Appliance Load Research Articles

Multiscale Spatio-Temporal Feature Fusion Based Non-Intrusive Appliance Load Monitoring for Multiple Industrial Industries

The appliance types and power consumption patterns vary greatly across different industries. This can lead to unstable identification results of traditional appliance load monitoring methods in different industries. A non-intrusive appliance load monitoring (NIALM) method for multiple industries based on multiscale spatio-temporal feature fusion has been proposed. Firstly, the ConvNeXt Block with efficient channel attention has strong feature extraction capability. Spatial features of appliance state changes and micro-variations generated during operation can be extracted from mixed industrial load information by it. Meanwhile, the bidirectional gated recurrent neural network is used to learn the bidirectional dependencies of the load data, obtaining temporal features. Then, the multi-scale feature extraction module is used to extract temporal and spatial features from different depths of the network layers. And the extracted multi-scale temporal and spatial features are fully integrated. Finally, the proposed model is optimized using the Stochastic Weight Averaging method. During the training process, a certain number of model weights are randomly averaged, which can improve the model's generalization ability and identification accuracy. The experiment was conducted on six different industries. The evaluation indexes such as accuracy, F1 score, and Wasserstein distance are also used to verify the effectiveness and superiority of the method.

Smart Non-Intrusive Appliance Load-Monitoring System Based on Phase Diagram Analysis

Much of today’s power grid was designed and built using technologies and organizational principles developed decades ago. The lack of energy resources and classic power networks are the main causes of the development of the smart grid to efficiently use energy resources, with stable and safe operation. In such a network, one of the fundamental priorities is provided by non-intrusive appliance load monitoring (NIALM) in order to analyze, recognize and determine the electricity consumption of each consumer. In this paper, we propose a new smart system approach for the characterization of the appliance load signature based on a data-driven method, namely the phase diagram. Our aim is to use the non-intrusive load monitoring of appliances in order to recognize different types of consumers that can exist within a smart building.

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

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

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

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

Designing An Appliance Load Monitor with Programmable Counter for Energy Efficiency

The decentralization of production sources is a way to increase the energy efficiency of domestic buildings. Energy is indeed an important source of human activity, but due to its limitation and mismanagement, energy is becoming increasingly scarce, especially in developing countries. The use of electrical energy is mainly concentrated in electrical installations for domestic, administrative use and in industry. We are faced with wasting energy and increasing electricity bills. To solve these problems, we propose a solution that involves designing an appliance load monitor with a programmable meter for energy efficiencyto manage and track electrical energy consumption in high energy consumption buildings. New technologies are changing the way we live, in transportation, communications, commerce and even the way we design our homes, providing the technology that allows us to improve our quality of life in our homes, providing the technology necessary to communicate, send and receive messages. The Internet makes it possible to observe and download consumption data over long distances. We have integrated a system for controlling and monitoring our electricity consumption via a mobile interface that we designed. This system will allow us to see the different electrical values of our consumption and to control the operating time of the device. Keywords: Microcontrollers, sensors, Arduino, Raspberry Pi, energy management, mobile application.

STNILM: Switch Transformer based Non-Intrusive Load Monitoring for short and long duration appliances

Non-Intrusive Load Monitoring (NILM) is a technique used by contemporary energy management systems to predict and optimize appliance load distribution in real time. The real-time reduction of energy consumption and improvement of electricity efficiency are two major benefits of energy disaggregation. Transformer models have made NILM far better at forecasting device power values. Due to the absence of inductive bias in the local context, transformers may not be able to capture local signal patterns in sequence-to-point settings. In this work, we present a Switch Transformer based Non-Intrusive Load Monitoring (STNILM). STNILM utilizes switching and routing layers by replacing the vanilla transformer final layers to accurately estimate the power signals of short and long duration domestic appliances. It also uses self attention mechanisms to extract global dependencies between the aggregate and the domestic appliance signals. STNILM works with minimal dataset pre-processing and unbalanced. With extensive experiments and quantitative analysis, we demonstrate the efficiency and effectiveness of the proposed STNILM with considerable improvements in terms of accuracy and F1-score compared to state-of-the-art baselines.

