Demand Side Management Approach Research Articles

Design, Deployment and Performance Evaluation of an IoT Based Smart Energy Management System for Demand Side Management in Smart Grid

The Smart Grid (SG) is the technological development that incorporates digital technologies and advanced communication methods to determine and respond to variations in electricity consumption in order to revolutionize power distribution, transmission, and generation. In the conventional electrical grid, customers remain unaware to their energy usage patterns that not only results in energy loss but also money. The consumers’ usage and consumption standards need to be regulated in order to improve energy efficiency (EE). SG utilizes demand side management (DSM) for energy savings by use of various approaches like financial incentives, subsidized tariffs, and awareness to alter consumers’ energy demand. In smart environments (SE), Internet-of-Things (IoT) is evolving as a significant partner for resource and energy management. DSM in SG must take advantage of smart energy management system (SEMS) developed on smart meters (SMs) and modern technologies like IoT. Using SMs with IoT based technologies makes SEMS more effective in the SG. This paper offers design, deployment, implementation, and performance evaluation of an IoT based SEMS, including SMs as well as IoT middleware module and its related benefits. The proposed SEMS operates online and offers real-time (RT) load profiles (LPs) to customers and suppliers remotely. The customers’ LPs allow suppliers to disseminate and regulate their incentives as well as incite the customers to alter their energy consumption. Furthermore, these LPs serves as an input for developing numerous DSM approaches. The proposed solution is installed and evaluated at 4 different locations of Stylo Pvt. Ltd. Pakistan, which can communicate commands and observe the efficiency of electricity supplied by the utility. Moreover, the RT impact of using separate SM for automated control of heating, ventilation and air conditioning (HVAC) system is shown in terms of power consumption. The RT case study presents the efficacy of the proposed IoT-based-SEMS.

Exploring willingness-to-pay for the restoration and maintenance of reserved forests in the Greater Accra Region of Ghana

In recent years, debates on the alarming rate of forest depletion emanating from growth in urban settlement and changed urban land-use patterns have gained prominence across the globe. The present study adopts a demand-side management approach to investigate household's willingness-to-pay for the restoration and maintenance of protected forest reserves in three municipalities in Ghana. Using survey data of 733 households from the Greater Accra Region of Ghana, we model the demand for forest restoration and maintenance, respectively, by means of the contingent valuation technique. As part of the findings, the study shows evidence that a household is willing to pay Gh¢50.99 ($8.67) and Gh¢31.12 ($5.29) per annum to restore and maintain the protected forests, respectively. These amounts constitute less than one percent of the average household income per month. Consequently, critical validity tests are conducted to validate the robustness of the results. This study provides willingness-to-pay estimates for forest and forest resources, and its associated determinants. These estimates seek to bridge the information gap and inform policy decisions toward the overarching aim of ensuring sustainable forest management in Ghana.

A Demand Side Management Approach For Optimal Sizing of Standalone Renewable-Battery Systems

This paper develops a novel demand side management (DSM) approach to incorporate in optimal sizing of solar photovoltaic (PV), wind turbine (WT), and battery storage (BS) for a standalone household. The DSM strategy is based on the state-of-charge level of battery and day-ahead forecasts of solar insolation and wind speed. The core of the DSM is a fuzzy logic method which decides for efficient load shifting and/or load curtailment. The day-ahead forecasting errors, obtained by an artificial neural network technique, are considered not only in the DSM strategy but also in maintaining an operating reserve. The battery capacity degradation is calculated using the Rainflow counting algorithm to obtain a realistic battery model and estimate its lifetime. A typical household in South Australia (SA) is considered as a case study. Three different configurations (PV-BS, WT-BS, and PV-WT-BS) of the electricity supply system are optimized using the proposed method. It is found that the PV-WT-BS system is the best configuration that provides the lowest cost of electricity for both with and without applying the proposed DSM strategy. Comparison of the results of the best system configuration with an actual system in SA and two recently published articles indicates that the proposed method is very effective in lowering the electricity cost with zero-emission.

