ABSTRACT. Decarbonization is a global value and the roadmap to the green future. The key enablers are defined as electrification, to build up a more efficient and low-emission energy system utilizing clean fuel, and digitalization to integrate different smart technologies and components in the system. At the dawn of victory over the global pandemic, we are preparing ourselves for a tougher challenge - climate change, from various aspects of the power system. In this panel, we will discuss the key implementation drivers and the challenges, including those related to regulation, market and security perspectives. The following specific issues will be addressed: 1. microgrids as building blocks for smart distribution systems 2. how new flexibility means can make power grids resilient by design 3. electrification such as converting the conventional and fossil fuel-based powertrains to electric and smart powertrains for road vehicles, marine, and aviation 4. the risk assessment based on cybersecurity aspect and the power market revolution to develop the future carbon-free electrified energy system 5. perspective of future aviation and marine power systems, in light of hydrogen propulsion 6. Issues and outlook of local energy market for a decentralized power system

ABSTRACT. Active power curtailment is a cost-effective technique for mitigating overvoltage issues as a consequence of distributed generation. However, many solutions treat prosumers at highly sensitive parts of the grid unfairly. The best solution to this problem is to explore the non-linear behavior between active power and voltage to find the fair amount of power that needs to be curtailed to satisfy grid codes. Current state-of-the-art techniques for power curtailment are computationally expensive and case-specific. In this work, a fair value for power curtailment is achieved analytically that is both computationally efficient and generic. The results obtained analytically are validated utilizing an iterative algorithm, which provides a near optimal value for fair power curtailment. The results demonstrate that for single-phase and balanced three-phase networks, analytical and iterative methods have similar control actions and there is a 50% increase in distributed generation infeed at sensitive parts of the grid.

ABSTRACT. The distribution system operator (DSO) is responsible for ensuring, in a safe and reliable way, the connections of distributed generation (DG) projects in the electric distribution system (EDS). In this regard, the hosting capacity analysis is a crucial tool to assist the DSO in the decision-making process. Therefore, this work proposes a strategy to evaluate the effects of high penetration levels of electric vehicles (EVs) on the DG hosting capacity problem. The problem viewed from the perspective of DSO is formulated as a mixed-integer linear programming model, where the objective function maximizes the DG hosting capacity and simultaneously minimizes the EDS energy losses. To validate the proposed strategy, several penetration levels of EVs are considered under three different cases. Results show that when the initial installed DG capacity (without EV penetration) is compared with the installed DG capacities at the maximum penetration level of EVs, increases of 25%, 50% and 66.67% are achieved, in cases of uncontrolled charging scheme to EVs, controlled charging of EVs and taking advantage of V2G capabilities, respectively.

ABSTRACT. The increasing penetration of distributed energy resources in distribution networks is expected to bring operational challenges to network operators. It is commonly assumed that congestion management mechanisms are required to counteract the anticipated problems. However, careful and detailed simulations are required to determine the exact impact in distribution networks. Simulations need to consider a detailed prosumer modelling, their participation in the energy markets, an accurate network representation and the effect of network tariffs, among others. Only then different congestion management mechanisms can be fairly assessed. This paper presents such a simulation platform as a first step towards a holistic assessment framework, and presents some exemplary case studies as a preliminary assessment.

ABSTRACT. Integrating large amounts of distributed renewable energy sources, such as wind and solar, significantly impacts the operation of the electric power system. Accurate monitoring is required to avoid frequency or voltage instability due to contingencies. To meet this need and ensure reliable operation, the Romanian company Transelectrica installed a modern synchrophasor system. This paper describes the time-synchronized measurement architecture and data applications of the system. It shares several event examples where access to time-synchronized measurements provided significant value to operators and engineers. It also discusses the plans and next steps for time-synchronized measurements in the Romanian power system.

ABSTRACT. Among the many “low carbon electricity” initiatives launched by the Horizon2020 programm, the OSMOSE project aims to assess all possible ways to provide flexibility to the European power system. This article provides a brief explanation, in terms of organization and technological implementation of the Italian demonstrator (Work Package five) belonging to the OSMOSE project.

ABSTRACT. In order to improve performance of range estimation of electric vehicles, parameters that affect energy consumption should be determined accurately. This paper presents a parameter estimation methodology for electric vehicles based on least squares method. In this study, the power and angular velocity of wheels are measured from the vehicle directly. In addition to those, the directional velocity data is extracted from the GPS signal, in order to avoid the parameter dependency between the angular velocity and directional velocity. The proposed estimation process is validated by means of a drivetrain simulator, which calculates the power consumption of different types of vehicles.

