ABSTRACT. This paper introduces a machine learning-based risk-constrained portfolio optimization framework to arbitrage congestion with virtual bids in the wholesale electricity market. The proposed trading strategy aims to maximize the profit from the perspective of a proprietary trading company. A deep neural network is designed to estimate the difference between congestion spreads in day-ahead and real-time markets. A clustering algorithm is adopted to separate pricing nodes into a few groups, between which the congestion spreads can be exploited. We validate the proposed algorithmic trading strategy using publicly available data from California Independent System Operator (CAISO). The empirical results indicate that our proposed algorithm attains significant profit with modest portfolio budget and low risk tolerance level.

ABSTRACT. Real-time wide-area monitoring of smart grids demands a low latency data processing of power system data. To enable the low latency requirements and to avoid the large overhead of communicating a large volume of time-sensitive data to central processing units, distributed and local processing of data is a promising approach that can improve system monitoring functions. Data-driven state estimation in power systems is an example of functions that can benefit from distributed processing of data and enhance the real-time monitoring of the system. In this paper, distributed state estimation is considered over multi-regions, identified based on geographical distance and correlations among the state of the power system's components. Bayesian Multivariate Linear Regression (BMLR) combined with Auto-Regressive AR(p) process for distributed state estimation is considered over the multi-region power system. The performance of the distributed data-driven state estimation method and the role of regions are evaluated using the IEEE 118 test case under normal conditions as well as partially unobservable scenarios.

ABSTRACT. Non-technical losses (NTL) constitute a major issue in many countries, both developing and developed. NLT can be considered as a bad data detection problem. Thus, classical approaches like the weighted least square method and statistical tests can be used for the detection and identification of bad data. They are suitable tools when the topology of the network and its parameters are known. While this assumption is widely accepted in transmission grids, it is not clear to hold in distribution grids, where grid reconfiguration is common and parameter values have an important dependence on the ambient conditions.

In this paper, we leverage the latest advances in mathematical and computational tools to detect NTL in distribution grids. Thus, NTL detection can be implemented in an automated system that does not require human interaction. For dealing with it, we use off-the-shelf machine learning algorithms. In particular, we introduce a new architecture that combines a deep learning network and convolutional and recurrent neural networks.

A thoughtful set of simulations over a realistic dataset is performed and compared among other model-free machine-learning approaches, namely, support vector machine, random forest, and gradient boosted trees.

ABSTRACT. This paper discusses the data post-processing aspects of measurement-based load model estimation. Events measured by DFR (Digital Fault Recorder) are used for estimating exponential load models by least squares algorithm. Values are used for comparing the impact of different event filtering approaches. Events are filtered based on voltage and current unbalance ratio, residual voltage of disturbance, voltage deviation, and estimated load model parameter values. In addition, two different methods for calculating representative model are compared.

ABSTRACT. To achieve the goal of stabilizing carbon dioxide levels at acceptable levels under the Kyoto Protocol, it is essential to control carbon emissions by implementing regulatory interventions. Carbon tax and emission trading system (ETS) are two effective tools to regulate carbon emissions. Various studies have investigated both methods, yet few have considered the impact of these two policies on the electricity market and the social welfare. The purpose of this paper is to fill in the gap, namely to quantify the influence of the carbon tax and ETS on the electricity market. In addition, an optimal hybrid policy that combines both carbon tax and ETS is developed. Our case studies demonstrates that carbon tax, ETS, and the hybrid carbon policy can achieve the goal of reducing the carbon emission and increasing the social welfare. For hybrid policy, the carbon tax rate and the emission marginal cost of allowance in ETS are optimized properly to maximize the social welfare.

ABSTRACT. Ferroresonance severely impacts power quality and can also cause potential damage to electrical equipment when excessive voltages and currents are induced. Ferroresonance is rarely demonstrated provided that all three source phases are energized. However, when series-compensated lines are involved, a special type of ferroresonance can occur where the power system symmetry is remained.

In Sweden, there is an increased demand to connect wind farms at the series-compensated sections. However, the transformers that link these wind farms increase the ferroresonance risk during certain operational conditions. The transmission system operator must be aware of these circumstances to ensure safe operation, but the complex nature of such nonlinear phenomena makes it difficult to understand and hard to predict. Therefore, this work offers a plausible understanding of ferroresonance that involves series-compensated lines. It also develops operational guidelines that ensure safe operation by raising awareness of the high-risk operational conditions. Moreover, real-world ferroresonance field measurements are shown from an operational experience that violated the operational guideline. The transformer model implementation is essential when dealing with ferroresonance simulations. Therefore, the transformer model implementation is discussed and the simulation model is validated based on the real-world operational experience.

ABSTRACT. Linearization of the multi-phase power flow equations is useful in settings where solving the exact, non-linear equations becomes prohibitive, such as multi-period problems with long time horizons, or large-scale mixed-integer problems. For multi-phase radial networks, Gan and Low proposed a power flow linearization known as the ‘simplified unbalanced DistFlow’ equations. This paper shows that this linearization can be simplifie further by discarding off-diagonal variables and constraints, without affecting the results. This reduced the solve time across a set of 128 instances by 20% on average. The reduced formulation is shown to be identical to other linearizations reported in the literature, also known as ‘LinDist3Flow’. Finally, a study across 128 LV networks illustrates that the linearization tends to overestimate voltage rise due to generation and underestimate voltage drop due to load, leading to too conservative or optimistic dispatch results.