Equipment- and Time-Constrained Data Acquisition Protocol for Non-Intrusive Appliance Load Monitoring

Energy behaviours will play a key role in decarbonising the building sector but require the provision of tailored insights to assist occupants to reduce their energy use. Energy disaggregation has been proposed to provide such information on the appliance level without needing a smart meter plugged in to each load. However, the use of public datasets with pre-collected data employed for energy disaggregation is associated with limitations regarding its compatibility with random households, while gathering data on the ground still requires extensive, and hitherto under-deployed, equipment and time commitments. Going beyond these two approaches, here, we propose a novel data acquisition protocol based on multiplexing appliances’ signals to create an artificial database for energy disaggregation implementations tailored to each household and dedicated to performing under conditions of time and equipment constraints, requiring that only one smart meter be used and for less than a day. In a case study of a Greek household, we train and compare four common algorithms based on the data gathered through this protocol and perform two tests: an out-of-sample test in the artificially multiplexed signal, and an external test to predict the household’s appliances’ operation based on the time series of a real total consumption signal. We find accurate monitoring of the operation and the power consumption level of high-power appliances, while in low-power appliances the operation is still found to be followed accurately but is also associated with some incorrect triggers. These insights attest to the efficacy of the protocol and its ability to produce meaningful tips for changing energy behaviours even under constraints, while in said conditions, we also find that long short-term memory neural networks consistently outperform all other algorithms, with decision trees closely following.

Modular sandwich panel system for non-loadbearing walls – Experimental mechanical, fire and acoustic testing

Investment in the construction industry has increased over the last years and conventional construction materials, techniques, solutions and practices cannot meet the current needs. Resources are not being used to their full potential and there is a lack of manpower. It is vital to evolve encouraging sustainable and quality cost-effective materials, techniques, solutions and practices, therefore also enhancing the construction workers’ quality of life. This work proposes a new modular sandwich panel system for interior walls partition applications. The modular sandwich panel is made of an extruded polystyrene boards (XPS) core and has three options as facesheets: gypsum plasterboard (PGC), fire resistant gypsum plasterboard (PGF) and magnesium oxide board (MGO). This is an alternative solution for partition walls based on five main factors: it has the potential to be reused; its modules are self-supported (they do not need vertical metal studs from floor to ceiling); it is easy to install (on-site) by only two people; it is quick to install; and it provides easy local access to the interior of sanitary facilities. This work also includes experimental research for evaluation of proposed the wall solution in terms of mechanical resistance and stability, fire safety and airborne sound insulation between adjacent rooms. Mechanical resistance testing includes impact resistance tests, simple compression load tests, sanitary appliance load tests and endurance cycle load tests. Fire resistance testing allows to assess the integrity and maintenance of the panel wall under fire conditions. Acoustic experiments aim to determine the weighted standardised sound level difference (DnT,w) between adjacent compartments separated by the proposed panel wall system. Mechanical resistance tests indicate that the panel wall system has acceptable mechanical performance, including the connections between the panel wall modules. Results indicate that fire resistance and sound insulation characteristics of the panel wall can be enhanced.

Bottom-up assessment of household electricity consumption in dynamic cities of the Global South—Evidence from Kigali, Rwanda

Data on electricity consumption is crucial for assessing and modeling energy systems, making it a key element of sustainable urban planning. However, many countries in the Global South struggle with a shortage of statistically valid, geocoded, and disaggregated household-level data. This paper aims to develop a generic methodology for the generation of such a database in terms of electricity consumption. The methodology was tested in Kigali, the capital city of Rwanda, with a focus on all single-family residential building types of the inner city. Discrete data on buildings is obtained through combined information products derived from very high resolution (VHR) satellite imagery, field surveys, and computer assisted personal interviewing. In total, 509 valid geocoded survey datasets were used to evaluate and model household electricity consumption, as well as electrical appliance ownership. The study's findings reveal that the arithmetic mean of specific electricity consumption was 3.66 kWh per household per day and 345 kWh per capita per year in 2015. By subdividing the data into distinct building types as well as their spatial location, and weighting the specific values according to their proportion in the study area, a more accurate mean value of 1.88 kWh per household per day and 160 kWh per capita per year was obtained. Applying this weighted mean to extrapolate household electricity consumption for the study area, in conjunction with the sample's precision level, resulted in an estimate of 126–137 GWh for the year 2015. In contrast, using the arithmetic mean would have led to values twice as high, even exceeding the total electricity consumption of the entire city, including multi-family and non-residential buildings. The study highlights the significance of on-site data collection combined with geospatial mapping techniques in enhancing of understanding of residential energy systems. Using building types as indicators to distinguish between households with contrasting electricity consumption and electrical appliance load levels can address the challenges posed by rapid urban growth in the Global South. This proposed method can assist municipal administrations in establishing a database that can be updated resource-efficiently at regular intervals by acquiring new satellite images.

