Search

A key component in superconducting magnetic energy storage (SMES) system design is the cooling system. A well-designed cooling apparatus allows a SMES to work with higher critical currents, have less AC losses and work in a stable way. Acknowledging the importance of cooling systems in a SMES and that there have been many different designs presented in the bibliography, we hereby analyze various thermal designs that utilize cryocoolers, and we estimate the total volume of the superconducting coils used in their experiments. This review article focuses on recent high temperature superconducting (HTS) materials, and presents the cooling methods employed to cool the coil in conjunction with other important characteristics, such as the coil geometry, the operating temperature of the magnet, the thermal insulation temperature and the cooling power of the cryocoolers. From the presented analysis, the authors propose a procedure to devise a proper cooling scheme for a SMES design. Additionally, a qualitative comparative assessment of the analyzed designs in literature, in eight assessment criteria, is presented.

In this paper a sizing tool is proposed for Photovoltaic (PV) and Battery Energy Storage Systems (BESS) in installations with Demand Side Management capability. Three objectives functions are defined, describing the Self-Sufficiency Rate, the Net-Present Value of the investment in PV and BESS, and their combination in a joint expression, respectively. Two analysis methods are developed within the tool to find the optimal solution. The first one applies an exhaustive search by examining all possible combinations of PV and BESS capacity, under pre-defined load shifting. The drawback of this method is the high computational burden due to unnecessary access in the solution space. The second method is implemented with a Particle Swarm optimization (PSO) variant, namely the Unified-PSO. Under this approach, the analysis algorithm is guided in the solution space towards efficient solutions under lower computational time. The comparative assessment of the two methods verifies that via the PSO algorithm the optimal solutions are found under almost half execution time.

In this paper, an unsupervised non-intrusive load monitoring approach is proposed in order to encounter the disaggregation problem for Non-Intrusive Load Monitoring (NILM) methodologies, using odd harmonic current amplitudes. The problem has been contemplated as a multi-class multi-label one and for the combinations examined the number of appliances operating simultaneously varies between from one to up to three appliances. K-means has been utilized to cluster the different combinations, using additionally the elbow technique, in order to obtain the most suitable number of clusters that should be created. The results indicate that the proposed technique performs efficient load identification even with few samples at the training stage especially under the consideration of the third and fifth odd harmonic currents.

Superconducting magnetic energy storage systems (SMES) store energy in the form of magnetic field generated by a DC current flowing through a superconducting coil which has been cooled at a low temperature, typically less than 77 K. Therefore, the optimization of the coil and the design of the thermal design are crucial. In this paper the first steps in the development of a laboratory scale SMES are presented. Initially, based on the available material, the geometry of the coil is optimized to store the highest possible amount of energy utilizing Finite Element Methods (FEM). Based on the final geometry, the coil is built and thermally modelled inside a vacuum chamber. An initial experiment is carried out to cool the coil at 70 K and verify the existence of superconductivity. The measurements are carried out using a LabVIEW program.

In the smart home context, one of the features that most users would desire is the knowledge of the operation of each electrical appliance without the need to install expensive smart plugs or measuring equipment at each outlet-appliance. The Non-Intrusive Load Monitoring (NILM) idea is an alternative method where only one meter is utilized. However, the accuracy and simplicity of NILM approaches are still under improvement and evolvement. In this paper we propose the development of load signatures by a limited number of harmonic current vectors. The rationale behind this is that simultaneously operating appliances may have opposite or displaced harmonic current vectors, which means that using only current amplitudes as load signatures will not result in reliable identification. Three alternative formations are examined aiming to highlight the contribution of the harmonic current phase angles to the robustness of the load signatures. The first formation considers only the current amplitude, the second takes into account the phase angles in the form of vector projection on the x-axis, and the third is based on harmonic current vectors. The respective disaggregation schemes are also proposed considering either amplitude summations or vector aggregations, in order to examine the correct appliance combination that matches the actual operating one. Measurements have been performed in a typical residential installation considering both stand-alone operation of appliances and scenarios with appliance combinations. The results indicate that disaggregation performance is significantly improved when relying on harmonic current vectors in respect to the case with only current amplitudes, which makes the methodology potentially suitable for the new smart meters that are expected to be widely installed.