ABSTRACT. This paper aims to present the results obtained from a comparative analysis based on laboratory tests performed on different battery technologies (flow, sodium, lead carbon, lithium iron phosphate, lithium manganese and lithium titanate) for application in a transportable BESS. The laboratory tests followed the standard IEC 61427-2, dedicated to on-grid applications, and evaluated the endurance and electrical performance of each battery technology. The tests performed were: capacity tests with different current intensities (discharge and recharge), self-discharge, and cycle tests – according to the methods of Load-Following (LF) and Peak-Shaving (PS). Because the transportable solution of this R&D project is not restricted to a single application, the tests focus on the most probable applications. The batteries were tested during more than 24 months. The laboratory studies showed that the lithium-ion battery technologies are the most recommended for application in TBESS, due to the high cyclic performance and high energy and power density. It is important to note that LMO technology has reached almost 9,000 cycles (PS + LF), without reaching the end of its useful life. And the LFP and LTO technologies have already reached the mark of 6,930 and 4,158 cycles, respectively, maintaining the capacity above 95%.

ABSTRACT. Islanded microgrids are discussed in the context of new operational concepts for the public distribution grid to ensure power supply during disruptive events. Since the connection to upper grid levels is not available in such cases, the grid voltage must be provided locally. This can be achieved by using by grid-forming inverters in islanded operation mode. The temporary operation of such islanded grids in public distribution grids has not been established and the stability has not been fully researched. One related topic is the harmonic stability, which can be assessed using the impedance-based stability criterion. For ist application, frequency-dependent impedaces of grid-forming inverters are required. However, test procedures and corresponding test facilities are generally not available. Moreover, typical test voltage sources cannot be operated in parallel with grid-forming inverters without any additional synchronization. In this paper, a test setup is implemented enabling the measurement of the frequency-dependent impedance of a grid-forming inverters using such a typical linear voltage amplifier. The results confirm that impedances can be extracted with the test setup and measurement results are shown for commercially available battery inverters. Additionally, limitations of the test setup and aspects that should be considered upon usage are presented.

ABSTRACT. Abstract—This paper presents a three-phase optimal power flow (TOFP) based method for mitigating the voltage unbalance factor (VUF) in each of the three-phase busbars with the help of demand-side flexibility. It can be expected that issues with the unbalanced voltage can increase in the future distribution system with significant levels of single-phase Distributed Energy Resources (DERs). Unhealthy voltage unbalance status reduces the efficiency of the distribution systems, leading to energy loss and heat problems in electrical devices. This paper has proposed a three-phase OPF (TOPF) based local flexibility market (LFM) framework, where Distribution System Operators (DSOs) can utilize the upwards or downwards flexibility to manage the voltage unbalance. A modified IEEE-34 bus distribution network is used to illustrate the implementation of the proposed methodology, with the simulation results showing that the VUF value varies with the unbalanced load and connected photovoltaic (PV) levels, while the VUF drops below the limiting threshold when the flexible phase-by-phase flexibility services are introduced. This means that the proposed approach could inform modification of flexible service product in the future local flexibility market framework to help DSOs mitigate the unbalanced voltage.

ABSTRACT. This paper assesses the ability of four integral-based indices calculated using the post fault rotor angles, speed deviations, and accelerations to evaluate power system’s transient stability. First, the impact of the integration time for the calculation of each of the indices to provide an adequate assessment of the system stability status is assessed. Then, a more detailed evaluation and calculation of the accuracies achieved by the indices is done by looking into the direct relationship between the index values and the stability status of test simulations. Results show that the proposed indices are able to represent correctly the instability degree of the system to different extents, while at the same time identifying the limitations in their use.

ABSTRACT. As a result of the current restructuring procedure in power systems, renewable energy sources (RESs) are increasingly integrated into distribution systems. Respectively, while, the installation of RESs would benefit the system by securing the power supply, decreasing the power losses, and addressing the environmental concerns; their high penetration could cause operational issues in the system. In this regard, the traditional operational procedures seem not to be adequate for ensuring the voltage regulation in active distribution systems. In this respect, efficient scheduling of flexible resources should be taken into account by distribution system operators (DSOs) in order to mitigate the potential curtailment of RESs due to voltage issues in the grid. Consequently, this paper aims to develop a hierarchical management scheme to efficiently re-schedule the flexible resources in a multi-agent distribution system to address voltage issues raised by the high penetration of RESs. Respectively, Stackelberg game concept is employed to model the interaction between DSO and independent agents in the system, where DSO acts as the leader and agents are conceived as the followers. Finally, the proposed approach is implemented on the 37-bus test system to analyze the effectiveness of the proposed methodology in mitigating the voltage issues in the grid.

ABSTRACT. International efforts for transport electrification have contributed to an exponential growth of the electric vehicles (EV) market. In developing countries, the purchasing power of users is low, which restricts rapid switch to EVs. In this scenario, battery swapping station (SWS) models where the EV batteries are owned by a swapping station which is also responsible for replacing a discharged battery with a fully charged one has been gaining a lot of attention. However, battery swapping unlike other commodities is not straightforward because matching two batteries with similar characteristics is challenging and the investment costs for SWS become too high. In this paper, we develop a comprehensive and flexible battery swapping economic model that takes into account multiple cost components. We also propose an opportunity profit based framework for fair comparisons with home charging station (HCS) and commercial charging station (CCS scenarios). Our analysis suggests that with appropriate consideration of opportunity profit for the time saved through battery swapping, this model has a lot of potential to become successful in developing countries, particularly for the rapid electrification of commercial two- and three-wheelers.