Application of attention mechanism enhanced neural network in non-invasive load monitoring of industrial power data

The non-intrusive appliance load monitoring (NILM) decomposes the total power consumption of a power system into its contributing appliances. Previous studies only considered using the total power consumption information of appliances to decompose the load consumption. Besides the total electricity consumption, there is also important information such as current, voltage, and time in the total electricity consumption data, which can be used to analyze the load consumption information. Therefore, we proposed a sequence-to-sequence network enhanced by an attention mechanism, which effectively integrated the external features besides the total electricity consumption in grid data. Finally, we applied and evaluated the proposed model on the electricity consumption data of a gas station with 12 appliances, and our model achieved a 90.5% accuracy in load decomposition. Our solution provides a new solution on the application of NILM in the industrial field and helps to manage energy more rationally.

Energy Internet-Based Load Shifting in Smart Microgrids: An Experimental Study

This study investigated a grid-connected smart microgrid (MG) system integrating solar photovoltaic (PV) panels and a battery energy storage system (BESS) as distributed energy resources (DERs) to locally serve residential loads. The load-shifting demand-side management (DSM) technique was employed to effectively manage the load appliances. The proposed load-shifting algorithm relies on minimum price incentives to allow customers to allocate their load appliances economically during minimum price periods. The algorithm considers the waiting times and minimum tariff periods for appliances, calculates precise operating durations for each appliance, and prioritizes powering the appliances from the MG first, followed by the main grid. The system comprises two non-shiftable and three shiftable loads. When the MG power is insufficient to activate all shiftable loads, the system transfers the remaining unsupplied shiftable appliances to periods with low-priced energy. The Energy Internet concept is adopted to manage energy and monitor usage when a customer is unable to check the accuracy of their energy meter by supervising the system’s features on-site. The proposed comprehensive system enables load management, continuous monitoring, customer awareness, and energy cost saving. Six cases were studied, both numerically and experimentally, with varying MG power generation and load pre-scheduling periods, with and without DSM application. In all adopted cases, the implemented system save energy costs by at least 50%.

Stochastic bottom-up load profile generator for Canadian households’ electricity demand

The residential energy demand timing and magnitude are highly impacted by occupants' behaviors and activities. However, acquiring a reliable data source for these activities is a vital challenge, especially on an urban scale. In this context, this paper presents a stochastic bottom-up model for generating electrical loads for residential buildings in Canada. The proposed model is developed to investigate the impact of different household characteristics, appliance stock, and energy behaviors on the timing and magnitude of non-HVAC energy loads at individual or multiple houses. The proposed tool includes four main modules for generating stochastic profiles of occupancy, lighting demand, appliance load, and domestic hot water demand (DWH). The model is calibrated using the Canadian time use survey (TUS), energy use statistics, and appliance ownership surveys. The model is scalable and can be extended to serve various applications by adding new modules and data sources in the future. This paper presents the model development methodology, generated high-resolution load profiles, and validation in comparison with actual measurements. Finally, the model is used for studying the impact of household characteristics on total energy use. Future work will include incorporating this model into a comprehensive agent-based model for designing and testing effective demand response programs.

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

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

A Reinforcement Learning Approach for Integrating an Intelligent Home Energy Management System with a Vehicle-to-Home Unit

