a domestic environment from the aggregate measures collected by a single smart meter. In order to coarsely describe how each type of appliance works, we use finite-state machines (FSMs) based on fuzzy transitions. An ad-hoc disaggregation algorithm exploits a database of these FSMs for, at each meaningful variation in real and reactive aggregate powers, hypothesizing possible configurations of active appliances. This set of configurations is concurrently managed by the algorithm which, whenever requested, outputs the configuration with the highest confidence with respect to the sequence of detected events. We implemented a prototype of a monitoring system based on the proposed methodology and installed it in a real domestic scenario. We discuss an experiment in which 11 appliances were connected to the same circuit and the aggregate power consumption was measured by a smart meter for approximately 12 h. At the end of the experiment, only two possible configurations were outputs from the system, including the correct one.
Home energy displays are emerging home energy management devices. Their energy saving potential is limited, because most display whole-home electricity consumption data. We propose a new approach to disaggregation electricity consumption by individual appliances and/or end uses that would enhance the effectiveness of home energy displays. The proposed method decomposes a system of appliance models into tuplets of appliances overlapping in power draw. Each tuplet is disaggregated using a modified Viterbi algorithm. In this way, the complexity of the disaggregation algorithm is linearly proportional to the number of appliances. The superior accuracy of the method is illustrated by a simulation example and by actual household data.
Home energy displays are emerging home energy management devices. However, their energy savings potential is limited, because most display whole-home electricity consumption data. We propose a new approach to disaggregation electricity consumption by individual appliances and/or end uses that would enhance the effectiveness of home energy displays.
There is much interest in the Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM) as a natural Bayesian nonparametric extension of the ubiquitous Hidden Markov Model for learning from sequential and time-series data. However, in many settings the HDP-HMM's strict Markovian constraints are undesirable, particularly if we wish to learn or encode non-geometric state durations. We can extend the HDP-HMM to capture such structure by drawing upon explicit-duration semi-Markov modeling, which has been developed mainly in the parametric non-Bayesian setting, to allow construction of highly interpretable models that admit natural prior information on state durations.
In this paper we introduce the explicit-duration Hierarchical Dirichlet Process Hidden semi-Markov Model (HDP-HSMM) and develop sampling algorithms for efficient posterior inference. The methods we introduce also provide new methods for sampling inference in the finite Bayesian HSMM. Our modular Gibbs sampling methods can be embedded in samplers for larger hierarchical Bayesian models, adding semi-Markov chain modeling as another tool in the Bayesian inference toolbox. We demonstrate the utility of the HDP-HSMM and our inference methods on both synthetic and real experiments.
A multilevel aggregation/disaggregation approach to solving
large-scale power flow problems is introduced. This approach makes use
of REI-type network equivalents in building the system models for each
level. The modeling procedure is explained, and the utilization of
parallel processors to further improve the performance of the approach
is described. Several examples to illustrate the effectiveness of the
method in solving large size power flow problems are given
A new model is proposed for the study of transient stability where the load is modeled as a PQ bus. Flux decay of the generator field winding is included. The original network topology is maintained explicitly. An energy function is proposed which differs from the traditional one in that it includes additional terms corresponding to the energy stored in the loads and field winding. A characterization of the stability region is derived based on this energy function.
Practical Optimization
- A Antoniou
- W S Lu
A. Antoniou and W. S. Lu, Practical Optimization. Springer, 2007.
Smart metering implementation programme
"Smart metering implementation programme," tech. rep., Dept. Energy
Climate Change, London, U.K., 2013.