A Comprehensive Review on Residential Demand Side Management Strategies in Smart Grid Environment

The ever increasing demand for electricity and the rapid increase in the number of automatic electrical appliances have posed a critical energy management challenge for both utilities and consumers. Substantial work has been reported on the Home Energy Management System (HEMS) but to the best of our knowledge, there is no single review highlighting all recent and past developments on Demand Side Management (DSM) and HEMS altogether. The purpose of each study is to raise user comfort, load scheduling, energy minimization, or economic dispatch problem. Researchers have proposed different soft computing and optimization techniques to address the challenge, but still it seems to be a pressing issue. This paper presents a comprehensive review of research on DSM strategies to identify the challenging perspectives for future study. We have described DSM strategies, their deployment and communication technologies. The application of soft computing techniques such as Fuzzy Logic (FL), Artificial Neural Network (ANN), and Evolutionary Computation (EC) is discussed to deal with energy consumption minimization and scheduling problems. Different optimization-based DSM approaches are also reviewed. We have also reviewed the practical aspects of DSM implementation for smart energy management.

A new internet of things enabled trust distributed demand side management system

Renewable energy resources (RESs) have come under significant focus to cover the massive demand for electricity. Microgrids are coupled with Demand Side Management (DSM) for saving energy and amplifying energy efficacy. CO2 discharges, frequency and voltage protocols, RES stochastic nature, Peak to Average Ratio (PAR), and load dynamics are still termed as the most significant challenges for smart microgrids. Novel regulation and management methodologies are required for overpowering them. Internet of Things (IoT) is able to offer adaptive supervision of energy intake and warrant a cost-effective and safe functioning of the smart microgrids. In this study, a new DSM system called Real-Time Electricity Scheduling (RTES) for domestic home energy management is introduced for running the microgrids. The proposed management system seeks to curtail cost payments by optimally programming smart devices and enhancing the consumption of energy. The system comprises two portions: software and hardware. The hardware comprises a Base Station Unit (BSU) and many Terminal Unit (TU). The software comprises the Wi-Fi network programming along with the system protocol. In this paper, Message Queue Telemetry Transportation (MQTT) broker was constructed on TU board and BSU. The proposed technique raises effectual use of energy, thus raising the viability of IoT empowered households in smart cities. The approach impulsively reacts to protective home equipment, power factor correction, and price-centred demand response programmes to contest the key concern of the demand response programs, i.e., constraints in terms of customer’s knowledge to react upon obtaining demand response signals. Lastly, a smart microgrids model is experimentally employed to accomplish and corroborate the proposed DSM and control approach. The outcomes of the experiment have substantiated the efficacy and viability of the presented approach and the competence of proposed technique for energy management functioning in various modes. The outcomes endorse the proposed method’s ability to obtain an optimum DSM scheme by decreasing the smart microgrids’ emission cost, energy cost, and PAR, along with power factor correction.

Clustering and Fuzzy Logic-Based Demand-Side Management for Solar Microgrid Operation: Case Study of Ngurudoto Microgrid, Arusha, Tanzania

Permanent electricity availability should not be taken for granted since grid sustainability and reliability are at stake when there is no balance between supply and demand. This paper employs a load balancing demand-side management (DSM) approach based on fuzzy logic, considering the low energy users who have insignificant influence on system peaks. Through the K-means clustering algorithm, suitable candidates for DSM are identified, and the control mechanism is based on energy utilization and load priority. The results reveal that about 3.7 kW in power saving was achieved per month. This result indicates that, with a proper energy management strategy for an individual customer, almost a flatter load profile and power saving can be achieved.

Domestic demand-side management: analysis of microgrid with renewable energy sources using historical load data

This paper provides a high-accuracy assessment of domestic demand-side management (DSM) approach in the context of distributed renewable energy sources (RES). To determine the potential of domestic DSM for households, a microgrid model of a typical UK residential estate was developed to simulate the impact of RES. The microgrid model comprises 15 UK households with appropriate allocation of washing machines, tumble dryers and dishwashers in accordance with the statistical data. In order to obtain a high-accuracy result, the power consumption of the microgrid model utilises real historical high-resolution data of household energy consumption and RES generation. Thereafter, 40% of distributed wind and solar energy is implemented in the model to produce two individual scenarios. The operation of the white appliances in the model is controlled using a domestic DSM based on a load shifting algorithm. The primary criterion of the DSM considered in this paper is the reduction in energy feedback to the grid in order to decrease the utilisation of the grid and to reduce the transmission losses. The results obtained from the model simulation are compared to the baseline model and discussed with respect to the possible benefits of implementation of domestic DSM under the impact of RES. It has been shown that the self-consumption ratio of the microgrid operating under the DSM is increased by 3% for both scenarios. The model analysis provides highly realistic results which can be used for efficiency assessment of various load shifting methods.

