Search

Non-intrusive load monitoring (NILM) is the process of obtaining appliance-level data from a single metering point, measuring total electricity consumption of a household or a business. Appliance-level data can be directly used for demand response applications and energy management systems as well as for awareness raising and motivation for improvements in energy efficiency. Recently, classical machine learning and deep learning (DL) techniques became very popular and proved as highly effective for NILM classification, but with the growing complexity these methods are faced with significant computational and energy demands during both their training and operation. In this paper, we introduce a novel DL model aimed at enhanced multi-label classification of NILM with improved computation and energy efficiency. We also propose an evaluation methodology for comparison of different models using data synthesized from the measurement datasets so as to better represent real-world scenarios. Compared to the state-of-the-art, the proposed model has its energy consumption reduced by more than 23% while providing on average approximately 8 percentage points in performance improvement when evaluating on data derived from REFIT and UK-DALE datasets. We also show a 12 percentage point performance advantage of the proposed DL based model over a random forest model and observe performance degradation with the increase of the number of devices in the household, namely with each additional 5 devices, the average performance degrades by approximately 7 percentage points.

The successful implementation of Renewable Energy Communities (RECs) involves maximizing the self-consumption within a community, particularly in regulatory contexts in which shared energy is incentivized. In many countries, the absence of a metering infrastructure that provides data at an hourly or sub-hourly resolution level for low-voltage users (e.g., residential and commercial users) makes the design of a new energy community a challenging task. This study proposes a non-intrusive machine learning methodology that can be used to generate residential electrical consumption profiles at an hourly resolution level using only monthly consumption data (i.e., billed energy), with the aim of estimating the energy shared by RECs. The proposed methodology involves three phases: first, identifying the typical load patterns of residential users through k-Means clustering, then implementing a Random Forest algorithm, based on monthly energy bills, to identify typical load patterns and, finally, reconstructing the hourly electrical load profile through a data-driven rescaling procedure. The effectiveness of the proposed methodology has been evaluated through an REC case study composed by 37 residential users powered by a 70 kWp photovoltaic plant. The Normalized Mean Absolute Error (NMAE) and the Normalized Root Mean Squared Error (NRMSE) were evaluated over an entire year and whenever the energy was shared within the REC. The Relative Absolute Error was also measured when estimating the shared energy at both a monthly (MRAE) and at an annual basis. (RAE). A comparison between the REC load profile reconstructed using the proposed methodology and the real load profile yielded an overall NMAE of 20.04 %, an NRMSE of 26.17 %, and errors of 18.34 % and 23.87 % during shared energy timeframes, respectively. Furthermore, our model delivered relative absolute errors for the estimation of the shared energy at a monthly and annual scale of 8.31 % and 0.12 %, respectively.

There is a growing demand for Green AI (Artificial Intelligence) technologies in the market and society, as it emerges as a promising technology. Green AI technologies are used to create sustainable solutions and reduce the environmental impact of AI. This paper focuses on describing the services of Green AI and the challenges associated with it at the community level. This article also highlights the accuracy levels of machine learning algorithms for various time periods. The process of choosing the appropriate input parameters for weather, locations, and complexity is outlined in this paper to examine the ML algorithms. For correcting the algorithm performance parameters, metrics like RMSE (root mean square error), MSE (mean square error), MAE (mean absolute error), and MPE (mean percentage error) are considered. Considering the performance and results of this review, the LSTM (long short-term memory) performed well in most cases. This paper concludes that highly advanced techniques have dramatically improved forecasting accuracy. Finally, some guidelines are added for further studies, needs, and challenges. However, there is still a need for more solutions to the challenges, mainly in the area of electricity storage. [ABSTRACT FROM AUTHOR]

This paper proposes a decision support model to define electricity consumers’ remuneration structures when providing consumption flexibility, optimized for different load regimes. The proposed model addresses the remuneration of consumers when participating in demand response programs, benefiting or penalizing those who adjust their consumption when needed. The model defines dynamic remuneration values with different natures for the aggregator (e.g. flexibility aggregator or curtailment service provider) and for the consumer. The preferences and perspective of both are considered, by incorporating variables that represent the energy price, the energy production and the flexibility of consumers. The validation is performed using real data from the Iberian market, and results enable to conclude that the proposed model adapts the remuneration values in a way that it is increased according to the consumers’ elastic, while being reduced when the generation is higher. Consequently, the model boosts the active consumer participation when flexibility is required, while reaching a solution that represents an acceptable g tradeoff between the aggregators and the consumers.

