Determination of electricity consumers’ load profiles via weighted evidence accumulation clustering using subsampling

With the electricity market liberalization, the distribution and retail companies are looking for better market strategies based on adequate information upon the consumption patterns of its electricity consumers. A fair insight on the consumers’ behavior will permit the definition of specific contract aspects based on the different consumption patterns. In order to form the different consumers’ classes, and find a set of representative consumption patterns we use electricity consumption data from a utility client’s database and two approaches: Two-step clustering algorithm and the WEACS approach based on evidence accumulation (EAC) for combining partitions in a clustering ensemble. While EAC uses a voting mechanism to produce a co-association matrix based on the pairwise associations obtained from N partitions and where each partition has equal weight in the combination process, the WEACS approach uses subsampling and weights differently the partitions. As a complementary step to the WEACS approach, we combine the partitions obtained in the WEACS approach with the ALL clustering ensemble construction method and we use the Ward Link algorithm to obtain the final data partition. The characterization of the obtained consumers’ clusters was performed using the C5.0 classification algorithm. Experiment results showed that the WEACS approach leads to better results than many other clustering approaches.

Typical load profiles in the smart grid context – a clustering methods comparison

The present research paper presents five different clustering methods to identify typical load profiles of medium voltage (MV) electricity consumers. These methods are intended to be used in a smart grid environment to extract useful knowledge about customer’s behaviour. The obtained knowledge can be used to support a decision tool, not only for utilities but also for consumers. Load profiles can be used by the utilities to identify the aspects that cause system load peaks and enable the development of specific contracts with their customers. The framework presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partition, which is supported by cluster validity indices. The process ends with the analysis of the discovered knowledge. To validate the proposed framework, a case study with a real database of 208 MV consumers is used.

A combined engineering and statistical model of UK domestic appliance electrical load profiles

The development of a combined engineering and statistical Artificial Neural Network model of UK domestic appliance load profiles is presented. The model uses diary-style appliance use data and a survey questionnaire collected from 51 suburban households and 46 rural households during the summer of 2010 and2011 respectively. It also incorporates measured energy data and is sensitive to socioeconomic, physical dwelling and temperature variables. A prototype model is constructed in MATLAB using a two layer feed forward network with back propagation training which has a 12:10:24 architecture. Model outputs include appliance load profiles which can be applied to the fields of energy planning (microrenewables and smart grids), building simulation tools and energy policy.

A Data Mining Framework for Electric Load Profiling

This paper consists in the characterization of medium voltage (MV) electric power consumers based on a data clustering approach. It is intended to identify typical load profiles by selecting the best partition of a power consumption database among a pool of data partitions produced by several clustering algorithms. The best partition is selected using several cluster validity indices. These methods are intended to be used in a smart grid environment to extract useful knowledge about customers’ behavior. The data-mining-based methodology presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partitions. To validate our approach, a case study with a real database of 1.022 MV consumers was used.

Load Profiling Tool to Support Smart Grid Operation Scenarios

This paper presents the characterization of high voltage (HV) electric power consumers based on a data clustering approach. The typical load profiles (TLP) are obtained selecting the best partition of a power consumption database among a pool of data partitions produced by several clustering algorithms. The choice of the best partition is supported using several cluster validity indices. The proposed data-mining (DM) based methodology, that includes all steps presented in the process of knowledge discovery in databases (KDD), presents an automatic data treatment application in order to preprocess the initial database in an automatic way, allowing time saving and better accuracy during this phase. These methods are intended to be used in a smart grid environment to extract useful knowledge about customers’ consumption behavior. To validate our approach, a case study with a real database of 185 HV consumers was used.