New Approach using Artificial Intelligence-Machine Learning in Demand Side Management of Renewable Energy integrated Smart Grid for Smart City

With the emergence of smart sustainable cities, renewable energy enabled smart grid with active consumer participation is in quest. But to entrap the potential of renewable energy integrated smart grid is strenuous. It has two major correlative power concerns: 1) To synchronize demand-supply and generation scarcity-excess situation with least curtailment 2) Administering variability-uncertainty issues due to renewable energy generation to proportionate adequacy of power supply. Demand-side management (DSM) can be the key solution. An Artificial Intelligence/ Machine learning-based DSM will offer fast responses on a real-time basis to trace consumer consumption patterns, predicting energy forecast, energy market participation, demand response strategies, manage power outages, predicting system loopholes along with sustaining low-cost per unit charges. This paper draws a conclusive summary in tabular form of the thorough review carried out on the latest demand-side management approaches using artificial intelligence/ machine learning to unhide gaps and predict its path for future expansion.

DEMAND SIDE MANAGEMENT OF RENEWABLE ENERGY INTEGRATED SMART GRID USING LOAD SHIFTING TECHNIQUES

The Paris Agreement on Climate Change has led to introduction of new reforms for clean power plan such as decarbonization of power sector, planned decommissioning of thermal power plants and inclusion of renewable energies for power production. But this desired integration of renewable energy resources to power system faces two technical challenges: variability and uncertainty. An effective energy management with help of smart grid engineering can be the key for its beneficial use. Demand Side Management (DSM) is a valuable strategy for energy management in smart grid. It supports numerous smart grid functionalities for instance electricity market control, Load scheduling, managing decentralized distributed energy resources. Identifying energy consumption patterns and to sketch electricity load profiles can be achieved through numerous DSM based programs. Load shifting based DSM can be linked to consumer’s behavior in understanding their pattern of energy consumption. Here, the practiced load shifting based demand side management approach can help in maximizing power efficiency, sustaining power reliability and resiliency of renewable sources. This paper reviews the various energy management strategies developed to minimize the impact of renewable energy intermittency using Load Shifting Demand Side Management (DSM) approach.

A Game Theoretic Approach for Demand-Side Management Using Real-Time Variable Peak Pricing Considering Distributed Energy Resources

The pricing scheme plays a vital role in the demand-side management program. It allows utility, indirect control over the users energy consumption. This article proposes a new pricing structure, named as a real-time variable peak pricing scheme with a self-inbuilt feature to capture the benefit of both the existing pricing scheme, real-time pricing (RTP), and time of use (TOU), by switching its behavior between RTP and TOU as per the requirements. A home energy management model with comprehensive modeling of home appliances together with distributed storage and dispatchable generation is considered for testing the impact of the proposed pricing scheme and compared to existing RTP and TOU. The privacy and comfort of users are taken care of in the system modeling. An aggregative game approach is used to capture the interaction between users and utility. The conditions for the existence of Nash equilibrium and convergence of the proposed algorithm is investigated for all three pricing schemes considering residential users. The analysis shows the effectiveness of the proposed pricing scheme compare to the existing pricing schemes in terms of consumers energy bill and system peak reduction, complexity, and incentive to active consumers to participate in demand-side management programs.

Artificial Intelligence, Accelerated in Parallel Computing and Applied to Nonintrusive Appliance Load Monitoring for Residential Demand-Side Management in a Smart Grid: A Comparative Study