These days, users consume more electricity during peak hours, and electricity prices are typically higher between 3:00 p.m. and 11:00 p.m. If electric vehicle (EV) charging occurs during the same hours, the impact on residential distribution networks increases. Thus, home energy management systems (HEMS) have been introduced to manage the energy demand among households and EVs in residential distribution networks, such as a smart micro-grid (MG). Moreover, HEMS can efficiently manage renewable energy sources, such as solar photovoltaic (PV) panels, wind turbines, and vehicle energy storage. Until now, no HEMS has intelligently coordinated the uncertainty of smart MG elements. This paper investigated the impact of PV solar power, MG storage, and EVs on the maximum solar radiation hours. Several deep learning (DL) algorithms were utilized to account for the uncertainties. A reinforcement learning home centralized photovoltaic (RL-HCPV) scheduling algorithm was developed to manage the energy demand between the smart MG elements. The RL-HCPV system was modelled according to several constraints to meet household electricity demands in sunny and cloudy weather. Additionally, simulations demonstrated how the proposed RL-HCPV system could incorporate uncertainty, and efficiently handle the demand response and how vehicle-to-home (V2H) can help to level the appliance load profile and reduce power consumption costs with sustainable power production. The results demonstrated the advantages of utilizing RL and V2H technology as potential smart building storage technology.

A new optimized demand management system for smart grid-based residential buildings adopting renewable and storage energies

Demand Side Management (DSM) implies intelligently managing load appliances in a Smart Grid (SG). DSM programs help customers save money by reducing their electricity bills, minimizing the utility’s peak demand, and improving load factor. To achieve these goals, this paper proposes a new load shifting-based optimal DSM model for scheduling residential users’ appliances. The proposed system effectively handles the challenges raised in the literature regarding the absence of using recent, easy, and more robust optimization techniques, a comparison procedure with well-established ones, using Renewable Energy Resources (RERs), Renewable Energy Storage (RES), and adopting consumer comfort. This system uses recent algorithms called Virulence Optimization Algorithm (VOA) and Earth Worm Optimization Algorithm (EWOA) for optimally shifting the time slots of shiftable appliances. The system adopts RERs, RES, as well as utility grid energy for supplying load appliances. This system takes into account user preferences, timing factors for each appliance, and a pricing signal for relocating shiftable appliances to flatten the energy demand profile. In order to figure out how much electricity users will have to pay, a Time Of Use (TOU) dynamic pricing scheme has been used. Using MATLAB simulation environment, we have made effectiveness-based comparisons of the adopted optimization algorithms with the well-established meta-heuristics and evolutionary algorithms (Genetic Algorithm (GA), Cuckoo Search Optimization (CSO), and Binary Particle Swarm Optimization (BPSO) in order to determine the most efficient one. Without adopting RES, the results indicate that VOA outperforms the other algorithms. The VOA enables 59% minimization in Peak-to-Average Ratio (PAR) of consumption energy and is more robust than other competitors. By incorporating RES, the EWOA, alongside the VOA, provides less deviation and a lower PAR. The VOA saves 76.19% of PAR, and the EWOA saves 73.8%, followed by the BPSO, GA, and CSO, respectively. The electricity consumption using VOA and EWOA-based DSM cost 217 and 210 USD cents, respectively, whereas non-scheduled consumption costs 273 USD cents and scheduling based on BPSO, GA, and CSO costs 219, 220, and 222 USD cents.

A novel technique for stand-alone hydropower plant

Stand-alone hydropower plant is a major energy source in grid isolated zone because they do not require dams to be built. In a stand-alone hydropower plant Voltage and frequency, instability is caused by the load variation. This can cause the alternator to spin very quickly or very slowly, causing the induced voltage and output frequency to rise or fall. Furthermore, if the rate of alternator growth is too high or too low, the load appliances connected to the generator can undergo huge damage. For optimal overall performance of a small hydropower plant, generated voltage and frequency must be within the required range. To overcome these issues, a system is discussed in this paper. The proposed system not only stabilizes stand-alone hydropower plant output through an electronic load control system but is also easy and price-effective. Electronic load controller managed small hydro energy generation that takes into vital parameters, i.e, stabilization of both voltage and frequency despite variable consumer load. The presented technique consists of PID and fuzzy logic control system and also a couple of DC motors interconnected with an H-bridge to control the water valve known as a gate. The proposed control system also consists of several dummy loads to dump the extra generated power to keep the output power constant and to overcome frequency variations. For increased efficiency and life span of the generator, dummy loads used are switched only at zero crossings of a sinusoidal signal. The proposed system is implemented on both Simulink/MATLAB and Proteus. The simulation results prove that the presented technique overcomes the challenges of stand-alone hydropower plants.