Integration of Photovoltaics (PV) with Energy Storage Systems (ESS) and Demand Side Management (DSM) is an innovative way that could transform a building into a self-sufficient nanogrid. Such an approach could act as an energy rehabilitation solution and contribute to alleviating the technical problems posed to the electrical grid from high consumption of buildings and high PV injections. The scope of this paper is to investigate the potential for achieving high level of building self-sufficiency using hybrid PV-ESS-DSM solutions. Towards this direction this work analyzes the self-consumption and self-sufficiency rates for various installations equipped with PV and ESS with different climatic and load demand characteristics. Moreover, the potential for increasing these indices under DSM is examined aiming to maximize the utilization of renewable energy produced on-site.

This paper examines how Low Voltage Distribution Networks (DNs) can optimize their voltage profile and minimize their energy losses through the planning ofcharging of Electric Vehicles (EVs).We consider variations concerning the charger and battery capacity of the EVs and we formulate two objective functions. A Particle Swarm Optimization (PSO) algorithm is utilized for the determination of the optimized charging planning of the EVs. The main idea is to examine both the Grid-to- Vehicle (G2V) and Vehicle-to-Grid (V2G) schemes for the EVs and evaluate how each of them contributes to the improvement of the operational characteristics of the DN. Thus, for the G2V scheme the EVs are considered only under charging mode and the algorithm provides their optimal charging planning for one day. Additionally, for the V2G scheme we consider the possibility for EV discharging as well. The simulations are carried out on a real DN with real loading data for the day of the year with the worst voltage profile. The results indicate that under proper charging planning the DN will have a smoother voltage profile with lower power losses. Moreover, the controlled V2G scheme can further improve this situation.

Being an engineering field, power systems provide an extensive subject for optimization to be applied upon. Modern power systems have evolved in an increasingly highly complex system. The liberalization of the energy market and the introduction of distributed generation and, in particular, distributed renewable energy resources (DRES) have raised both opportunities and challenges that need to be tackled. Thus, complex issues related to the operation and planning of the distribution systems have emerged. Such issues involve many variables and refer to nonlinear objectives; thus their optimization is significantly based on heuristic techniques, such as particle swarm optimization (PSO). In this chapter, the implementation of PSO when contemplating various problems in power systems is presented. In particular, the utilization of PSO is demonstrated in the optimal distributed generation placement problem (ODGP), also known as optimal siting and sizing of distributed generation problem, and in the network reconfiguration problem. Finally, PSO is implemented in an optimal schedule of electric vehicles (EVs) charging, providing an apt example of the variety of problems for which PSO can be utilized and providing useful aid to important decisions, in the field of power systems.

This chapter introduces the Network Reconfiguration (NR) concept in Distribution Networks (DNs) as an efficient scheme to face various operational issues like reliability improvement and loss reduction. Furthermore, the potential for utilizing NR to perform voltage profile improvement under high DG or Renewable Energy Sources (RESs) penetration is presented. Finally, the coordination of the NR along with the optimal siting and sizing of DG units aiming to maximize their impact on loss reduction is also analyzed. The basic aim of this chapter is to demonstrate how specific automation upgrade in modern DNs regarding the replacement of manual switching equipment by automated controlled sectionalizers or tie-switches could allow Distribution System Operators (DSOs) to integrate real time management techniques of the DN under relatively low investment plans. Specific examples regarding both real and benchmarked DNs are included and the proposed algorithms are explained in detail.

The Optimal Distributed Generation Placement problem (ODGP) towards energy loss minimization depends basically on the network's layout and its load composition. Under load variations, different load compositions result, for each one of them, is highly possible to come up with a different optimal solution regarding the optimal siting and sizing of DG units. This paper examines the impact of these variations in order to verify how optimal solution should adapt to any load composition. A Local Particle Swarm Optimization Variant algorithm is proposed as the solution algorithm and numerous load composition snapshots for the IEEE-33 bus system are examined. Moreover, a methodology is proposed in order to highlight the critical nodes that prove to have an essential role to the solution. Finally, the possibility for the determination of a fixed solution with fixed installation nodes and constant power output that could yield near optimal energy loss reduction is examined.