A smart grid is a promising use-case of AIoT (AI (artificial intelligence) across IoT (internet of things)) that enables bidirectional communication among utilities that arises with demand response (DR) schemes for demand-side management (DSM) and consumers that manage their power demands according to received DR signals. Disaggregating composite electric energy consumption data from a single minimal set of plug-panel current and voltage sensors installed at the electric panel in a practical field of interest, nonintrusive appliance load monitoring (NIALM), a cost-effective load disaggregation approach for (residential) DSM, is able to discern individual electrical appliances concerned without accessing each of them by individual plug-load power meters (smart plugs) deployed intrusively. The most common load disaggregation approaches are based on machine learning algorithms such as artificial neural networks, while approaches based on evolutionary computing, metaheuristic algorithms considered as global optimization and search techniques, have recently caught the attention of researchers. This paper presents a genetic algorithm, developed in consideration of parallel evolutionary computing, and aims to address NIALM, whereby load disaggregation from composite electric energy consumption data is declared as a combinatorial optimization problem and is solved by the algorithm. The algorithm is accelerated in parallel, as it would involve large amounts of NIALM data disaggregated through evolutionary computing, chromosomes, and/or evolutionary cycles to dominate its performance in load disaggregation and excessively cost its execution time. Moreover, the evolutionary computing implementation based on parallel computing, a feed-forward, multilayer artificial neural network that can learn from training data across all available workers of a parallel pool on a machine (in parallel computing) addresses the same NIALM/load disaggregation. Where, a comparative study is made in this paper. The presented methodology is experimentally validated by and applied on a publicly available reference dataset.

Internet of things and cloud computing‐based energy management system for demand side management in smart grid

A smart grid is an electricity network, which deals with electronic power conditioning and control of production, transmission, and distribution of electrical power by employing digital communication technologies to monitor and manage local changes in electricity usage. In the traditional power grid, energy consumers remain oblivious to their power consumption patterns, resulting in wasted energy as well as money. This issue is severely pronounced in the developing countries where there is a huge gap between demand and supply, resulting in frequent power outages and load-shedding. For electrical energy savings, the smart grid employs demand side management (DSM), which refers to adaptation in consumer's demand for energy through various approaches such as financial incentives and awareness. The DSM in future smart grid must exploit automated energy management systems (EMS) built upon the state-of-the-art technologies such as the internet of things (IoT) and cloud and/or fog computing. In this paper, we present the architecture framework, design, and implementation of an IoT and cloud computing-based EMS, which generates load profile of consumer to be accessed remotely by utility company or by the consumer. The consumers' load profiles enable utility companies to regulate and disseminate their incentives and incite the consumers to adapt their energy consumption. Our designed EMS is implemented on a Project Circuit Board (PCB) to be easily installed at the consumer premises where it performs the following tasks: (a) monitors energy consumption of electrical appliances by means of our designed current and voltage sensors, (b) uploads sensed data to Google Firebase cloud over many-to-many IoT communication protocol Message Queuing Telemetry Transport (MQTT) where consumer's load profile is generated, which can be accessed via a web portal. These load profiles serve as input for implementing the various DSM approaches. Our results demonstrate generated load profiles of consumer load in terms of current, voltage, energy, and power accessible via a web portal.

Scheduling of Air Conditioning and Thermal Energy Storage Systems Considering Demand Response Programs

The high penetration rate of renewable energy sources (RESs) in smart energy systems has both threat and opportunity consequences. On the positive side, it is inevitable that RESs are beneficial with respect to conventional energy resources from the environmental aspects. On the negative side, the RESs are a great source of uncertainty, which will make challenges for the system operators to cope with. To tackle the issues of the negative side, there are several methods to deal with intermittent RESs, such as electrical and thermal energy storage systems (TESSs). In fact, pairing RESs to electrical energy storage systems (ESSs) has favorable economic opportunities for the facility owners and power grid operators (PGO), simultaneously. Moreover, the application of demand-side management approaches, such as demand response programs (DRPs) on flexible loads, specifically thermal loads, is an effective solution through the system operation. To this end, in this work, an air conditioning system (A/C system) with a TESS has been studied as a way of volatility compensation of the wind farm forecast-errors (WFFEs). Additionally, the WFFEs are investigated from multiple visions to assist the dispatch of the storage facilities. The operation design is presented for the A/C systems in both day-ahead and real-time operations based on the specifications of WFFEs. Analyzing the output results, the main aims of the work, in terms of applying DRPs and make-up of WFFEs to the scheduling of A/C system and TESS, will be evaluated. The dispatched cooling and base loads show the superiority of the proposed method, which has a smoother curve compared to the original curve. Further, the WFFEs application has proved and demonstrated a way better function than the other uncertainty management techniques by committing and compensating the forecast errors of cooling loads.