Load and Season Adapted Solar Collectors

In Sweden solar irradiation and space heating loads are unevenly distributed over the year. Domestic hot water loads may be nearly constant. Test results on solar collector performance are often reported as yearly output of a certain collector at fixed temperatures, e g 25, 50 and 75 C. These data are not suitable for dimensioning of solar systems, because the actual performance of the collector depends heavily on solar fraction and load distribution over the year.At higher latitudes it is difficult to attain high solar fractions for buildings, due to overheating in summer and small marginal output for added collector area. Solar collectors with internal reflectors offer possibilities to evade overheating problems and deliver more energy at seasons when the load is higher. There are methods for estimating the yearly angular irradiation distribution, but there is a lack of methods for describing the load and the storage in such a way as to enable optical design of season and load adapted collectors.This report describes two methods for estimation of solar system performance with relevance for season and load adaption. Results regarding attainable solar fractions as a function of collector features, load profiles, load levels and storage characteristics are reported. The first method uses monthly collector output data at fixed temperatures from the simulation program MINSUN for estimating solar fractions for different load profiles and load levels. The load level is defined as estimated yearly collector output at constant collector temperature divided be yearly load. This table may examplify the results:CollectorLoadLoadSolar Improvementtypeprofile levelfractionover flat plateFlat plateDHW 75 %59 %Load adaptedDHW 75 %66 %12 %Flat plateSpace heating 50 %22 %Load adaptedSpace heating 50 %28 %29 %The second method utilises simulations with one-hour timesteps for collectors connected to a simplified storage and a variable load. Collector output, optical and thermal losses, heat overproduction, load level and storage temperature are presented as functions of solar incidence angles. These data are suitable for optical design of load adapted solar collectors. Results for a Stockholm location indicate that a solar combisystem with a solar fraction around 30 % should have collectors that reduce heat production at solar heights above 30 degrees and have optimum efficiency for solar heights between 8 and 30 degrees.

Load profiling and data mining techniques in electricity deregulated market

This paper presents load profiles of electricity customers, using the knowledge discovery in databases (KDD) procedure, a data mining technique, to determine the load profiles for different types of customers. In this paper, the current load profiling methods are compared using data mining techniques, by analysing and evaluating these classification techniques. The objective of this study is to determine the best load profiling methods and data mining techniques to classify, detect and predict non-technical losses in the distribution sector, due to faulty metering and billing errors, as well as to gather knowledge on customer behaviour and preferences so as to gain a competitive advantage in the deregulated market. This paper focuses mainly on the comparative analysis of the classification techniques selected; a forthcoming paper will focus on the detection and prediction methods.

Connection design without accurate LV feeder load data:an argument for LV monitoring

Since privatisation, maintenance of DNO LV feeder maximum demand information has gradually demised in some Utility Areas, and it is postulated that lack of knowledge about 11kV and LV electrical networks is resulting in a less economical and energy efficient Network as a whole. In an attempt to quantify the negative impact, this paper examines ten postulated new connection scenarios for a set of real LV load readings, in order to find the difference in design solutions when LV load readings were and were not known. The load profiles of the substations were examined in order to explore the utilisation profile. It was found that in 70% of the scenarios explored, significant cost differences were found. These cost differences varied by an average of 1000%, between schemes designed with and without load readings. Obviously, over designing a system and therefore operating more, underutilised transformers becomes less financially beneficial and less energy efficient. The paper concludes that new connection design is improved in terms of cost when carried out based on known LV load information and enhances the case for regular maximum feeder demand information and/or metering of LV feeders. © 2013 IEEE.

GA-ANN Short-Term Electricity Load Forecasting

This paper presents a methodology for short-term load forecasting based on genetic algorithm feature selection and artificial neural network modeling. A feed forward artificial neural network is used to model the 24-h ahead load based on past consumption, weather and stock index data. A genetic algorithm is used in order to find the best subset of variables for modeling. Three data sets of different geographical locations, encompassing areas of different dimensions with distinct load profiles are used in order to evaluate the methodology. The developed approach was found to generate models achieving a minimum mean average percentage error under 2 %. The feature selection algorithm was able to significantly reduce the number of used features and increase the accuracy of the models.

Chronological categorization and decomposition of customer loads

The majority of distribution utilities do not have accurate information on the constituents of their loads. This information is very useful in managing and planning the network, adequately and economically. Customer loads are normally categorized in three main sectors: 1) residential; 2) industrial; and 3) commercial. In this paper, penalized least-squares regression and Euclidean distance methods are developed for this application to identify and quantify the makeup of a feeder load with unknown sectors/subsectors. This process is done on a monthly basis to account for seasonal and other load changes. The error between the actual and estimated load profiles are used as a benchmark of accuracy. This approach has shown to be accurate in identifying customer types in unknown load profiles, and is used in cross-validation of the results and initial assumptions.