ABSTRACT. This paper introduces an algorithm that quantifies an electric vehicle (EV) user's utility from a charging schedule. The existing utility formulations in the bidirectional charging literature consider only the final quantities of a charging option such as final state-of-charge (SOC), charging completion time or cumulative vehicle-to-grid (V2G) discharge. On the contrary, the formulation suggested in this paper considers the whole trajectory that ends with such final quantities. In this way, the algorithm accurately models the impact of i) the future mobility scenarios and ii) the battery degradation due to V2G discharge. The developed utility function is applied in on-the-move and park-and-ride charging scenarios in order to estimate user utility from multiple schedule alternatives that have identical final SOC and cumulative V2G discharge. The results demonstrate that, despite the identical final conditions, the schedules denote significantly different utilities to the EV users due to the uncertainty of the parking duration, the aimed range in the following trips, the convenience of additional charging between the current and next planned charging events, and the battery degradation's dependence on depth of discharge. These results indicate the importance of an assessment based on the entire schedule of a charging option rather than its final conditions.

ABSTRACT. The global electric vehicle uptake and the in- creasing electrification of final energy consumption could bring grid management issues to the distribution system in a near future. Demand response programs are seen as a direct solution through the aggregation of distributed flexibility resources, ex- pecting electric vehicles to be one of the main flexibility sources. However, the charging demand depends on the variability of daily drivers’ behaviour, and calculate the aggregated flexibility potential from forecasted individual sessions may result in a highly inefficient process and uncertain result. Therefore, to reduce this uncertainty, this work proposes a methodology to discover distinct user profiles with a model-based clustering method followed by the aggregation of the clusters into daily user profiles. The clustering method uses only the two basic connection variables available for any charging infrastructure: the connection start time and the duration of the connection. The method has been tested with data from the public charging infrastructure of the city of Arnhem, in The Netherlands. The clustering process has resulted in 45 different clusters aggregated into 7 user profiles, some of them with very low variability and therefore clearly useful to consider in flexibility programs.

ABSTRACT. Many bus operators worldwide have started with the electrification of their fleets. Analysis of the total cost of ownership is an often-used tool in this process allowing the bus operators to compare different technologies, find the cost optimum composition of their fleet and make strategic decisions. This paper provides a unique combination of analyzing the total cost of ownership for two electric bus depots depending on the impact of route and charging scheduling. Two different approaches to both route and charging scheduling were analyzed enabling a quantification of their effect on the total costs. As the analysis shows, the optimized scheduling can have a significant effect on the costs, emphasizing the importance of intelligent management systems for the future electric bus depots. Additionally, this paper provides a sensitivity analysis investigating the effects of diesel and electricity prices, CO2 tax and prices for electric buses on the total cost of ownership and on the break-even point compared to the conventional fleets. The analysis was conducted using real timetables from two existing bus depots in the City of Hamburg in Germany.

ABSTRACT. The growth of Electric Vehicles (EV) penetration into the distribution network around the world is expected to reduce greenhouse gas emissions from the transportation sector. Simultaneously transportation electrification is becoming a point of concern for power utilities and Distribution System Operators (DSO). Because of this, a simulation framework to understand the system conditions with extensive EV penetration, from the perspective of DSO, is needed.

For this purpose, the modified CIGRE-MV 14-node test distribution network was modeled and simulated using the real-time open-source simulator, Dynamic Phasor Simulator (DPSim). It offers features such as Common Information Model import, real-time capability, and interfaces for communication with other simulators. This makes the simulator an ideal choice for performing real-time power flow simulation studies and controller hardware-in-the-loop simulations. The power flow data is communicated to the Power Management (PM) algorithm for analysis with the help of lightweight Message Queuing Telemetry Transport (MQTT) protocol.

This setup is demonstrated by applying it for analyzing the impacts of increasing penetration levels of EVs in the network that results in critical loading conditions during the peak-load hours. The anticipated impact is addressed by power injections from the Battery Energy Storage System (BESS) determined and initiated by the PM algorithm.

ABSTRACT. Rooftop photovoltaics (PVs) and electric vehicles (EVs) have gained significant attention recently due to their key factor of decarbonising the electricity and transportation sectors. Introducing new charging stations and increasing the integration of PVs into power distribution systems can bring new operational challenges, including overvoltage and voltage unbalance. Therefore, this paper investigates the impact of EVs connection on an unbalanced power distribution system considering local voltage control of PV inverters. Several cases were studied using the unbalanced IEEE 37-node test feeder over a 24-hour simulation period. The unbalanced load flow calculation results argue that EVs connection can increase the severity of voltage unbalance, and PVs connection can potentially reduce the voltage unbalance issues. On the other hand, the reactive power capability of a PV system does not necessarily minimise power losses, especially at a high penetration level of PV.

ABSTRACT. With the increasing use of information and communication technology in electrical power grids, the security of energy supply is increasingly threatened by cyber-attacks. Traditional cyber-security measures, such as firewalls or intrusion detection/prevention systems, can be used as mitigation and prevention measures, but their effective use requires a deep understanding of the potential threat landscape and complex attack processes in energy information systems. Given the complexity and lack of detailed knowledge of coordinated, timed attacks in smart grid applications, we need information and insight into realistic attack scenarios in an appropriate and practical setting. In this paper, we present a man-in-the-middle- based attack scenario that intercepts process communication between control systems and field devices, employs false data injection techniques, and performs data corruption such as sending false commands to field devices. We demonstrate the applicability of the presented attack scenario in a physical smart grid laboratory environment and analyze the generated data under normal and attack conditions to extract domain-specific knowledge for detection mechanisms.