Smart Energy Applications are particularly impacting, especially due to energy resource scarcity and its high associated costs. Smart management of energy consumption derives both from the user lifestyle, in terms of efficient and responsible behaviors, and from automatic algorithms that control and counteract energy waste and inefficient management. Focusing the attention on the latter, the development of methodologies and well-working techniques to monitor and optimize consumption often requires an important effort in long measurement campaigns to get raw data to work with. Whenever this should be too much expensive or proper instrumentation is unavailable, public datasets could solve the problem. The current literature review on the dataset availability showed a large presence of information, especially related to electrical energy consumption. Nevertheless, several limitations affect them, from the low number of calculated electrical parameters (i.e. 4-5 in most cases) to short analysis periods, passing by the lack of detailed frequency domain information or poor consumption habit transitions analysis. Accordingly, this work aims to overcome current dataset limitations, by proposing a real-measurement based simulated dataset, extracting more than 400 discriminative electrical parameters on 36 different home appliances, discussing preliminary acquisition set-ups, simulation process, extracted electrical parameters and examples of applicability to smart energy applications. To provide a data quality index, a validation procedure has also been carried out, showing how simulated data match real acquisition with a reference measurement instrument. The produced dataset is available for downloading and analysis in public free access and its repository link is provided in the reference section.

The evolution of smart meters has led to the generation of high-resolution time-series data - a stream of data capable of unveiling valuable knowledge from consumption behaviours for different applications. The ability to extract hidden knowledge from such massive amounts of data requires that it be analysed intelligently. Hence, for a clear representation of the various consumption behaviours of consumers, a good number of data mining technologies are usually employed. This paper presents a systematic review of the various data mining techniques and methodologies employed while profiling energy data streams. The review identifies the strengths and shortcomings of existing data mining methods as applied in research, focusing more on data processing techniques and load clustering. Also discussed are data mining methods used to profile consumption data, their pros and cons. It was inferred during the research that the choice of data mining technique employed is highly dependent on the application it is intended for and the intrinsic nature of the dataset.

There is a growing demand for Green AI (Artificial Intelligence) technologies in the market and society, as it emerges as a promising technology. Green AI technologies are used to create sustainable solutions and reduce the environmental impact of AI. This paper focuses on describing the services of Green AI and the challenges associated with it at the community level. This article also highlights the accuracy levels of machine learning algorithms for various time periods. The process of choosing the appropriate input parameters for weather, locations, and complexity is outlined in this paper to examine the ML algorithms. For correcting the algorithm performance parameters, metrics like RMSE (root mean square error), MSE (mean square error), MAE (mean absolute error), and MPE (mean percentage error) are considered. Considering the performance and results of this review, the LSTM (long short-term memory) performed well in most cases. This paper concludes that highly advanced techniques have dramatically improved forecasting accuracy. Finally, some guidelines are added for further studies, needs, and challenges. However, there is still a need for more solutions to the challenges, mainly in the area of electricity storage.

With a significant deployment of smart meters across end-user platforms, the dynamic visibility of energy flow among the end-users has been increased significantly. The granular information of smart meters can be used to improve the load forecast accuracy and to influence energy consumption patterns with demand side management (DSM) schemes. This paper addresses the challenges of smart meter data size, complexity, variability and volatility for efficient use in load forecast and DSM. A novel clustering-based approach for analysis of smart meter data, aimed at more accurate and detailed load profiling, reduced profile complexity, improved load forecast accuracy and providing optimal DSM solutions is proposed. The proposed approach utilizes an advanced clustering algorithm to reduce the data size. The approach addresses data complexity, variability and volatility by linearizing the load profiles and minimizing the errors. The validity of the approach is demonstrated on an Irish smart meter dataset and on a simulated solar photovoltaic (PV) data and showed an improved load forecast accuracy, improved DSM solutions, and reduced computational burden. The improvements in the DSM solution are evidenced by a higher cost saving with a higher peak load reduction at the lower level of demand flexibility.