Robust Optimal Energy Management of a Residential Microgrid Under Uncertainties on Demand and Renewable Power Generation

Smart microgrids are experiencing an increasing growth due to their economic, social, and environmental benefits. However, the inherent intermittency of renewable energy sources (RESs) and users' behavior lead to significant uncertainty, which implies important challenges on the system design. Facing this issue, this article proposes a novel robust framework for the day-ahead energy scheduling of a residential microgrid comprising interconnected smart users, each owning individual RESs, noncontrollable loads (NCLs), energy- and comfort-based CLs, and individual plug-in electric vehicles (PEVs). Moreover, users share a number of RESs and an energy storage system (ESS). We assume that the microgrid can buy/sell energy from/to the grid subject to quadratic/linear dynamic pricing functions. The objective of scheduling is minimizing the expected energy cost while satisfying device/comfort/contractual constraints, including feasibility constraints on energy transfer between users and the grid under RES generation and users' demand uncertainties. To this aim, first, we formulate a min-max robust problem to obtain the optimal CLs scheduling and charging/discharging strategies of the ESS and PEVs. Then, based on the duality theory for multi-objective optimization, we transform the min-max problem into a mixed-integer quadratic programming problem to solve the equivalent robust counterpart of the scheduling problem effectively. We deal with the conservativeness of the proposed approach for different scenarios and quantify the effects of the budget of uncertainty on the cost saving, the peak-to-average ratio, and the constraints' violation rate. We validate the effectiveness of the method on a simulated case study and we compare the results with a related robust approach. Note to Practitioners-This article is motivated by the emerging need for intelligent demand-side management (DSM) approaches in smart microgrids in the presence of both power generation and demand uncertainties. The proposed robust energy scheduling strategy allows the decision maker (i.e., the energy manager of the microgrid) to make a satisfactory tradeoff between the users' payment and constraints' violation rate considering the energy cost saving, the system technical limitations and the users' comfort by adjusting the values of the budget of uncertainty. The proposed framework is generic and flexible as it can be applied to different structures of microgrids considering various types of uncertainties in energy generation or demand.

A novel billing approach for fair and effective demand side management: Appliance level billing (AppLeBill)

Demand side management (DSM) is an important part of smart electrical grid studies due to high energy efficiency it provides. Effective DSM techniques are hinged on participation level of the customers and control manner of their loads so if there is no fair and appropriate management, customers will not embrace demand side management applications. Besides, the customers might not participate to the demand response process willingly if their comforts are not considered. Therefore, the DSM systems must consider fairness and customer’s comfort as well as distribution system objectives. Multifarious DSM systems have been proposed in the literature, however, the fair billing has not included as a parameter in most of these studies. This paper introduces a novel billing approach based upon appliance level billing (AppLeBill) for fair and effective demand side management. The proposed AppLeBill idea offers a billing strategy such that time-shiftable appliances (TSAs) in a residence may have a separate billing rate rather than a single billing rate for total electricity usage. The proposed AppLeBill concept includes three different schemes which are explained and simulated in order to compare each other in terms of their potential benefits.

Mixed data-driven decision-making in demand response management: An empirical evidence from dynamic time-warping based nonparametric-matching DID

As an important approach for demand-side management in the power sector, demand responses (DRs) are increasingly important in guiding scientific energy consumption behaviour. However, most related prior studies are based on small-scale experimental or survey data with a rule-based optimization algorithm; scientific DR management and strategy formulation studies driven by large-scale, hybrid frequency data are rare. This paper integrates a large-scale controlled trial, 15 min high-frequency power consumption data, and individual residents’ monthly low-frequency power consumption data on a micro-scale. The data-driven and causal analysis methods are combined and a machine-learning algorithm have been adopted to propose a dynamic time-warping (DTW) clustering-based difference-in-differences (DID) method. This non-parametric matching method successfully results in an intra-group randomized experiment. Empirical results reveal that cash-incentive-based DR can effectively stimulate electricity-saving behaviour, and families from the treatment groups save an average of 27.3% of their total electricity consumption in the experimental period. Further, a dynamic response process analysis indicates that a substantial discrepancy exists in the degree of demand response and the response modes of residents with different power consumption patterns. More importantly, prior empirical studies proved this method's effectiveness and feasibility: based on the DTW non-parametric matching method, the control and treatment groups can well support the parallel trend hypothesis. This work provides important implications for the accurate, efficient implementation and scientific decision-making of subsequent DR programs.