ABSTRACT. As the power grid is a cyber-physical system, it is vulnerable to cyberattacks. Therefore, research on the cybersecurity of the grid is in critical need. In response to this, research has been focused on the various cybersecurity issues of the grid, including the formulation of cybersecurity frameworks. However, it is noteworthy that a significant proportion of the literature is concerned with centralized mechanisms, which are prone to single point failures. The literature also appears focused on the transmission system, so that certain solutions are impracticable at the distribution level. In this paper, a cybersecurity architecture is proposed for the electric energy distribution system. The proposed method is a decentralized mechanism for attack correlation and mitigation. A prototype of the proposed architecture is also presented, simulations of which are performed using the IEEE 13-Node Test Feeder. The simulations have validated the performance of the prototype, and consequently the advantage of the proposed architecture.

ABSTRACT. One of the key aids in ensuring security in the substation is data collection. Collected substation data can be applied to data analytics to infer the substation’s security status. However, most of the data collected are measurement logs and Ethernet-based traffic data. The industrial network market share shows that serial-based protocols still exist, and it will take some time before they are phased out. Also, data from serial-based traffic is sourced mostly from SCADA; thus, when SCADA is compromised, the data inferred from it can be misleading. To the best of our knowledge, there is no work on data collection involving serial-based protocols such as DNP3 serial, Modbus-RTU and IEC 60870-5-101; hence, we propose a custom serial sniffer that captures data from serial devices and logs them. Such complementary data can be transmitted to a Security Operations Centre (SOC) to provide a second view to confirm the substation’s status. Our bench-marking of the serial sniffer shows negligible utilization of CPU and memory resources for DNP3 serial and Modbus-RTU. Our sniffer code can be embedded into serial-to-Ethernet protocol converters due to its lightweight nature. Our work can complement data collection within substations until the migration to Ethernet-based protocols is complete.

ABSTRACT. Denial-of-Service (DoS) attacks in wide-area control loops of electric power systems can cause temporary halting of information flow between the generators, leading to closed-loop instability. One way to counteract this issue would be to recreate the missing state information at the impacted generators by using the model of the entire system. However, that not only violates privacy but is also impractical from a scalability point of view. In this paper, we propose to resolve this issue by using a model-free technique employing neural networks. Specifically, a long short-term memory network (LSTM) is used. Once an attack is detected and localized, the LSTM at the impacted generator(s) predicts the magnitudes of the corresponding missing states in a completely decentralized fashion using offline training and online data updates. These predicted states are thereafter used in conjunction with the healthy states to sustain the wide-area feedback until the attack is cleared. The approach is validated using the IEEE 68-bus, 16-machine power system.

ABSTRACT. Synchrophasor technology contributes significantly to the power system transformation into smart grid since it enables the wide area monitoring and control concept. At the same time, its integration to the smart grid imposes new cyber security challenges in these cyber-physical systems. The key element of the synchrophasor technology is the Phasor Measurement Unit (PMU) which provides synchronized measurements to the control center with a fast reporting rate. The PMU measurements are kept in a Phasor Data Concentrator (PDC) which collects and time aligns measurements from different PMUs. The communication between PMUs and PDC is defined by the synchrophasor communication standard, C37.118.2-2011, which lacks security mechanisms and hence it is vulnerable to cyber-attacks. Consequently, an intrusion detection system (IDS) that can detect PMU measurement manipulation is necessary to maintain the integrity and reliability of the PMU measurements. In this paper, a Synchrophasor Specific Intrusion Detection System that utilizes a PMU’s behavioral model to detect data manipulation attacks against PMU measurements is proposed. The proposed IDS approach is tested and validated using a hardware in

ABSTRACT. One of the main challenges of residential demand-side management (DSM) is achieving enough aggregated flexibility to participate in the different energy markets. Aggregators enable such operation by managing a large portfolio of individual customers. However, this also involves a continuous flow of information from the individual users to the Aggregator. Due to the sensitive nature of energy data, from which behaviour and usage patterns can be inferred easily, not every customer is willing to share this information. What is more, considering the current efforts to protect personal data, the limitation of data collection and exposure should be a priority of any modern system. To achieve such functionality, this work presents a novel privacy-friendly aggregation algorithm that combines market-based control (MBC), public-key cryptography and secret sharing, with the goal to provide the same energy aggregation services but without revealing the individual user profiles.

ABSTRACT. The development of modern LV distribution networks is characterized by the growth in the number of converter-interfaced prosumers using electronic elements with switching frequencies in the kHz range. It causes the emission of so-called “supraharmonics” into electrical networks. For the analysis of the propagation of supraharmonics it is necessary to known the corresponding frequency-dependent parameters of LV distribution lines. Modern LV distribution lines are mainly cable lines and designed for the operation at the fundamental frequency (e.g. 50 Hz). The calculation of line impedances at all other frequencies is the subject of special studies. The paper deals with the analysis of positive sequence series impedances of LV sector-shaped cables in the frequency range 2 to 9 kHz. The analysis is based on FEM simulation results. Novel formulas are suggested for the simplified determination of series resistances and inductances of LV 4-core cables with sector-shaped aluminium conductors.

ABSTRACT. Electric Arc Furnace (EAF) introduces several problems in the power system such as harmonics, voltage flicker, unbalance and voltage fluctuation. Research community has found the solution for these problems by using SVC and STATCOM, but these solutions do not solve all frequency variation issues, which EAF causes in the weak power system. Weak power systems with low inertia and frequency regulation capacity present frequency problems due to active power demand of EAF. According to research presented in this paper, the frequency regulation can be realized by using a special and low energy capacity battery energy storage system (BESS) synchronized with the EAF. It is possible to reduce the frequency problems of the EAF cost-efficiently and therefore, EAF can be installed also to the weak power system.

ABSTRACT. The new underground distribution network of University City at UNAM was designed to be a smart grid, which in addition to operating in an automated and efficient way, allows the implementation of new technologies and the development of all kinds of research. This article presents the network characteristics as a first instance, and shows how its modeling and implementation through a commercial simulator provide a tool for conducting energy quality studies, planning and integration of new elements to the network. As a sample, a substation was chosen within this network which presents power quality problems and also requires planning for a future load increase.