Nowadays, knowing load profiles of different customers in electricity market-oriented operation of power system is very important. Smart metering system provides more than enough information about it. However, even if there is a load measurement on each customer in the system, it is economically unjustified and it is time consuming to conduct a detailed analysis each of them. In this study, unlike literature approaches, the different industrial load profiles are observed. The fuzzy c-means clustering is applied to get typical load curves representing different type of industry. For that purpose, the detailed statistical analysis and calculations of load shape factors of the industrial customers for different period of the year have been carried out. Customer's 15-min load was collected by developing Advanced Metering Infrastructure in the power system of Bosnia and Herzegovina. According to available data, statistical analysis and applied fuzzy c-means clustering, three typical load profiles for different industrial customers are proposed. [ABSTRACT FROM AUTHOR]

The traditional power grid system is evolving toward a smart grid system, which will improve the user experience. However, this system is capable of generating high-dimensional data at a very high sample rate. A common technique for reducing high-dimensional data is dimensionality reduction. With this technique, we are able to reduce the data to a lower dimension, making it suitable for smart grid applications including transmission, storage, and visualization. Linear dimensionality reduction techniques are mostly explored in the context of smart grid applications. Due to the nonlinear nature of data generation in the smart grid, we anticipate that the nonlinear dimensionality reduction techniques can perform better. This work evaluates different nonlinear dimensionality reduction techniques and compares them with principal component analysis, which is a widely used linear dimensionality reduction technique in the smart grid environment. We use the visualization of load profile data and adjusted rank index (ARI) for comparison of dimensionality reduction techniques. Load profiling is an important task to complement the demand-side management and tariff selection. The visual depiction of the load profiles and ARI suggest that the nonlinear dimensionality reduction techniques perform better compared with linear dimensionality reduction techniques. [ABSTRACT FROM AUTHOR]

This article presents an integrated and intelligent price based self-scheduling approach for demand response (DR) aggregators in the wholesale market considering load profile attributes of aggregated loads. The load profile attributes namely utilization and availability factors are derived as a function of load type and temporal characteristics of responsive loads. The proposed integrated approach uses the concept of conventional random willingness factors coupled with load profile attributes to resolve the ambiguity around customer/load behavior. The response of the customer to load profile attributes and the willingness factor is formulated using three different models, namely, linear, exponential, and nonlinear models. Further, to consider the nondeterministic and indistinct behavior of the customer, a fuzzy inference system (FIS) is developed to create a relationship between the load profile attributes and willingness to DR participation cost. Further, an online FIS membership function parameter tuning mechanism is developed to improve the performance of DR aggregator as well as overall day ahead market. The proposed approach is simulated for a combined price based scheduling of the generation company and the DR aggregator with responsive loads spread over various load sectors such as industrial, commercial, residential, agricultural, and the municipal sector. The simulation results of the online intelligent-integrated framework are compared to conventional willingness model, nonfuzzy, and untuned FIS models with and without considering load profile and customer willingness factor. The comparison of the same demonstrates the effectiveness of the proposed online framework in improving the overall and DR aggregator surplus over other approaches. [ABSTRACT FROM AUTHOR]

Virtual power plants (VPPs) of secondary distribution network have come into existence due to the installation of generators in the premises of domestic consumers and the recent legislative emphasis on net‐zero energy operation. A predetermined operation of VPP as required by a system operator is possible only when its power output is assessed beforehand. A simple approach for assessment based on load profiling using descriptive statistical parameters such as z‐scores of the load data is proposed in this study. The proposed method collates historical customer data according to seasons, days of the week and specified time‐intervals of the day. Then, z‐scores are computed for each of the historical load and generator data. The calculated z‐scores are subsequently grouped into specified clusters. The grouped data in each of the clusters are then employed to predict the VPP output. Demonstration of the assessment of VPP output using available Pecan street data is presented to validate the proposed approach. [ABSTRACT FROM AUTHOR]