Integration of distributed generation (DG) in existing distribution networks has been studied thoroughly during the past years as a measure of reducing grid's power losses. However, the optimal DG placement, known as ODGP, toward loss minimisation, has not been studied in depth by considering the possible impact of the reverse power flow (RPF) caused by extended penetration of distributed energy resources. This study uses a constriction factor embedded local particle swarm optimisation algorithm along with the appropriate particle formulation that solves the ODGP problem by taking into account the impact of possible RPF. In this study, the idea of RPF modelling is introduced by providing extended versions of IEEE test systems. Modified versions of the IEEE 30-bus and IEEE 33-bus test systems are modelled and results are presented in order to highlight the impact of RPF on the ODGP problem solution. The mathematical formulation is given, results and analysis for both extended systems are presented, while the importance of RPF for different conditions is assessed.

Scope of this paper is to examine the performance of two disaggregation schemes for NILM algorithms based on the utilization of current harmonic vectors. The analysis is based on measurements performed on real residential installation and the utilized features regarding the formulation of the load signatures of the appliances refer to the odd harmonic current vectors up to the fifth harmonic order. Moreover, a large number of measured appliance combinations is used for the application of the disaggregation schemes in order to evaluate their performance. The first scheme relies on sequential disaggregation for the harmonic orders, starting from the fifth and moving on to the first. The second scheme concerns simultaneous disaggregation of all harmonic orders in a parallel manner. The merits and flaws of both schemes are highlighted and the results indicate that for the parallel disaggregation scheme the appliance identification performance is higher at the expense of the computational burden.

The performance of neural networks in load forecasting tasks is highly influenced by the selection of the inputs. This selection is either problem specific or is relied on the literature. The scope of the present study is to compare two features (i.e. inputs) selection methods based on metaheuristics. Thus, the inputs selection process is treated as an optimization process. The test case involves the day-ahead load forecasting of the aggregated load covered by a distribution substation in a large urban area.

This paper examines the application of machine learning techniques in NILM methodologies based on the first three odd harmonic order current vectors as the only attributes of the appliances. Proper formulation of the measured current waveform of appliances' combinations is also presented. We apply our methodology on performed measurements of typical Low Voltage residential installations considering harmonic order currents as the input features for both the training and disaggregation scheme. Our results support the hypothesis that the identification performance is enhanced when higher harmonic currents are included in the NILM methodology.

The scope of the present study is the assessment of a Particle Swarm Optimization (PSO) method for optimal input selection for Short-Term Load Forecasting (STLF) tasks. The test case involves the STLF of a high to medium voltage transformer, where a Feed-Forward Neural Network (FFNN) is trained and applied. An effective training process is crucial to the forecasting model's success. Low number of inputs can lead to low potential for generalization. On the contrary, a high number may lead to large training duration. Thus, a trade-off between the generalization and duration should be accomplished. Using the PSO based method, the most correlated inputs are selected, a fact that leads to low prediction errors.

The present paper proposes an integrated Demand Response (DR) methodology for eligible consumers in day-ahead markets. The test case refers to a high voltage industrial consumer of the Greek energy system. The first stage of the methodology involves a load profiling process to derive the typical load patterns of the consumers. Based on the shapes of the load profiles, a dedicated DR measure is designed for the purpose of manifesting a specific load modification target. The DR measure is supported with a Short-Term Load Forecasting (STLF) routine in day-ahead basis. In the paper, load predictions are treated as a tool that can provide information about the potential of the DR measures. Based on the forecasts, the DR are designed, implemented and evaluated.

Natural gas is considered as a crucial power source both for electrical and thermal loads. Natural gas markets are continually expanding, a fact that raises the need of formulating new energy services to the end-users. The aim of the present paper is to develop a two-stage clustering process that leads to the formation of natural gas demand profiles per distribution point. Natural gas demand profiles can be further utilized in a variety of applications such as forecasting, tariff designs and others. The test case refers to the distribution points of the national natural gas system of Greece. The results show that hierarchical algorithms are the suitable clustering solution for the dataset under study.