ABSTRACT. The rapid growth of power electronics-based devices over the past years has increased the harmonics distortion along the power grid. This phenomenon may pose challenges to distribution network operators for managing harmonic levels. This paper proposes an approach to use grid-connected photovoltaic inverters as active filters. The proposed control algorithm consists of a proportional resonance controller for the inner current loop, notch filter for harmonic extraction and an automatic gain functionality to assist the state of the inverter. Active filter operation is activated based on irradiance conditions and system harmonic levels. The main contributions of this work consist in: applying this control scheme to multiple inverters connected at different locations, and using real irradiance and load data to assess the benefits of the proposed application.

ABSTRACT. This paper carries out an inter-area oscillation study using the accurate model of the Greek power system. Taking into account the complexity of the system under examination, an automatic toolbox is developed that can define the oscillation modes and provide a graphical representation in order to detect if there exist any area separation in the selected frequency. As case study, an inter-area oscillation between Greece and Bulgaria was simulated. The results showcase that the toolbox can effectively define the frequency, amplitude, damping and modes of the system and detect if there exist any significant phase angle difference between the buses, which is an important indicator of the stress in the power system. This information can be used to determine the critical points for employing wide area monitoring and damping control systems in order to damp oscillations and improve the overall reliability of the power system.

ABSTRACT. This paper summarises results and experiences from several demonstration projects across European countries in the field of battery energy storage system (BESS) integration to the power system. These research projects are selected among research institutes and universities that are part of the European Energy Research Alliance (EERA) Joint Program on Smart Grids. The paper categorizes these projects according to the demonstrated applications of BESS and then reviews specific aspects of each project. This paper provides an opportunity to find out the summary of the most recent results as well as challenges and open research questions in projects focusing on different BESS application in the power system.

ABSTRACT. In today’s power systems, different trends can be observed, which reduce the total system strength. The result is a weakened reference voltage at the terminals of inverter-based generation. Present-day control systems have been designed under the assumption of a stiff reference voltage. Due to the reduced stiffness, it imposes a challenge for the inverters’ synchronization units. Without a precise synchronization, the entire converter control will become unstable. Therefore, advanced synchronization units are necessary. The performance of selected synchronization units, i.e., the phase-locked loop (PLL) and frequency-locked loop (FLL), is analyzed under weak grid conditions. When utilizing the PLL, additional filtering is beneficial to extract the positive-sequence voltage vector. In general, the FLL turns out to be a more robust solution. Furthermore, the dependency of the voltage at the coupling node on the impedances of the converter and the grid is investigated analytically. In the case of weak grids, the influence of the load on the node voltage must be considered. Thus, a novel impedance ratio is introduced to assess the system strength under weak grid conditions.

ABSTRACT. This paper deals with positive- and negative-sequence current control of voltage-source converters. An enhanced weak-grid tolerant state-feedback controller design based on direct pole placement approach is proposed. The controller is synthesized from a body of literature on multi-frequency current control of grid converters. The resulting design yields consistent dynamic performance for varying grid strengths, and remains stable even under very weak grids. Due to the explicit parameterization of the pole locations, complex optimization methods often associated with robust control designs are avoided, which simplifies the controller design process.

ABSTRACT. This paper presents a methodology to estimate the distributed circuit topology and analyze the occurrence of faults at different locations in a distribution circuit. The experiments discussed here show the performance of a convolutional neural network (CNN) based circuit topology estimation model. The proposed approach uses a data-based approach to estimate the circuit configuration with faulty and normal data. The results are compared with that of a standard Support Vector Machine (SVM). Furthermore, fault classification using an SVM is analyzed. The effectiveness of the proposed method is tested using data obtained from power simulations on the modified IEEE 123 bus system using MATLAB Simulink.

ABSTRACT. Grid converters need information about the point-of-common-coupling voltage for their control. The voltage is typically measured for current and ac-bus voltage control, and for synchronization. In order to increase reliability by decreasing dependence on this measurement, this paper proposes a multifunctional grid-voltage sensorless controller that can operate either in current- or voltage-control mode. The control system is designed to be robust against variations in grid strength, covering strong and very weak grid conditions. The controller is experimentally tested and its robustness is verified in these conditions. The results show that sensorless ac-bus voltage control can provide grid voltage support.

ABSTRACT. Integration of power electronic-based renewable energy sources into weak grids brings new concerns regarding the harmonic stability of modern power systems. This paper studies harmonic stability and impedance characteristics of power synchronization control (PSC) which has been introduced for control of converters under weak grid operating condition. Due to the varying grid impedance in weak grids, the harmonic stability of converters with LCL output filters would be a demanding prospect. Thus, this paper addresses this issue by studying the admittance characteristics of PSC based on the dynamics of the control scheme, the delay and the sampling frequency of the control system, and the linearized model of the grid-connected converter. Finally, the critical grid inductance and the sufficient conditions for achieving harmonic stability are introduced. Time-domain simulations are carried out to support the proposed harmonic stability analysis.

ABSTRACT. One of the distinctive features of smart grids is feeder automation. Fault location is an important part of this feature and has become an essential function for distribution system operators. Reliable fault location expedites the restoration of power following an outage caused by a permanent fault. The most common type of faults in distribution networks is the single phase to ground fault. To locate earth faults in non-effectively earthed medium-voltage distribution networks, a number of methods have been put forward among which methods that are based on fault passage indication appear to be promising. This paper discusses the technical apparatus required for implementing two FPI-based methods in practice.