Now, Smart Energy Meters (SEMs) have become very popular in controlling and managing electricity consumption in residential, commercial, industrial sectors, to name a few. SEMs collects the fine-grained energy-consumption data at a regular time interval (every hour). Such a low sampling data rate can be useful in research, evaluating the higher consumption of electricity, etc. However, existing research work cannot adequately assess the qualitative, quantitative, and logical properties of the data that could theoretically improve the performance of classification and analytics. The purpose of this bibliometric analysis is to understand the reach of SEMs and its data analysis worldwide and across the various applications to make the concept clear for future researchers. The Scopus, Semantic Scholar and Crossref databases are used for performing the bibliometric analysis. This research study revealed that the insights of the SEMs data-enabled significant advancement in many fields such as energy consumption pattern analysis and prediction, demand response, load profiling, and direct-indirect phone charging analysis. [ABSTRACT FROM AUTHOR]

The utilization of energy consumption data is crucial for efficient operation and planning in smart grids. Nonetheless, certain obstacles need to be addressed, such as high computational costs, data security and privacy concerns, and significant expenses associated with installing smart meters across the electrical grid. To address these challenges, generating synthetic data has emerged as a promising approach, providing an opportunity to enhance energy efficiency, demand flexibility, and power grid operation. Therefore, this study proposes a nonlinear model of independent component estimation (NICE) with convolutional layers to produce realistic load profiles. This research aims to evaluate the potential of deep generative models (DGMs) through the characterization and quantification of electricity consumption profiles obtained from an actual smart grid on a university campus. The Kullback–Leibler divergence is used to evaluate the performance of the proposed model. Simulation results show that the proposed model can accurately capture the spatiotemporal correlation of actual samples, leading to synthetic load profiles that closely resemble actual profiles. The performance of the proposed NICE model is compared with a NICE model with dense layers, as well as with Generative Adversarial Networks (GAN) with dense layers, and GAN with convolutional layers (cGAN), all methods previously used in the literature to generate synthetic load profiles. It was observed that the proposed NICE model with convolutional layers leads to better results. This model produces more significant similarity between the probability distributions of actual and synthetic data, in addition to a more extraordinary ability to reproduce more realistic load variability curves. • A methodology for generating synthetic load profiles for the smart grids. • Convolutional NICE model for producing realistic load profiles. • Using deep generative models to obtain electricity consumption profiles. • Learning patterns from actual load profiles to reproduce synthetic load profiles. • Generating load profiles for buildings to assist in management and control. [ABSTRACT FROM AUTHOR]

Demand response exchange (DRX) presents a pool-based approach for optimal clearing/scheduling of demand response (DR) services through bid clearing. The same is intended to minimize the limitations of price-based and incentive-based mechanisms. However, unlike the price-based or incentive-based DR schemes, the DR sellers in DRX do not have a clear motivation to formulate the bid offers. Similarly, the DR buyers’ bid formulation should also include optimal benefit analysis. This paper formulates and analyzes the bid formulation for DR sellers and buyers considering respective operation objectives. The DR seller bid formulation is derived considering load behavior through utilization index and availability index obtained from load profile of respective loads. On the other hand, DR buyer bid formulation is derived from power exchange operation attributes/cost of generation. Therefore, the DR clearing ultimately affects the operational cost of operation which, in turn, is reflected in DR buying bid and DR clearing finally. This paper develops an iterative approach in which the continuous interactions between ISO/power exchange would finally converge to a stable operation. The load profile-based strategic bid formulation is modeled using mathematical as well as fuzzy inference system (FIS). In addition, an adaptive FIS has been devised to improve the performance of DR scheduling. The simulation results illustrate the impact of customer behavior, intelligent decision-making, and penetration level on the performance and convergence of power markets (DRX and power exchange). [ABSTRACT FROM AUTHOR]