A DSM Approach for Distribution Systems with High Wind Power Penetration

The basic objectives of demand side management (DSM) are shifting load from peak hours to off-peak hours and reducing consumption during peak hours. The DSM operation is cleared when deregulated electricity market is considered where the retailer purchases electricity from the electricity market to cover the end users requirements of energy. In this paper, DSM techniques (load shifting and peak clipping) are used to maximize the profit for retailer company by reducing total demand at peak hours and achieve an optimal daily load schedule using linear programing (LP) and genetic algorithm (GA). These techniques are implemented on the 33-bus radial network included wind generation penetration. A short term artificial neural network technique (ANN) is used to get forecasted wind speed and forecasted users load for date 25-March-2018. The ANN uses an actual hourly load data and an actual hourly wind speed data. Then the forecasted data is used in the optimization to get optimal daily load schedule to maximize the profit for retailer company. Finally, comparison is carried out between profit using LP and GA. The optimized DSM succeeded to increase the profits of the company by around 4.5 times its old profit using LP and around 2.5 times using GA.

Approach for Dynamic Price-Based Demand Side Management in Cyber-Physical Production Systems

Nowadays, the potential of intelligent energy management concepts varies in many areas, e.g. smart energy usage. Trends in the energy sector for electricity depend on various factors. One factor is the fluctuating supply and distribution of renewable energy, leading to a necessary load control of smart grids. Another one is the need for a more flexible demand side management, especially for the manufacturing industry. Based on the increasing connectivity of modular Cyber-Physical Production Systems (CPPS), new opportunities of demand side management concepts can be identified. The aim of this paper is to show an approach for a dynamic price-based Demand Side Management (DSM) in CPPS, with regards to economic and technological aspects. We focus on DSM and dynamic pricing for a CPPS-side interaction with smart grids and markets in the energy sector for electricity. Furthermore, the meaning of collecting, transforming and learning from data about energy consumption by using CPPS and the resulting possibilities for DSM will be shown. Based on these findings, possible impacts for manufacturers will be identified, e.g. possibilities for a better resource adaptation of electrical energy, which is related to production planning data or the identification for reconfiguration of production processes.

The Mode Most Traveled: Parking Implications and Policy Responses

A disproportionate number of Americans drive to work alone – at the individual state level, the figure ranges from a low of 58% in New York to 85% in Alabama. What factors explain this travel behavior and what sort of policy responses are required to effect changes? In responding to this question, we used multiple test beds and made the case for using a demand side management approach to remedy parking problems particularly observed in cities central business districts. To this end, we provide an overview of travel behavior and information on policy levers by generating detailed profiles that capture the travel behavior of commuters in the Pacific states of the continental United States. Building off the result that revealed San Francisco as an outlier, we examine the efficacy of modifying parking rates, via elasticity measures, to influence the demand for parking by developing a two-stage panel data regression optimization model for managing parking in the City of San Francisco. A key contribution of the research is deriving these price elasticities of parking demand estimates using panel data methods. Coefficient estimates from the panel data regression are used to fit a linear prediction model that is the primary input to the optimization model. The balance of the thesis focuses on parking information by discussing the design and implementation of ParkPGH, a novel smart parking application that provides real time and predictive information on garage parking availability in downtown Pittsburgh. At its core is a predictive model that uses as input historical parking, weather and event data to provide estimates of available parking spaces. We provide an example of the model implementation using data from the Theater Square garage where we utilize neural network-based predictors and multiple net searches to generate both continuous and binary estimates of parking availability. Provision was made for the binary classifier given the need to reduce the possibility of Type II errors.

Semisupervised Multilabel Deep Learning Based Nonintrusive Load Monitoring in Smart Grids

Nonintrusive load monitoring (NILM) is a technique that infers appliance-level energy consumption patterns and operation state changes based on feeder power signals. With the availability of fine-grained electric load profiles, there has been increasing interest in using this approach for demand-side energy management in smart grids. NILM is a multilabel classification problem due to the simultaneous operation of multiple appliances. Recently, deep learning based techniques have been shown to be a promising approach to solving this problem, but annotating the huge volume of load profile data with multiple active appliances for learning is very challenging and impractical. In this article, a new semisupervised multilabel deep learning based framework is proposed to address this problem with the goal of mitigating the reliance on large labeled datasets. Specifically, a temporal convolutional neural network is used to automatically extract high-level load signatures for individual appliances. These signatures can be efficiently used to improve the feature representation capability of the framework. Case studies conducted on two open-access NILM datasets demonstrate the effectiveness and superiority of the proposed approach.