The penetration of Electric Vehicles (EVs) is rapidly increasing at the Low Voltage (LV) level, thus increasing the electrification rate of transport. Moreover, Photovoltaic (PV) systems at the LV level have been constantly increasing as well. Buildings with PVs on their roof and EVs as the main transport means may well become the norm the following years, considering additionally that all new buildings in the EU should be Nearly Zero Energy Buildings (NZEBs). Such NZEBs should cover the majority of their low energy demand by Renewable Energy Sources, and PV technology is the main candidate to achieve that. Both these facts pose considerable challenges on the LV networks. In this context, this paper examines the potential of efficiently utilizing the stored energy of Battery Energy Storage Systems (BESSs) in order to face the increased load demand by the penetration of EVs in LV Distribution Networks (DNs), with the presence of PVs. A Particle Swarm Optimization (PSO) algorithm is developed in order to perform optimal charging schedule for the EVs under the aim of optimizing the DN voltage profile. The analysis is performed on a real LV DN with real load data and the results indicate that under proper charging schedule both the voltage profile and the energy losses of the DN could be improved.

In this paper a novel approach regarding the optimal penetration of Distributed Generation (DG) in Distribution Networks (DNs) towards loss minimization is proposed. More specific, a Local Particle Swarm Optimization (PSO) variant algorithm is developed in order to define the optimal active and reactive power generation and/or consumption requirements for the optimal number and location of nodes that yield loss minimization. Thus, the proposed approach provides the optimal number, siting and sizing of DGs altogether. In addition, based on the optimal power requirements of the resulted nodes, a combination of potential DG types to be installed is recommended. The proposed objective function in this paper is also innovative since it embeds the constraint of reverse power flow to the slack bus by the formation of a new penalty term. The proposed methodology is applied to 30 and 33 bus systems. The results indicate the optimal number, locations, and capacity of DG units, which were calculated simultaneously. Finally, the impact of the predefined amount of permissible reverse power flow to the optimal solution is also examined through two scenarios: the first considers zero reverse power flow and the second unlimited reverse power flow.

This paper focuses on the optimal siting of Battery Energy Storage Systems (BESS) in a Distribution Network with installed Photovoltaic Generation, in order to minimize the energy losses of the system. The bus voltage limit, as well as the ampacity level of the lines are taken into consideration as constraints, while the technical constraints of the BESSs have also been taken into account. Unified Particle Swarm Optimization is used as the solving optimization technique. Simulations are being carried out on IEEE-33 bus system regarding different scenarios and the results are presented and compared. A significant improvement in energy losses, voltage and line ampacity profile is achieved by the introduction of BESS units in a Distribution Network with high PV Penetration.

This paper focuses on the optimal siting of Battery Energy Storage Systems (BESS) in a Distribution Network (DN) with installed Wind Generation (WG), in order to minimize the energy losses and reduce the power load fluctuation of the DN. The bus voltage limit, as well as the ampacity level of the lines are taken into consideration as constraints, while the technical characteristics of BESSs and WG have also been taken into account. In order to propose a solution for this optimization problem, the Unified version of the Particle Swarm Optimization technique is utilized, called UPSO. Simulations have been carried out on the typical 33-Bus system and the results are presented and compared. It is deduced that the introduction of BESSs can aid to the high penetration of WG, blunting the latter's severe consequences on the DN, reducing energy losses, and line ampacity levels and improving the voltage profile of the DN.

This paper examines voltage quality problems in Low Voltage (LV) Distribution Networks (DNs) due to high Photovoltaic (PV) and Electric Vehicles (EVs) penetration. The proposed methodology considers Battery Energy Storage Systems (BESSs) along with the PV units in order to store the energy surplus that causes overvoltage during the high PV generation. This stored energy is utilized at night when the majority of the EVs is expected to be charged. Moreover, a Unified Particle Swarm Optimization (UPSO) algorithm is used in order to optimally schedule EVs charging during the night and under different charging modes. This scheduling is implemented under either an assumed continuous charging mode or an intermittent one. The objective function is set to be the optimal voltage profile. A real LV DN with real measured load data is examined. Variations have been taken into account about some EVs' parameters and the results indicate that the voltage profile is significantly improved under the proposed charging schedule by the UPSO. If this charging schedule is combined with efficient exploitation of the BESS stored energy then the improvement is even higher. The proposed approach also improves energy efficiency towards the implementation of the Nearly Zero Energy Buildings (NZEB) concept.