This paper describes a two-stage unsupervised clustering based on the g-Means approach, which can employ any partitional or crisp clustering algorithm for load profiling of high-voltage (HV) feeders of a distribution system. The methodology proposed can evaluate number of clusters and its respective members from input data. The outcome of g-Means is dependent on a significance level— $\alpha$. This paper is based on feeder loading data, which are recorded at a transformer secondary at a distribution substation. Real-life active power data of 33-kV feeders acquired through existing supervisory control and data acquisition (SCADA) system of Central Indian state distribution company have been taken as a case study, as it is being processed for the first time, and hence, no prima facie information regarding the number of clusters or classes was available. The first stage entails checking for the randomness in daily load patterns and to extract month-wise working day (weekday) and nonworking day (holidays/Sundays) patterns. The second stage is based on this outcome, where clustering is performed to group similar patterns into clusters to generate month-wise “typical monthly load profiles”. The results are acceptable and are validated by using appropriate validity indices. [ABSTRACT FROM AUTHOR]

Highlights • Proposing an extended clustering algorithm for big-data analysis in smart power systems. • Proposing a novel and uniform method to model the big-data extracted from smart meters. • Proposing an innovative load profiling approach, incorporating reduced data complexity, to use in smart power system applications. Abstract Increasing penetration of distributed energy resources, varying load demands and big data of smart meters require new load models to support power system studies. The big data of smart meters and non-linearities in the load demand require the smart meter data to be represented in an alternative way to use in stochastic simulations to enhance processing. This paper proposes a novel method for stochastic load modelling of smart meter data. The approach turns smart meter data to a manageable level by linearizing energy consumption patterns producing energy classifications. A case study, using real world smart meter data, simulated scenarios to prove the robustness and accuracy of the method. The accuracy of results validates the stability and robustness of the approach and model validation provided substantiation for application in probabilistic studies. [ABSTRACT FROM AUTHOR]

This paper proposes a novel feature construction methodology aiming at both clustering yearly load profiles of low-voltage consumers, as well as investigating the stochastic nature of their peak demands. These load profiles describe the electricity consumption over a one-year period, allowing the study of seasonal dependence. The clustering of load curves has been extensively studied in literature, where clustering of daily or weekly load curves based on temporal features has received the most research attention. The proposed feature construction aims at generating a new set of variables that can be used in machine learning applications, stepping away from traditional, high dimensional, chronological feature sets. This paper presents a novel feature set based on two types of features: respectively the consumption time window on a daily and weekly basis, and the time of occurrence of peak demands. An analytic expression for the load duration curve is validated and leveraged in order to define the the region that has to be considered as peak demand region. The clustering results using the proposed set of features on a dataset of measured Flemish consumers at 15-min resolution are evaluated and interpreted, where special attention is given to the stochastic nature of the peak demands.

Transportation vehicles are a large contributor of the carbon dioxide emissions to the atmosphere. Electric Vehicles (EVs) are a promising solution to reduce the CO2 emissions which, however, requires the right electric power production mix for the largest impact. The increase in the electric power consumption caused by the EV charging demand could be matched by the growing share of Renewable Energy Sources (RES) in the power production. EVs are becoming a popular sustainable mean of transportation and the expansion of EV units due to the stochastic nature of charging behavior and increasing share of RES creates additional challenges to the stability in the power systems. Modeling of EV charging fleets allows understanding EV charging capacity and demand response (DR) potential of EV in the power systems. This article focuses on modeling of daily EV charging profiles for buildings with various number of chargers and daily events. The article presents a modeling approach based on the charger occupancy data from the local charging sites. The approach allows one to simulate load profiles and to find how many chargers are necessary to suffice the approximate demand of EV charging from the traffic characteristics, such as arrival time, duration of charging, and maximum charging power. Additionally, to better understand the potential impact of demand response, the modeling approach allows one to compare charging profiles, while adjusting the maximum power consumption of chargers.