ABSTRACT. Successful cyber attacks targeting the electric power system can have severe impacts on the security of supply. Quantification of the impact is needed to develop proper countermeasures that enhance power system resilience. The aim of this paper is to present a method to assess the impact of manipulated control signals sent to distributed energy resources (DERs) on stable power system operation. Therefore, manipulation scenarios and assessment criteria are developed. A case study is performed with the CIGRE medium voltage benchmark grid to test and verify the proposed method using a dynamic time-domain simulation environment. Results show that critical scenarios exist, in which the attacker can force the bus voltages out of the voltage band and thus trigger plant protection of other DERs. The criticality of the scenarios is dependent on the grid topology as well as the operating point and the amount of compromised DER power. The proposed method is able to rate the criticality of these scenarios and can be applied to other grid models to further investigate the impact of cyber attacks on stable power system operation.

ABSTRACT. Graph signal processing has been shown to provide a unique platform and new perspective for representing and analyzing power system measurements. In this paper, the effects of various cyber and physical stresses as well as their differences on the graph signals of the system and their spectral domain are presented and discussed. Specifically, graph Fourier transform (GFT) and the local smoothness corresponding to the graph signals are presented as tools for the characterization of the effects of stresses. Particularly, various cyber-attacks including denial-of-service, replay attack, ramp attack, and delay attack, are considered on the time-series associated with the voltage angle measurements. An analysis of the relation of the degrees of the nodes under cyber-attack with the GFT of the associated graph signal has also been presented. Finally, a comparative analysis of the effects of cyber and physical stresses on the graph spectrum in the context of detecting and locating stresses in the smart grid are presented.

ABSTRACT. A good functioning of power transformers must be assured, since they play a vital role in distribution and transmission systems, an early diagnosis of incipient faults in such equipment enables their predictive maintenance. Traditional dissolved gas analysis interpretation methods are the most popular approach for fault diagnosis in power transformers, however, their application suffers from several drawbacks; some studies have applied machine learning techniques towards overcoming them. This article introduces a novel methodology for incipient fault diagnosis in power transformers that uses artificial neural networks (ANN). It considers feature selection for multilayer perceptron (MLP) networks in a cascade structure for classifying faults. Its performance was compared with those of two other ANNs, i.e., a traditional MLP and another cascade structure with no feature selection. The methodology is of simple implementation, highly accurate, and capable of correctly classifying over 85% of the test samples.

ABSTRACT. In the context of carbon dioxide emissions reduction, hybrid energy systems including both classical and renewable power plants require additional technologies to achieve depollution targets. Power-to-methane concept gains a lot of popularity, enabling both carbon dioxide utilization and renewable energy surplus management on long term time frames. This paper investigates the potential of power-to-methane technologies to contribute to renewable energy penetration augmentation and more reliable load supply. Moreover, the trends of renewable energy share in each sector are presented and discussed, in the perspective of cross-sectoral integration.

ABSTRACT. Tap-changing transformer models used in power system studies tend to neglect the change in the short-circuit impedance of the device at the different tap positions. However, the variation of the short-circuit impedance can be significant in transformers with a wide voltage-regulation range. Fortunately, in those cases in which the voltage variation of ±5% is exceeded, the manufacturer is obliged to test and provide data for the short-circuit impedance at terminal taps. The present contribution provides an update of a recently proposed model of the tap-changing transformer so that it can take advantage of the additional information available in the aforementioned cases. The increased accuracy of the resulting model has the potential to improve the quality of the results from power system studies with embedded tap-changing transformers.

ABSTRACT. This paper formulates, and then adapts a Mixed Integer Linear Programming MILP model for partitioning a graph into a desirable number of internally connected sub-graphs to identify the maximal coherent cut-set for a power grid. For a particular operational snapshot of the grid, a coherent cut-set refers to a set of lines within the grid with the same flow directionality whose removals splits the grid into exactly two internally connected sub-grids. Such cut-sets are of interest as they can represent operational vulnerabilities. This paper suggests a novel methodology that directly detects the maximal coherent cut-sets for power grids. Two important applications are considered for the suggested methodology: i) a novel N-k analysis, ii) identifying the worst coherent cut-set that each line is involved in. Moreover, a novel Sankey diagram-based visualisation technique is introduced for portraying the grids and the detected cut-sets, resulting in improving the situational awareness in the control room.

ABSTRACT. This paper proposes an algorithm for real-time congestion management in a distribution network. It sets up a peer-to-peer market allowing the distribution system operator to inject network charges. This enables him to obtain flexibility from distributed agents with heterogeneous preferences. These network charges vary in real time and are related to the network's congestion. Prosumers minimize their cost function, and find a consensus through ADMM decomposition. This formulation allows the management of the large number of agents present in the distribution networks only using one price broadcast by the DSO to prosumers. We demonstrate with the CIGRE low voltage test case that this strategy is efficient to manage congestion and presents limited sub-optimality compared to the OPF.

ABSTRACT. In this work, we present a novel modular approach for control of an energy storage device towards multiple objectives simultaneously. The proposed control method is explained and its usefulness is demonstrated using simulations of a 16 house neighbourhood. The simulation results of the proposed approach are compared to those using a standard battery setup. It is concluded that, when a storage device is divided in virtual storage devices, which each have their own goal (e.g. peakshaving, grid-balancing and economic benefits in energy markets) and their own (dedicated) fraction of the storage device capacity, improvements in storage device utilization can be achieved.

