Load Monitoring Method Research Articles

STFT Cluster Analysis for DC Pulsed Load Monitoring and Fault Detection on Naval Shipboard Power Systems

Current trends in Naval shipboard power system architecture indicate that the electrification of future warships is inevitable, and it will be equipped with loads that draw periodic pulsed currents from the dc microgrid or have large transients while switching state. In order to monitor the operation of those loads, solely time-based features are not enough as they do not provide sufficiently unique information to differentiate various transient stages of the load profile. The focus of more recent research has been on extracting time–frequency features. However, no comprehensive solution exists that could work for any general load profile. The proposed load monitoring and fault detection method presented in this article outlines a data clustering-based approach to extract unique feature vectors from short-time Fourier transform analysis for any pulsed load. These features can then be used to identify various events in the load transient as well as shunt faults and series arcing faults. Implementation and performance of the scheme for several load profiles and fault scenarios are included.

An Event-Driven Convolutional Neural Architecture for Non-Intrusive Load Monitoring of Residential Appliance

Nowadays, the advancement of non-intrusive load monitoring (NILM) is hastened by the everincreasing requirements for smart power utilization and demand side management. Thus, an intelligent event-driven non-intrusive load monitoring method based on convolutional neural network (CNN) is proposed in this article to profile residential consumer behavior, which focused on limited applicability and low recognition accuracy of existing NILM techniques. The deep convolutional neural architecture improves the performance of load monitoring procedure, and extends its application range to monitoring process in terms of various and combined characteristics. Meanwhile, the event-driven procedure is conducted to find the start-stop time and state change of the appliance accurately, including zero-cross detection, current similarity detection, threshold evaluation, and event current acquisition. Subsequently, the current-to-image conversion is carried out to represent the characteristics of residential appliance as the input of CNN. And the convolutional neural architecture is proposed to identify the power consumption load and its working condition. Selected experimental results are performed that the average recognition accuracy is over 93%, which demonstrates the effectiveness and superiority of the proposed methodology for home energy management.

Nonintrusive on-site load-monitoring method with self-adaption

Nonintrusive load monitoring (NILM) is an effective way to measure the demand side of electricity consumption. An on-site NILM method in high-frequency acquisition mode was explored, which makes the entire process, including the construction of load-waveform decomposition and a short-term dynamic signature database, as well as continuous on-site load identification, automatic and executable in real time. First, a load decomposition model was established based on the additivity principle of load current to obtain the independent load waveform. According to the operational signatures of the load, there was no need to obtain prior data by conducting a preliminary experiment. The load type was judged by Bayesian classification model, and a dynamic load signature database was adaptively built for independent users. Based on the dynamic signature database, load identification was realized by optimization model, obtaining the electricity consumption of load in real time. The effectiveness of the method was verified by measuring electricity-consumption data. According to the experiment, the method can be automatically executed on-site to adapt to different users, and the dynamic signature database established improves the weak universality caused by establishing the database in advance. The fast optimization based on the signature database ensures the identification is efficient and accurate.

Non-intrusive load monitoring and decomposition method based on decision tree

In order to realize the problems of non-intrusive load monitoring and decomposition (NILMD) from two aspects of load identification and load decomposition, based on the load characteristics of the database, this paper firstly analyzes and identifies the equipment composition of mixed electrical equipment group by using the load decision tree algorithm. Then, a 0–1 programming model for the equipment status identification is established, and the Particle Swarm Optimization (PSO) is used to solve the model for equipment state recognition, and the equipment operating state in the equipment group is identified. Finally, a simulation experiment is carried out for the partial data of Question A in the 6th “teddy cup” data mining challenge competition.