Energy is indispensable to human existence, and its need is on the increase daily due to technological advancement. In recent times, there is sporadic increase in the development of modern buildings, and these buildings require a cost-effective and sustainable source of energy. Nigeria and some other African countries have major energy shortage issues due to the insufficient power generation and distribution facilities. This energy poverty creates a challenge for the growing population, and as such, it is vital to ensure that the energy supplied to a building is duly optimized and delivered cost effectively. In this study, an energy audit of an eight-floor multipurpose business complex was performed to determine the nature and type of loads within the building. Based on the load profile and energy consumption over a 10-year period, three alternative energy sources (national grid, diesel generator and photovoltaic [PV] system) were considered using various load-sharing ratios. The result reveals that though PV solar system is a renewable energy source that reduces the production of greenhouse gases generated by burning diesel, its continuous application in low–middle-income country like Nigeria may be challenged by the initial start-up capital which raises the cost of its unit energy.

The chief aim of this paper is to develop an effective approach to the issue of load profile clustering by applying Improved Replacement In Bee Optimization algorithm (IRIBO). While, intelligent metering solutions like Automated Meter Reading (AMR), Automated Meter Infrastructure (AMI) are in place to address the current issues prevailing in the domain of electricity markets, algorithm using Improved Replacement In Bee Optimization has been proved beneficial and uncomplicated to apply within a selective database. In this study Load Profile (LP) clustering distribution networks based on the shape of the load profile was studied for fitness function in the selected LP clustering. The results clearly indicate that LP clustering has advantages in providing metering solutions to consumers who do not possess digital metering which can be easily operated with trivial changes in the calibrations.

A case study of residential electricity consumption patterns mining and abnormal user identification using hierarchical clustering is presented in this paper. First, based on a brief introduction of hierarchical clustering, a process model and the specific steps of electricity consumption patterns mining in smart grid environment are proposed. Then, a case study using the daily electricity consumption data of 300 residential users in an eastern city of China, Kunshan, from November 16, 2014 to December 16, 2014, is presented. Through the implementation of hierarchical clustering, 9 abnormal users and 4 types of monthly electricity consumption patterns are successfully identified. The results show that most residential users in Kunshan city, nearly 81%, have a similar monthly electricity consumption pattern. Their average daily electricity consumption is about 7.73 kWh in the early winter with small fluctuations. Also, their daily electricity consumption is significantly associated with the temperature changes. However, it is worth noting that the special electricity consumption patterns of a small proportion of electricity users cannot be ignored, which is of great significance for the planning, operation, policy formulation and decision-making of smart grid. [ABSTRACT FROM AUTHOR]

This paper investigates the combined price-based scheduling/participation of generation company (GENCO) and demand response aggregation company (DRACO) in energy and reserve markets. The temporally coupled customer behavior can be better represented using the load profile attributes, when compared to the traditional approach with random willingness assignment. The proposed cost models for energy and reserve offerings consider the effect of load type, load pattern consumption, and availability/flexibility patterns of each type of load with time of use constraints. The load curtailment (LC) cost model accounts for criticality and willingness of the responsive loads via utilization factor and availability factors, respectively. The proposed cost models present a realistic picture of LC cost by eliminating the random willingness factor of the existing LC cost models. Thereafter, various cases of market participation with different reserve payment policies are formulated for combined participation of GENCO and DRACO. In addition, the sensitivity of participation decision of various entities to the seasonal load variation is examined for summer and winter loading profiles. The proposed cost models and scheduling framework is simulated using GENCO with ten thermal units and DRACO with various load types, profiles distributed across different load sectors comprising of commercial, residential, industrial, municipal, and agricultural loads. The combined participation resulted in improved market surplus with reduced GENCO surplus. Also, the energy and reserve market surplus dependence on seasonal load patterns is observed across different test cases and payment policies. [ABSTRACT FROM PUBLISHER]