Due to the high installation rate of Photovoltaics (PV) systems, the challenging task of data processing arises. As the number of intermittent PV systems grows especially in the distribution networks, this task becomes even more important. In many cases, the data present inconsistencies, i.e. missing, incomplete or profoundly wrong values. The present paper proposes a method built on machine learning algorithms for missing data completion. The original incomplete PV generation time series are filled and restored with low error. The proposed method can be easily applied in real installations with high rate of data collection and storage needs.

This chapter introduces the Optimal Distributed Generation Placement problem towards power and energy loss minimization. Several solving methods are applied in order for the most suitable to emerge. Apart from technical and DG constraints, recent raised issues due to high Distributed Generation penetration like the reverse power flow effect is considered as well. The load and generation variability and their impact in integrating Renewable Energy Sources are examined, aided by the use of Capacity Factors implementation. In addition, the impact of Optimal Distributed Generation Placement problem in conjunction with Network Reconfiguration and Optimal Energy Storage Systems Placement is introduced aiming to examine how joined management schemes could be efficiently combined in order to maximize the potential loss and energy reduction.

This study examines how environmental and socioeconomic criteria affect renewable energy resources (RES) distribution strategic plans regarding national energy policies. Four criteria are introduced with respective coefficients properly formulated to quantify their capacity. Moreover, these coefficients are properly normalised to combine the effect of each criterion under a uniform formulation. The base case scenario in this work considers an initially available capacity of RESs to be equally distributed among the candidate regions. Six scenarios about different prioritisation are examined. The results prove that different prioritisation criteria yield significant variations regarding the assigned regional RES capacity. The proposed algorithm defines optimisation only by terms of predefined prioritisation criteria; each solution could be considered optimal given that the respective installation strategic plan is subject to specific weighted criteria. The advantages of the proposed algorithm rely on its simplicity and expandability, since both coefficients formulation and resizing procedure are easily performed, as well as additional criteria could be easily incorporated in the resizing procedure. Thus, this algorithm could be considered as a multi-objective planning tool regarding long-term strategic plans for nationwide RES distribution.

This paper examines the development of load signatures based on odd order harmonic current vectors. A measurement setup is developed, and measurements concerning stand-alone and combined operation of several appliances in steady state have been performed using the fundamental voltage phase angle as the angular reference. The proposed load signatures consist of the first three harmonic order currents and each one of them is considered a vector quantity. An alternative formulation that considers only the projections of these vectors on the x-axis is also proposed in order to simplify the load signatures structure. The disaggregation scheme regarding the NILM approach assumes that the contribution of each appliance's harmonic current to the aggregated one, under simultaneous operation of specific appliances, is linear and results by summation of the respective components. Results indicate that the accuracy of this approach is higher when harmonic currents are treated as vectors and their phase angle is taken into account in the computations. The proposed load signature formulation is simple, the linear approach for the disaggregation scheme is kept regardless the harmonic current vectors consideration and the results seem to be more accurate.

In this paper, a comparative analysis and evaluation of several heuristic techniques, when applied to the Optimal Distributed Generation Placement (ODGP) problem, is presented. Loss minimization is considered as the objective, while the network's technical characteristics as the constraints. Three versions of Particle Swarm Optimization (PSO), Global, Local and Unified (GPSO, LPSO and UPSO, respectively), Genetic Algorithm (GA), Artificial Bee Colony (ABC), Cuckoo Search (CS) and Harmony Search (HS) are compared. The implemented analysis demonstrates that all Heuristic Techniques examined can solve the ODGP problem efficiently, although UPSO emerge as the most promising, in terms of solution and convergence performance, whereas regarding execution time, HS is the most prominent. The study is evaluated upon typical 33 and 30 bus systems.

This paper proposes a measurement based methodology towards the development of behavioral patterns regarding the lighting usage in offices. Based on measurements concerning illumination in working environment, in relation to work place occupation and outdoors weather conditions, the presented methodology highlights the fundamental aspects regarding the involved parameters towards the development of new and sophisticated lighting control algorithms. The results derived by the measurements indicate that subjective criteria, related to personal preferences, differentiate behavior patterns for users with similar objective characteristics. Thus, a user-oriented algorithm for lighting control should take into account crucial information about personalized behavioral patterns in order to adjust illumination in respect to history data, current weather conditions and feedback by the user.