An improved non-intrusive load disaggregation algorithm and its application

The non-intrusive load monitoring (NILM) method is proposed to acquire the energy consumption of appliances in a building. Steady-state current decomposition is one of the most effective and applicable methods in NILM. Although many load features and decomposing models have been developed in the previous work, the harmonic vectors have been rarely discussed. In this study, weighted current harmonic vectors are proposed to increase the weight of the useful harmonic. The harmonic vectors are calculated geometrically so that all the information in features can be retained. A multi-objective particle swarm optimization based model is built to carry out the disaggregation task, where both summation and standard deviation of the errors are considered as objective functions. Moreover, to make the public datasets available in current decomposition, a current superimposing method is proposed. It generates aggregated currents from the currents of appliances running independently. Finally, the proposed model and two contrast models are performed on the WHITED dataset. The experimental results indicate that the proposed method is more robust and has higher current disaggregation precision than the involved comparing methods.

A new approach of electrical appliance identification in residential buildings

This paper proposes a simple algorithm about non-intrusive appliance load monitoring (NIALM) method. The main objective is to analyze the overall power consumption of a given building and to identify the different operating appliances. This approach aims to reduce the overall energy expense of maintaining a specified level of comfort. In our approach, we firstly replace the main signal by a shorter form in order to reduce computing time. This criterion is important to guarantee real-time operation mode. Furthermore, we can classify the operating devices through their type and the mean electrical power consumed. Finally, for identification, we use the template’s waveform matching to identify the individual energy consumption with an optimized manner. To validate the proposed algorithm, satisfactory simulation results showing the reliability of the proposed NIALM method are found.

A nonintrusive power load monitoring method using coupled allocation mechanism

A nonintrusive power load monitoring method using coupled allocation mechanism

An index concentration method for suspended load monitoring in large rivers of the Amazonian foreland

Abstract. Because increasing climatic variability and anthropic pressures have affected the sediment dynamics of large tropical rivers, long-term sediment concentration series have become crucial for understanding the related socioeconomic and environmental impacts. For operational and cost rationalization purposes, index concentrations are often sampled in the flow and used as a surrogate of the cross-sectional average concentration. However, in large rivers where suspended sands are responsible for vertical concentration gradients, this index method can induce large uncertainties in the matter fluxes. Assuming that physical laws describing the suspension of grains in turbulent flow are valid for large rivers, a simple formulation is derived to model the ratio (α) between the depth-averaged and index concentrations. The model is validated using an exceptional dataset (1330 water samples, 249 concentration profiles, 88 particle size distributions and 494 discharge measurements) that was collected between 2010 and 2017 in the Amazonian foreland. The α prediction requires the estimation of the Rouse number (P), which summarizes the balance between the suspended particle settling and the turbulent lift, weighted by the ratio of sediment to eddy diffusivity (β). Two particle size groups, fine sediments and sand, were considered to evaluate P. Discrepancies were observed between the evaluated and measured P, which were attributed to biases related to the settling and shear velocities estimations, but also to diffusivity ratios β≠1. An empirical expression taking these biases into account was then formulated to predict accurate estimates of β, then P (ΔP=±0.03) and finally α. The proposed model is a powerful tool for optimizing the concentration sampling. It allows for detailed uncertainty analysis on the average concentration derived from an index method. Finally, this model could likely be coupled with remote sensing and hydrological modeling to serve as a step toward the development of an integrated approach for assessing sediment fluxes in poorly monitored basins.

An online non-intrusive load monitoring method based on Hidden Markov model

Non-intrusive load monitoring (NILM) can decompose the total power consumption measured by the smart meter into the power consumed by the individual appliances, so as to achieve the purpose of saving energy. In this paper, an improved method of Daubechies9 (DB9) which is a discrete wavelet is proposed, which can effectively remove the noise of the low-frequency components. On this basis, an online NILM method based on Hidden Markov model (HMM) is proposed. The model of load switching can be built using apparent power of transient-state with this method. Besides, the improved forward algorithm which effectively suppressing the data underflow in load classification is proposed. The proposed methods are embedded in the smart meter and can increase the overall recognition rate of the load over 90% in the experiments which prove that they have good applicability.