An FE investigation simulating intra-operative corrective forces applied to correct scoliosis deformity

Background: Adolescent idiopathic scoliosis (AIS) is a deformity of the spine, which may 34 require surgical correction by attaching a rod to the patient’s spine using screws 35 implanted in the vertebral bodies. Surgeons achieve an intra-operative reduction in the 36 deformity by applying compressive forces across the intervertebral disc spaces while 37 they secure the rod to the vertebra. We were interested to understand how the 38 deformity correction is influenced by increasing magnitudes of surgical corrective forces 39 and what tissue level stresses are predicted at the vertebral endplates due to the 40 surgical correction. 41 Methods: Patient-specific finite element models of the osseoligamentous spine and 42 ribcage of eight AIS patients who underwent single rod anterior scoliosis surgery were 43 created using pre-operative computed tomography (CT) scans. The surgically altered 44 spine, including titanium rod and vertebral screws, was simulated. The models were 45 analysed using data for intra-operatively measured compressive forces – three load 46 profiles representing the mean and upper and lower standard deviation of this data 47 were analysed. Data for the clinically observed deformity correction (Cobb angle) were 48 compared with the model-predicted correction and the model results investigated to 49 better understand the influence of increased compressive forces on the biomechanics of 50 the instrumented joints. 51 Results: The predicted corrected Cobb angle for seven of the eight FE models were 52 within the 5° clinical Cobb measurement variability for at least one of the force profiles. 53 The largest portion of overall correction was predicted at or near the apical 54 intervertebral disc for all load profiles. Model predictions for four of the eight patients 55 showed endplate-to-endplate contact was occurring on adjacent endplates of one or 56 more intervertebral disc spaces in the instrumented curve following the surgical loading 57 steps. 58 Conclusion: This study demonstrated there is a direct relationship between intra-59 operative joint compressive forces and the degree of deformity correction achieved. The 60 majority of the deformity correction will occur at or in adjacent spinal levels to the apex 61 of the deformity. This study highlighted the importance of the intervertebral disc space 62 anatomy in governing the coronal plane deformity correction and the limit of this 63 correction will be when bone-to-bone contact of the opposing vertebral endplates 64 occurs.

A flexible tool for integrated planning of active distribution networks

This paper presents a flexible and integrated planning tool for active distribution network to maximise the benefits of having high level s of renewables, customer engagement, and new technology implementations. The tool has two main processing parts: “optimisation” and “forecast”. The “optimization” part is an automated and integrated planning framework to optimize the net present value (NPV) of investment strategy for electric distribution network augmentation over large areas and long planning horizons (e.g. 5 to 20 years) based on a modified particle swarm optimization (MPSO). The “forecast” is a flexible agent-based framework to produce load duration curves (LDCs) of load forecasts for different levels of customer engagement, energy storage controls, and electric vehicles (EVs). In addition, “forecast” connects the existing databases of utility to the proposed tool as well as outputs the load profiles and network plan in Google Earth. This integrated tool enables different divisions within a utility to analyze their programs and options in a single platform using comprehensive information.

Non-convex dynamic economic/environmental dispatch with plug-in electric vehicle loads

Electric vehicles are a key prospect for future transportation. A large penetration of electric vehicles has the potential to reduce the global fossil fuel consumption and hence the greenhouse gas emissions and air pollution. However, the additional stochastic loads imposed by plug-in electric vehicles will possibly introduce significant changes to existing load profiles. In his paper, electric vehicles loads are integrated into an 5-unit system using a non-convex dynamic dispatch model. The actual infrastructure characteristics including valve-point effects, load balance constrains and transmission loss have been included in the model. Multiple load profiles are comparatively studied and compared in terms of economic and environmental impacts in order o identify patterns to charge properly. The study as expected shows ha off-peak charging is the best scenario with respect to using less fuels and producing less emissions.

Optimatization of hybrid renewable energy systems on isolated microgrids : a smart grid approach

Tese de doutoramento, Sistemas Sustentáveis de Energia, Universidade de Lisboa, Faculdade de Ciências, 2016

Data mining contributions to characterize MV consumers and to improve the suppliers-consumers settlements

This paper deals with the establishment of a characterization methodology of electric power profiles of medium voltage (MV) consumers. The characterization is supported on the data base knowledge discovery process (KDD). Data Mining techniques are used with the purpose of obtaining typical load profiles of MV customers and specific knowledge of their customers’ consumption habits. In order to form the different customers’ classes and to find a set of representative consumption patterns, a hierarchical clustering algorithm and a clustering ensemble combination approach (WEACS) are used. Taking into account the typical consumption profile of the class to which the customers belong, new tariff options were defined and new energy coefficients prices were proposed. Finally, and with the results obtained, the consequences that these will have in the interaction between customer and electric power suppliers are analyzed.