In this paper loss reduction in Distribution Networks is performed under proper consideration regarding Distributed Generation (DG) siting and sizing and network reconfiguration. The optimal solution for each of the aforementioned techniques, when applied individually, is highly dependent on the network layout and on its load composition. Thus, the question that arises when both techniques are utilized, concerns their application order towards the most efficient solution. The analysis in this work examines how the sequence regarding the implementation of these two techniques affects the optimal solution in terms of DG capacity penetration and number of switching operations. Various scenarios with different available DG units are examined regarding their optimal siting and sizing before and after the implementation of the network reconfiguration technique. Finally, an alternative scenario with the simultaneous application of both techniques is also examined. In this latter case the complexity of the problem is highly increased and the solution efficiency is also affected. A special PSO variant is utilized as the solution algorithm, namely Unified-PSO, accordingly modified for each scenario and the IEEE 33 bus system is utilized as the reference distribution network. The analysis concludes about the optimal application order of these two techniques for practical implementation.

In this paper we present a new methodology for the formulation of efficient load signatures towards the implementation of a near-real time Non-Intrusive Load Monitoring (NILM) approach. The purpose of this work relies on defining representative current values regarding the 1st, 3rd and 5th harmonic orders to be utilized in the load signatures formulation. A measurement setup has been developed and steady-state measurements have been performed in a Low Voltage residence. A data processing methodology is proposed aiming to depict representative current values for each harmonic order in order to keep the load signature short and simple. In addition, a simple disaggregation scheme is proposed under linear equations for the disaggregation mode in order to examine the near-real time application of the methodology. The analysis indicates that the developed load signatures could be efficient for a per second application rate of the NILM algorithm. The results show that the higher harmonic currents facilitate the identification performance. Finally, the analysis concludes that for combinations that include appliances with intense harmonic content, the phase angle of the higher for harmonic currents should also be considered to the load signatures formulation.

a b s t r a c t A simple and novel methodology for the configuration of robust and distinct load signatures is presented in this paper. A technique based on a spectral analysis of the current waveform of each load is proposed, leading to load signatures consisting of three distinct spectral coefficients. Data required for each load signature come from only one period of the current waveform, under a sampling frequency at the order of kHz. A simple identification algorithm has also been developed in order to estimate the signatures adequacy. Each appliance has been measured and classified as an individual load, using a simple identification procedure. Data obtained from two different measurement sets show that the proposed technique provides unique data formations for each appliance tested by developing an efficient recognition procedure with high levels of identification rate. Due to the simplicity of both the database configuration and the identification procedure, the proposed methodology could be considered appropriate for online monitoring and real time smart grid applications. © 2014 Elsevier B.V. All rights reserved.

In this paper the Optimal Distributed Generation Problem (ODGP) towards energy minimization is solved for a large number of scenarios regarding power loss minimization. Load variations are taken into account by the formulation of different snapshots concerning the network's operational status. These snapshots refer to various load compositions and for each one the ODGP problem is applied. Load variations are formed stochastically under a uniform distribution while the initial loading conditions are considered as the mean load profile of the network. The solution algorithm relies on a Local PSO Variant and the results indicate that for not extreme load variations some specific nodes tend to participate in the majority of the different solutions. Thus, the analysis proposes a fixed solution that could yield the highest energy reduction despite the fact that it is not the optimal for each individual operating state with different load composition.

This paper examines the contribution of the Odd-Order Harmonic Currents to the development of robust and distinct Load Signatures that could, in turn, facilitate load identification in Non-Intrusive Load Monitoring (NILM) algorithms. The analysis is based on field measurements in various residences with single-phase grid connections. Measurements have been conducted both for stand-alone operation of each appliance and for various combination scenarios. The former enables the determination of the active and reactive power consumption, the instantaneous current values, and the magnitude of odd-order harmonic current under steady state operation, for realistic operating conditions regarding the operational time of each appliance. The latter, allows considering candidate aggregated signals for decomposition. The results indicate that the utilization of more distinct and enhanced Load Signatures, using odd-order harmonic currents, improves the efficiency of the identification procedure.