Appliance Flexibility Analysis Considering User Behavior in Home Energy Management System Using Smart Plugs

The home energy management system (HEMS) plays an important role in enabling residential customers to participate in demand response (DR) programs autonomously via appliance-level dispatch with smart plugs, and the flexibility analysis could help the HEMS to dispatch appliances efficiently without harming user's comfort level. This paper presents a bottom-up method to obtain the flexibility of appliances. With the use of smart plugs in the HEMS, the type of the appliance attached to each smart plug is inferred with consideration of measurement uncertainty as well as user behavior. Then, the flexibility of an appliance is analyzed based on the extrapolated power consumption of the appliance as well as the owner's usage behavior. In the case study, the types of appliances recognized for individual smart plugs as well as the corresponding flexibilities are used as reference for appliance dispatching in the HEMS under various DR applications. The case study demonstrates the performance of the proposed appliance load monitoring method and the application of the proposed flexibility analysis.

Incorporating Appliance Usage Patterns for Non-Intrusive Load Monitoring and Load Forecasting

This paper proposes a novel non-intrusive load monitoring (NILM) method which incorporates appliance usage patterns (AUPs) to improve performance of active load identification and forecasting. In the first stage, the AUPs of a given residence were learned using a spectral decomposition based standard NILM algorithm. Then, learnt AUPs were utilized to bias the priori probabilities of the appliances through a specifically constructed fuzzy system. The AUPs contain likelihood measures for each appliance to be active at the present instant based on the recent activity/inactivity of appliances and the time of day. Hence, the priori probabilities determined through the AUPs increase the active load identification accuracy of the NILM algorithm. The proposed method was successfully tested for two standard databases containing real household measurements in USA and Germany. The proposed method demonstrates an improvement in active load estimation when applied to the aforementioned databases as the proposed method augments the smart meter readings with the behavioral trends obtained from AUPs. Furthermore, a residential power consumption forecasting mechanism, which can predict the total active power demand of an aggregated set of houses, 5 min ahead of real time, was successfully formulated and implemented utilizing the proposed AUP based technique.

Uma revisão bibliométrica sobre a identificação de cargas similares em Smart Grid

Diante da necessidade de uso consciente dos recursos energéticos para o crescimento sustentável, o conceito Smart Grid tem servido de referência para o uso eficiente de tais recursos, sobretudo o elétrico. Entretanto, a aplicação de tal conceito nas instalações elétricas requer o desenvolvimento de dispositivos que permitam o monitoramento autônomo de aparelhos eletroeletrônicos em tais instalações. Este artigo tem como objetivo realizar uma revisão bibliométrica sobre os métodos não-intrusivo de monitoramento de cargas, a fim de mostrar o estado da arte dessa temática, isto é, os principais avanços e entraves para o desenvolvimento dessa tecnologia. É realizada uma análise quantitativa e qualitativa dos trabalhos referentes ao tema buscados via base de dados Web of Science. Como resultado de tal investigação, observa-se a relevância dos estudos nessa área, que está em pleno crescimento. Cita-se ainda o fato de o tema estar sendo investigado tanto em países desenvolvidos, como Canadá e Estados Unidos, quanto em países em desenvolvimento, como Malásia e Índia, o que mostra que existe uma preocupação mundial em se avançar em tal problema que independe da situação econômica da nação.

Detection of unidentified appliances in non-intrusive load monitoring using siamese neural networks

Non-intrusive load monitoring methods aim to disaggregate the total power consumption of a household into individual appliances by analyzing changes in the voltage and current measured at the grid connection point of the household. The goal is to identify the active appliances, based on their unique fingerprint. Most state-of-the-art classification algorithms rely on the assumption that all events in the data stream are triggered by known appliances, which is often not the case. This paper proposes a method capable of detecting previously unidentified appliances in an automated way. For this, appliances represented by their VI trajectory are mapped to a newly learned feature space created by a siamese neural network such that samples of the same appliance form tight clusters. Then, clustering is performed by DBSCAN allowing the method to assign appliance samples to clusters or label them as ‘unidentified’. Benchmarking on PLAID and WHITED shows that an F1,macro-measure of respectively 0.90 and 0.85 can be obtained for classifying the unidentified appliances as ‘unidentified’.