ABSTRACT. The target of this paper is to demonstrate the prospective of Synchronous Condenser (SC) and Super Conducting Energy Storage (SMES) for catering additional inertial support in renewable dominated power systems. In recent times, prolific renewable penetration in many power systems has caused numerous conventional power plants to retire. However, unlike conventional synchronous generators, several types of wind turbine generators do not provide inertial support following a disturbance. This causes overall reduction in system inertia and elevates the risk of unacceptable system response subsequent to a large disturbance. However, SC and SMES can emulate the inertial behavior and cater supplementary inertial support following a contingency. To this end, this paper proposes analytical sizing schemes to determine the minimum required sizes of these devices to keep the Rate of Change of Frequency (ROCOF) above a predefined value. Simulations show that the proposed methodology yields satisfactory system performances and system ROCOF stays below the preset limit. In addition, detailed cost analysis is executed for these devices to figure out the more financially viable and worthwhile choice

ABSTRACT. Traditional district heating networks have been built around few central heat production units. Now, many more decentral heat sources based on renewable energies or waste heat are being included to reduce carbon emissions. Given the resulting complex flow patterns in the network reliable and fast heat grid state estimation becomes mandatory for efficient grid control. In this paper we first show an approach to reduce computational efforts for various grid computations by summarising pipe segments of the network. We then develop a probabilistic state estimator based on the reduced model by locally linearising the non-linear grid equations around the best state estimate. We show that the linearisation approach with a significantly lower computational burden achieves a prediction quality comparable to those of a sampling-based Monte-Carlo approach that uses the full model. This allows state estimation to become an online routine even in complex heating networks.

ABSTRACT. Nowadays, high penetration of photovoltaic (PV) generation is being integrated into the power system due to its economic and environmental benefits. However, this may affect the angular stability of the system in particular when the power systems are weak. Motivated by this issue, in this paper angular stability analysis for two-area benchmark system is carried out with a PV penetration level of 30%. Comparisons of the stability impact when the PVs are online and offline are presented. It is demonstrated by evaluating the damping ratio and clearing critical time index that both the PVs’ location and industry requirements can mitigate the negative effects caused by the displacement of conventional generators.

ABSTRACT. There are several challenges along the path towards 100% renewable-based power systems. One of them is ensuring the system's frequency stability and control. Provided they are large enough; isolated power systems represent an excellent testing ground for stability-related research due to the similarity between their dynamics and those of a continental-size power system while keeping a manageable size. In this work, the recently published Cape Verde Benchmark System is used to analyze the effects of phasing out the different thermal units on frequency stability. It is assumed that those units are substituted with wind generation, and then the possibility of keeping them running as synchronous condensers (SC) for inertia support is explored. A relevant scenario with 70% renewable penetration is defined and analysed after a sudden load increase (10%) and a single-line short circuit fault. The results show the benefits of keeping Synchronous generators as SC, providing a certain level of inertia to support frequency stability. Additionally, the study also points out that to fulfill the replacement of the role of SGs, a primary frequency reserve needs to be implemented to coordinate with SCs for secure frequency stability in renewable-based systems.

ABSTRACT. We consider the problem of frequency synchronization and power sharing in a droop-controlled autonomous community microgrid. We address this problem by developing models of hierarchical community EMS, which provide synchronization conditions for an autonomous microgrids community by means of graph theory and Kuramoto-based consensus algorithms for multi-agent reinforcement systems. In our research, we suggest that there exists a natural corollary between Kuramoto’s oscillators and the consensus algorithm. These two models can be used in tandem to craft a much more robust control algorithm. Moreover, we consider the optimal consensus of multi-agent systems using reinforcement learning control. The control objective is to design the controllers for each agent such that all the agents will be consensual with the leader agent. Our analysis is further validated via a simulation example of a community microgrid based on the CIGRE benchmark medium voltage distribution network.

ABSTRACT. This paper frames itself in the area of solar energy where harvesting is conducted with airborne moving farms. The goal of these farms is to harvest radiation to reach the storage capacity in the shortest time possible. To that end, we propose the use of reinforcement learning (RL) as a means to optimally plan the path of an airborne solar farm. In particular, the airborne solar farm dynamically plans its path driven by the forecasted irradiation; this schema allows minimization of the collection time by maximizing the irradiance of the farm while taking into consideration the costs of airship movement. The use of reinforcement learning promotes the quick adaptation of the airborne farm to the dynamically changing ambient conditions leading to increased radiation harvest yield. The proposed irradiation-driven RL planning method is tested on a set of various locations that exhibit characteristically different weather patterns. Our results demonstrate that the benefit of using RL is maximized as compared to static solar farms.

ABSTRACT. In this paper, the role of CCHP technologies for increased integration of variable renewable energy (VRE) into a low carbon neighborhood energy system has been explored. Microgrid model is developed for a hypothetical model community in Trondheim city of Norway. The model run minimises the total system cost and optimises both the investments and operation costs of the microgrid. The analysis is done for two distinct scenarios: using the main grid as a balancing source or connected mode and (2) without balancing source or in island mode. The results reveal that the main grid is a least-cost balancing source for integrating VRE in a microgrid. The optimal VRE integration level, however, is 65% and 79% in connected mode and island mode, respectively. The contribution of the main grid is dependent on the assumed grid connection capacity. The results also showed that hydronic heating is a least-cost solution compared to direct heating and has the leverage to integrate more VRE.