Efficient deep generative model for short-term household load forecasting using non-intrusive load monitoring

Home energy management systems (HEMS) enable key strategies and methods to improve residential efficiency and energy utilization. To make informed decisions, HEMS depend on energy monitoring and forecasting systems. However, short-term load forecasting (STLF) at the household level is challenging due to the high uncertainty in load demand caused by the difficulty of expecting customer behavior. Traditional solutions only use household load data at low temporal resolution and cannot adequately consider the users’ power consumption profile, limiting prediction accuracy. This paper addresses these issues and proposes a two-stage approach for the household STLF that leverages past and future appliance load data. In the first stage, the past and future appliance load data is estimated from the aggregate load measurement through a non-intrusive load monitoring (NILM) approach based on a deep generative model using variational autoencoders (VAE). This granular information is then augmented through a novel household load predictive model. We implement six configurations combining three stage-1 NILM models with two stage-2 predictive models based on VAE and temporal convolutional networks. Given the volatile nature of residential loads, identifying the running appliances and their consumption benefits the household STLF. The proposed approach is designed to leverage short-term appliance load forecasts rather than being limited to historical features, which is a key advantage, especially when multi-state appliances are used. The proposed approach is compared to state-of-the-art approaches on the UK-DALE and REFIT datasets and yields competitive results with improvements of 16% and 19% on average for the MAE and MAPE metrics, respectively.

A novel dynamic load-priority-based scheduling strategy for home energy management system

In order to increase renewable energy consumption and enhance energy efficiency, residents participate in the utility in the form of demand response. To address the challenges of imprecise energy efficiency services and inadequate response of household energy use, a novel dynamic-load-priority-based scheduling strategy is proposed. Firstly, according to the operation law of household appliances, the control models of three kinds of flexible loads are established respectively. Secondly, according to the characteristics of household appliances, the priority evaluation index of household appliances including user comfort, electricity cost and task urgency index is proposed. The Technique for Order Preference by Similarity to Ideal Solution based on entropy weight is used to prioritize the household appliances that control the scheduling queue. Finally, according to the sorting results, the load control strategy is implemented to reasonably arrange the orderly work of household electrical equipment. Compared with the uncontrolled state, the proposed strategy can limit the total power of appliance load operation to the demand range in the peak load condition, and save 3% of electricity cost and reduce 5.4 kW of peak load under the demand limit of 12 kW, and 32% of electricity cost and 10.6 kW of peak load under the demand limit of 7 kW, without affecting the basic life of customers. In the period of new energy saturation, the load can also be dispatched to participate in the new energy consumption, thus realizing the coordinated interaction between residential energy consumption and the utility.

Smart Energy Management and Load Monitoring of Individual Loads

As India continues to urbanise with its manufacturing sector growing, its energy demands have increased rapidly. In this period, the need for energy monitoring and conservation must be evident. So the consumers can monitor their load consumption by a digital wattmeter. But the digital wattmeter measures and shows the total power consumed by consumer appliances. The consumer has no idea about which appliance consumes more power. Also, the consumer will not be able to know whether any appliance is getting damaged. To resolve these issues, we have come up with this project which is an IOT-based energy monitoring and controlling system, used to measure the power consumed by each appliance (Load). The power consumed by each appliance can be monitored individually by the consumer and it also informs us if there is an issue through a web application. Also the web application stores the historical data of the power consumed by each load. An Internet of Things (IoT) technique was employed to make the system smart. A microcontroller with built-in Wi-Fi - ESP32 - was used to monitor and control each load using a web application.

Attention-Based Multitask Probabilistic Network for Nonintrusive Appliance Load Monitoring

Monitoring individual appliances’ operating state and energy consumption in a building enables significant energy-saving opportunities. These days, smart meters perform this task non-intrusively using sophisticated signal processing, machine learning, or deep learning approaches. To this end, this paper proposes a novel multi-task deep learning model that uses readily available low-frequency energy data from the smart meter for simultaneous appliance state detection and energy disaggregation. The model creatively adopts and customizes the famous transformer model from the field of language modeling for the above task. Further, the model output is produced as a mixture of probability density functions to handle uncertainties. The model performance is evaluated using the publicly available REFIT and UKDALE datasets. The test results indicate the proposed model’s superiority, generalizability, and transferability compared to other state-of-the-art models.