Energy consumption analysis of residential swimming pools for peak load shaving

Peak load shaving is a very important issue, however, most of peak load shaving methods either require extra investments or reduce the comfort of the consumers. This paper analyzes the impact of circulating pumps of residential swimming pools on the peak load, and shows that shifting the active time of circulating pumps of residential swimming pools could shave the peak load without requiring extra investments or reducing the comfort of the consumers. First, based on an extensive dataset containing hourly energy consumption readings of 1005 residents during March 2011 and October 2012 in South Ontario, this paper analyzes the features of the energy consumptions of residents with and without swimming pools. Second, this paper proposes a novel non-intrusive appliance load monitoring method to estimate the energy consumption of circulating pumps of residential swimming pools. The advantages of the proposed method are that, compared with other non-intrusive appliance load monitoring methods, it does not require high sampling rate data or prior information of the appliance, therefore the cost of implementation is reduced and users’ privacy is protected. Finally, this paper shows that, the average hourly energy consumption of CPRSP is 0.7429 kW h. During the peak hour, circulating pumps of residential swimming pools contributes 20.11% energy consumption of residents with swimming pools, as well as 8.79% peak load of all neighborhoods. When considering minimizing the peak load, by postponing circulating pumps of residential swimming pools for 8 h rather than turn them off, the peak load can be shaved by 4.64%. When considering minimizing the cost of circulating pumps of residential swimming pools, the peak load can be shaved by 3.15% by postponing circulating pumps of residential swimming pools for 6 h, meanwhile the peak hour is postponed from 18:00 to 19:00.

Training load monitoring in elite English soccer: a comparison of practices and perceptions between coaches and practitioners

ABSTRACTPurpose: To survey practices and perceptions of training load monitoring among soccer coaches and practitioners.Methods: A questionnaire assessed factors influencing training planning, training load practices, and training load feedback and usefulness. The questionnaire was distributed via email and as an online version (Bristol Online Survey Tool) to relevant staff working within elite English Soccer.Results: Respondents represented two groups; those involved with player tactical (coach, n = 94) or physical (practitioner, n = 88) preparation. Coaches worked predominantly with younger players at lower standing clubs while practitioners worked with older players at higher standard clubs. With exception for the influence of current match schedule in training planning, there was coach-practitioner agreement for all training planning questions. There was agreement on some purposes for training load monitoring (maximise fitness, evaluate training) but not others (enhance fitness, reduce injury). For load monitoring methods, the greatest proportion of coach answers was for coach perception (22%); whereas the greatest proportion of practitioner responses was for GPS (22%). Largely, load reports were perceived positively and 84.1% of respondents felt training load monitoring was beneficial to their club.Conclusion: This survey shows coaches and practitioners perceive training load monitoring as worthwhile, with differences in practices and perceptions likely reflecting club infrastructure.

Session-RPE Method for Training Load Monitoring: Validity, Ecological Usefulness, and Influencing Factors.

Purpose: The aim of this review is to (1) retrieve all data validating the Session-rating of perceived exertion (RPE)-method using various criteria, (2) highlight the rationale of this method and its ecological usefulness, and (3) describe factors that can alter RPE and users of this method should take into consideration.Method: Search engines such as SPORTDiscus, PubMed, and Google Scholar databases in the English language between 2001 and 2016 were consulted for the validity and usefulness of the session-RPE method. Studies were considered for further analysis when they used the session-RPE method proposed by Foster et al. in 2001. Participants were athletes of any gender, age, or level of competition. Studies using languages other than English were excluded in the analysis of the validity and reliability of the session-RPE method. Other studies were examined to explain the rationale of the session-RPE method and the origin of RPE.Results: A total of 950 studies cited the Foster et al. study that proposed the session RPE-method. 36 studies have examined the validity and reliability of this proposed method using the modified CR-10.Conclusion: These studies confirmed the validity and good reliability and internal consistency of session-RPE method in several sports and physical activities with men and women of different age categories (children, adolescents, and adults) among various expertise levels. This method could be used as “standing alone” method for training load (TL) monitoring purposes though some recommend to combine it with other physiological parameters as heart rate.