Due to high implementation rates of smart meter systems, considerable amount of research is placed in machine learning tools for data handling and information retrieval. A key tool in load data processing is clustering. In recent years, a number of researches have proposed different clustering algorithms in the load profiling field. The present paper provides a methodology for addressing the aforementioned problem through Multi-Criteria Decision Analysis (MCDA) and namely, using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). A comparison of the algorithms is employed. Next, a single test case on the selection of an algorithm is examined. User specific weights are applied and based on these weight values, the optimal algorithm is drawn. [ABSTRACT FROM AUTHOR]

Global climate change has driven many manufacturing organizations to consider renewable energy as an alternative source of power for their operations. However, the instability and unpredictability of renewable energy has made it difficult to design an optimal solution. This is particularly critical for food manufacturing which is constrained by timing and seasonable fluctuations. This paper describes a feasibility study of offsetting grid supplied power by local generation using a roof top photovoltaic system. The study shows that the company can significantly reduce their carbon emitting energy usage by a 100 kW photovoltaic solar system that helps to reduce the annual electricity consumption by 39%. The system sizing methodology has taken into account the load profile of the food manufacturing plant and seasonal solar energy availability with the constrain to consume all solar power immediately.

The Kingdom of Saudi Arabia (KSA) has a large solar and wind energy resource. Through its Vision 2030 to exploit such resources, KSA is planning to install 9.5 GW of renewable energy power generation systems by 2030, through a mix of solar and wind energy. The government is planning to invest 109 billion US$ over the next 20 years for solar energy. The focus will be on solar photovoltaic (PV) and concentrated solar technologies at a national level. So far, the electricity demand in KSA is almost entirely dependent on fossil fuels for generating power. This paper addresses the potential to utilize the solar radiation resource at a different scale and reduce the power demand from the grid, bringing collateral benefits for householders and the government alike. The work presents the results from monitoring the electricity consumption of two typical domestic buildings (villas) in Jeddah, KSA. The electricity consumption observations were associated with indoor environmental conditions to study how and when cooling demand affects final demand. The study investigated options to serve the observed demand profile of the villas with simulated power generation from arrays of PV panels installed on two buildings’ roofs. Finally, a model of dynamic solar radiation simulation was developed to assess the hourly electricity generation, and a cost-benefit analysis was conducted for different capacity PV systems scenarios. The results indicate that locally used rooftop PV output could reduce the household electrical demand from the grid by around 80% at the housing level and in combination with building refurbishment solutions, could result in additional energy savings. The economic analysis discusses the implications of a proposed feed-in tariff with the associated payback periods and ROI, as well as proposals for PV system deployment at a large scale on the roof of buildings in KSA.

Clustering in energy markets is a top topic with high significance on expert and intelligent systems. The main impact of is paper is the proposal of a new clustering framework for the automatic classification of electricity customers’ loads. An automatic selection of the clustering classification algorithm is also highlighted. Finally, new customers can be assigned to a predefined set of clusters in the classification phase. The computation time of the proposed framework is less than that of previous classification techniques, which enables the processing of a complete electric company sample in a matter of minutes on a personal computer. The high accuracy of the predicted classification results verifies the performance of the clustering technique. This classification phase is of significant assistance in interpreting the results, and the simplicity of the clustering phase is sufficient to demonstrate the quality of the complete mining framework. [ABSTRACT FROM AUTHOR]

The changes and evolution of the electricity distribution has provided new possibilities to the electricity providers for developing a better marketing and trading strategies. A key aspect for designing specific tariff structures is the identification of customers groups exhibiting similar consumption patterns. This paper presents a new methodology for the classification of electricity customers on the basis of their electrical behaviour. Approach is based on the periodicity analysis and well known clustering technique - k-means. The paper presents the classification results obtained on a set of 3753 industrial users, whose consumption has been monitored for 3 years. [ABSTRACT FROM AUTHOR]

Due to high implementation rates of smart meter systems, considerable amount of research is placed in machine learning tools for data handling and information retrieval. A key tool in load data processing is clustering. In recent years, a number of researches have proposed different clustering algorithms in the load profiling field. The present paper provides a methodology for addressing the aforementioned problem through Multi-Criteria Decision Analysis (MCDA) and namely, using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). A comparison of the algorithms is employed. Next, a single test case on the selection of an algorithm is examined. User specific weights are applied and based on these weight values, the optimal algorithm is drawn.