ABSTRACT. Battery energy storage systems (BESSs) for residential photovoltaic (PV) prosumers are becoming more popular as battery prices are decreasing, as a means of increasing self-consumption. When considering prosumers that also own an electric vehicle (EV), smart charging can also fulfill the same objective. This paper assesses the benefits of a PV prosumer with an EV under two options: installing a BESS or applying smart charging. Smart charging is applied with the goal of reducing energy imports by charging the EV when there is excess of production. The effect of using BESSs of varying inverter and storage size is also investigated, when operated under a simple energy import minimization logic. We use a yearly dataset of real 5-min measurements collected from a Danish household with a 6 kWp PV plant. The prosumer also owns a 62 kWh Nissan LEAF. We find that smart charging increases self-consumption from 29% to 54%, leading to yearly savings of 220 euro. Through a sensitivity analysis we find that a BESS of 8 kWh / 2 kW yields similar yearly results. Given the approximately 20 times higher payback period of the BESS we conclude that smart charging is economically much more attractive.

ABSTRACT. A critical issue in a hybrid energy storage system (HESS) is the control strategy, especially the power distribution between the individual energy storage devices. In this paper, the power distribution strategy based on the low-pass filter (LPF) controller with a variable time constant is introduced. The adjustable range of the variable time constant is determined by the spectrum analysis of the imbalanced power in a stand-alone household-prosumer system. In addition, the variation of the time constant is based on the feedback of the state of charge (SoC) of the supercapacitor (SC). The simulation results show that the power distribution strategy keeps the SoC of the SC in a moderate range and utilizes the SC more properly. A scaled-down experimental setup is built to verify the effectiveness of the power distribution strategy and the simulation results. Therefore, the proposed power distribution strategy ensures the effective operation of the HESS, avoids the unnecessary enlarging of the SC, and achieves cost reduction.

ABSTRACT. Renewable energy sources (RES) are currently being deployed worldwide with the focus to reduce grid load, secure tax incentives, and promote RES. Meanwhile, the Dutch government has introduced a new regulation for net-metering policy and nearly zero energy building (n-ZEB). To combat these new regulations, wind turbines and photovoltaics (PV) panels along with a battery energy storage (BES) are placed to improve the tuning possibilities of RES and load demand. An energy management system (EMS) is required to optimize the performance and lifecycle of the BES and simultaneously reducing the electricity bill. In this paper, a mixed-integer linear programming (MILP) optimization problem is formulated to schedule energy profile for the BES considering the day-ahead predicted RES and load demand. Henceforth, a prediction algorithm is designed to forecast the day-ahead prediction of RES and load demand from the day-ahead weather forecast. A case study has been performed for 130 house apartment building in Alkmaar, the Netherlands. The simulation results show at least 90% correlation between the weather parameters, RES, and load demand. Further, results demonstrate the effectiveness of the proposed EMS for different seasons throughout the year.

ABSTRACT. Common Information Model (CIM) is a well established open standard in the power systems domain due to its common semantics used to construct communication messages between the applications. It is considered as one of the enablers for smart grids because of its robust framework for accurate data sharing, merging, and transformation into reusable information. It has multiple utilisation ranging from developing an interoperable software application to a cyber-physical system. However, CIM provides a basic framework for information exchange, and it needs to extend as and when new business requirements arise. The latest power system digital twins' trend to establish smarter energy grids highlights new requirements for CIM utilisation. In this context, this paper aims to offer a comprehensive understanding by summarising and categorising the research on CIM's practical use for interoperability in smart grids.

ABSTRACT. The modern LV distribution networks are characterized by high presence of converter-interfaced prosumers using electronic elements with switching frequencies in the kHz range. It results in the harmonic emission of the current and voltage components at the frequencies over 2 kHz (“supraharmonics”) into electrical networks. Unbalanced operating conditions of three-phase prosumers and the high part of single-phase installations produce the zero sequence components both for triplen and for other harmonics. The emitted zero sequence components flow in the phase and neutral conductors of distribution lines and can propagate in the earth if the neutral is grounded. The spread of zero sequence harmonic components in LV electrical networks is substantially depending on the frequency-dependent zero sequence impedances of LV distribution lines, which are mainly cable lines. These parameters are not present in standard cable catalogues and are the subjects of special studies. Zero sequence series impedances of LV sector-shaped cables with aluminium conductors are analyzed in the paper for the frequency range 2 to 9 kHz, using FEM simulation results. Novel formulas are suggested for the determination of series resistances and inductances in the frequency range under study.

ABSTRACT. Electrical equipment constitutes a considerable economic asset value for the transmission and distribution system operators. The electrical equipment has long intrinsic lifetimes, most of which exceed 40 years. However, some components of equipment age faster than others, or become obsolete due to the evolutions of the technologies used and induce premature replacement of the complete equipment. Economic, environmental or safety requirements may also be imposed and make the compliance of certain equipment incompatible with the existing or new regulations. The panel focuses on power grids electrical equipment lifetime extension strategies. Within a CIRED working group, a review is performed on the currently used models for lifetime extensions. Panelists describe: • Iifetime extension of equipment and/or substation solutions alternative to the renewal of equipment; • lesson learned, good practices of lifetime extension options for electrical equipment; • obstacles and/or limitations of electrical equipment life extension strategies.

ABSTRACT. The integration of renewable resources raises multiple challenges to the microgrid operation due to their intermittent characteristics with high levels of uncertainty and complexity, including voltage and frequency variations, stability, and protection. Future intelligent microgrid will become more complex as a cyber-physical ecosystem which would require an integrated paradigm for energy management and control systems based on various (big) data resources. This paradigm shift leveraged by emerging digital technologies enables cyber-physical solutions to enhance resilience and security for the microgrids, as well as for the coupled energy systems. This special session presents the latest development of cyber-physical solutions to secure and to optimize operation of the future integrated microgrids. Topics include the following: • PV system integration in microgrids: impact and solutions • New approaches for coupling of microgrids • Control Prototyping and Power Hardware in-the-Loop for Microgrids Assessment • Micro grid Distributed Control validation in an hardware in the loop environment