Appliance classification using VI trajectories and convolutional neural networks

Non-intrusive load monitoring methods aim to disaggregate the total power consumption of a household into individual appliances by analysing changes in the voltage and current measured at the grid connection point of the household. The goal is to identify the active appliances, based on their unique fingerprint. An informative characteristic to attain this goal is the voltage–current trajectory. In this paper, a weighted pixelated image of the voltage–current trajectory is used as input data for a deep learning method: a convolutional neural network that will automatically extract key features for appliance classification. The macro-average F-measure is 77.60% for the PLAID dataset and 75.46% for the WHITED dataset.

Non-intrusive load monitoring under residential solar power influx

This paper proposes a novel Non-Intrusive Load Monitoring (NILM) method for a consumer premises with a residentially installed solar plant. This method simultaneously identifies the amount of solar power influx as well as the turned ON appliances, their operating modes, and power consumption levels. Further, it works effectively with a single active power measurement taken at the total power entry point with a sampling rate of 1Hz. First, a unique set of appliance and solar signatures were constructed using a high-resolution implementation of Karhunen Loéve expansion (KLE). Then, different operating modes of multi-state appliances were automatically classified utilizing a spectral clustering based method. Finally, using the total power demand profile, through a subspace component power level matching algorithm, the turned ON appliances along with their operating modes and power levels as well as the solar influx amount were found at each time point. The proposed NILM method was first successfully validated on six synthetically generated houses (with solar units) using real household data taken from the Reference Energy Disaggregation Dataset (REDD) - USA. Then, in order to demonstrate the scalability of the proposed NILM method, it was employed on a set of 400 individual households. From that, reliable estimations were obtained for the total residential solar generation and for the total load that can be shed to provide reserve services. Finally, through a developed prediction technique, NILM results observed from 400 households during four days in the recent past were utilized to predict the next day’s total load that can be shed.

Noncontact electrical system monitoring on a U.S. Coast Guard cutter

Modernization in the U.S. Navy and U.S. Coast Guard (USCG) includes an emphasis on automation systems to help replace manual tasks and reduce crew sizes. This places a high reliance on monitoring systems to ensure proper operation of equipment and to maintain safety at sea. Recently developed noncontact current and voltage sensors, [1] combined with nonintrusive load monitoring (NILM) methods [2], provide a nonintrusive, low-cost, and easily installed package for machinery monitoring. This paper presents an application case study using these NILM enabled sensors installed on the main electrical feeders of the USCG Famous Class Cutter SPENCER. The system records the power demand on the ship and disaggregates this demand by identifying transients corresponding to loads changing states, e.g., a pump turning on. Results of this study showcase these novel sensors' ability to monitor both generation and load side equipment while at sea or in port and provide information useful for tracking operation schedules, energy usage, and maintenance needs.

A Novel Nonintrusive Transient Event Detection Based on the Current

Load monitoring is of great significance to the power system. Traditional load monitoring requires the installation of sensors and other hardware devices at each monitored load. This intrusive load monitoring method requires a lot of manpower, material and financial resources. For this reason, a non-intrusive load monitoring method is proposed, in which the load operation can be analyzed by installing the monitoring equipment at the power inlet and monitoring the current and voltage signals. Therefore, noninvasive power load monitoring is one of the important development directions of power load monitoring in the future. Different power loads typically exhibit different characteristics in switching and cutting. Accordingly, we propose an automatic detection algorithm for transient events based on difference of current intensity, it can according to the identification of the load current signal accurately switching occurs at the moment and moment load shedding occurred and trigger the relevant program to these time points recorded, and then the application of the procedure for further processing