The importance of Short-Term Load Forecasting (STLF) in power systems planning and management is reflected by the plethora of the related researches. STFL is a popular technical field in the power systems community and already counts many years of research activities and applications. The vast majority of the studies focus at the aggregated system load. Little attention is placed at small size loads, i.e. in buses of the transmission and distribution systems. Since there is a continuous advancement of smart grids technologies involving small size loads, bus STLF is a potentially important tool in smart grid applications. In contrast to system load, bus load presents a high level of stochasticity. Thus a robust forecaster should be able to capture and simulate the attributes of bus loads. The scope of the study is to develop bus forecasting models for day-ahead and hour-ahead load predictions. The models are based on Artificial Neural Networks (ANNs). Using a clustering methodology, the forecasting accuracy of the ANNs is enhanced leading to the formulation of hybrid forecasting models that are characterized by high level of parameterization and efficiency. The developed models are tested on a set of buses covering urban, sub-urban and industrial loads. [ABSTRACT FROM AUTHOR]

In a competitive retail market, large volumes of smart meter data provide opportunities for load serving entities to enhance their knowledge of customers’ electricity consumption behaviors via load profiling. Instead of focusing on the shape of the load curves, this paper proposes a novel approach for clustering of electricity consumption behavior dynamics, where “dynamics” refer to transitions and relations between consumption behaviors, or rather consumption levels, in adjacent periods. First, for each individual customer, symbolic aggregate approximation is performed to reduce the scale of the data set, and time-based Markov model is applied to model the dynamic of electricity consumption, transforming the large data set of load curves to several state transition matrixes. Second, a clustering technique by fast search and find of density peaks (CFSFDP) is primarily carried out to obtain the typical dynamics of consumption behavior, with the difference between any two consumption patterns measured by the Kullback-Liebler distance, and to classify the customers into several clusters. To tackle the challenges of big data, the CFSFDP technique is integrated into a divide-and-conquer approach toward big data applications. A numerical case verifies the effectiveness of the proposed models and approaches. [ABSTRACT FROM PUBLISHER]

In order to improve low voltage (LV) network visibility without extensive monitoring and integrate low carbon technologies (LCTs) in a cost-effective manner, this paper proposes a novel three-stage network load profiling method. It uses real-time information monitored from selective representative areas to develop network templates. The three stages are: clustering, classification and scaling. It can be used to identify the loading conditions of unmonitored LV systems with similar fixed data to those monitored LV substations. In the clustering stage, hierarchical clustering and K-means are used to cluster substations into groups based on the shape of the monitored load profiles. The classification tool designed with multinomial logistic regression maps an unmonitored LV substation into the most probable templates by using routinely available fixed data. Finally, clusterwise weighted constrained regression is employed to estimate peak for individual LV substations and the developed templates. The three-stage profiling is demonstrated on a practical system in the U.K. under the umbrella of a smart grid trail project. Ten LV templates are developed by using the metered data from 800 monitored LV substations. A comprehensive comparison between the estimated peaks using the three-stage process and the metered peaks suggests that the methodology can achieve superior accuracy. This is part I of the paper, introducing clustering and classification. The scaling (peak estimation) process will be introduced in part II of the paper. [ABSTRACT FROM PUBLISHER]

This paper presents an electricity medium voltage (MV) customer characterization framework supported by knowledge discovery in database (KDD). The main idea is to identify typical load profiles (TLP) of MV consumers and to develop a rule set for the automatic classification of new consumers. To achieve our goal a methodology is proposed consisting of several steps: data pre-processing; application of several clustering algorithms to segment the daily load profiles; selection of the best partition, corresponding to the best consumers’ segmentation, based on the assessments of several clustering validity indices; and finally, a classification model is built based on the resulting clusters. To validate the proposed framework, a case study which includes a real database of MV consumers is performed. [ABSTRACT FROM AUTHOR]