ABSTRACT. Existing electrical networks are going through a transition and distributed energy resource, if not managed properly, can hinder this transition. Uncontrolled introduction of photovoltaic and electric vehicles in distribution networks would lead to substantial issues such as commitment mismatches, line congestion, voltage deviations, etc. This paper presents the use of a classical approach, mixed integer linear programming optimization, and a novel approach, adaptive multi-agent system, to solve the highlighted distribution side challenges by utilizing electric vehicles storage capacity. This comparison serves as a great tool to benchmark the performance of the under-development adaptive multi-agent system methodology.

ABSTRACT. The low voltage (LV) distribution network configuration is a mostly radial system, so the possibility of the voltage source inverters (VSI) being connected to a grid with a large grid impedance is high. Consequently, harmonic interaction between the VSI and the LV grid is a concern. This work studies the various parameters that influence the harmonic stability of the VSI in the LV distribution grid using the impedance-based method. Analytic expressions for the small-signal output impedance of the VSI are developed and validated through a comparison with simulation results from MATLAB/Simulink. The main parameters that influence harmonic resonance in the VSI-grid system are assessed to ensure the stable operation of the VSI-grid system. It has been revealed that the shape of the VSI output impedance in the high-frequency region is highly influenced by the switching frequency and the current controller bandwidth. Increased switching frequency results in the use of smaller VSI output filters, which can give rise to harmonic interactions and a deterioration in the stability of the VSI. Simulation and validation results will be shown to prove this analysis.

ABSTRACT. The disruptive innovations of electric vehicle (EV) technology create many problems that have not been surmounted satisfactorily. One such problem is the charging/discharging cycles of EVs. Although charging stations are the first choice in this regard, battery swap station (BSS) has emerged as a distinctive solution owing to the differentiation of electric vehicle designs and long battery charging times. BSSs have the potential to provide fast and effective solutions especially for the electrification of public transportation systems. This paper proposes the optimal battery charging schedule of BSS by calculating the battery swap demand of electric buses (EBs) run for public transportation with the help of a dynamic model that does not require high-resolution measurements which are inefficient for large bus networks. Also, the capacity estimation that BSS can allocate for the ancillary services is given. The presented approach is implemented using data of the Berlin public transport network.

ABSTRACT. Integrating renewable energy resources with dedicated energy storage systems in building regimes is becoming viable with numerous well-documented advantages. Multiple buildings, i.e., primary building(s) (with solar & storage) and secondary building(s) (ordinary building dependent entirely on the grid), have bilateral contracts with each other, resulting in an optimized exchange bilaterally and with the grid to reduce grid costs. Further, the allowance of electric vehicles (EVs) in these buildings enable added options to optimize the power flows, focusing on cost reduction. In this work, a mixed-integer linear program (MILP) is formulated to maximize the primary building's profits, along with the financial gains to secondary buildings and a fleet of EVs. The charging infrastructure at the primary building also facilitates the EVs’ load by accounting for factors such as feeder time-of-use (TOU) tariffs, contracted rates between the buildings, and EV charging prices. Results for actual load profiles for buildings in Auckland, New Zealand show that primary buildings earn up to 87 % of daily profits along with the 10–41 % daily cost savings for secondary buildings. The insights presented in this work can benefit local aggregators with options to maximize local profits lowering the grid dependence.

ABSTRACT. Increased renewable penetration propelled by several international climate change initiatives and political thrusts of sovereign governments has changed the energy distribution arrangements. Microgrids (MGs) and peer-to-peer (P2P) energy sharing are considered as attractive solutions. However, investigation of feasibility for such solutions is often dependent on large amounts of user data. We propose a web-based platform to explore potential sites and assist the process of establishing MGs and P2P energy sharing. The platform incorporates psychological aspects and behaviour of users to collect relevant user data. The envisaged system can work with minimal amount of data to provide intermediate results, which further improve based on availability of more information from users. Additionally, the tool is capable of being coupled with other optimisation and simulation solutions. The uniqueness of this system lies in the fact that this platform is competent of validating the viability of solutions in multiple dimensions including technical, economic, and legal.

ABSTRACT. Cloud computing is a megatrend in industry digitalization. The virtual power plant (VPP) is especially suited for cloud deployment since it can manage energy resources that are not co-located. However, there is a lack of research in VPP cloudification, especially related to multi-tenancy, a key technology behind the efficiency benefits of cloud computing. Cloudification is complicated by the fact that power systems and distributed energy resources frequently employ interfaces and protocols from an earlier era. In particular, IEC 60870-5-104 is a well-established standard for telecontrol in electrical engineering and power system automation applications. In this paper, a new perspective to VPP interoperability is proposed through the application of cloud computing. IEC 60870-5-104 enabled systems are integrated into third-party systems in the cloud. Examples of such third-party systems are the Internet of Things enabled distributed energy resources or electricity market information systems. This paper proposes a new problem definition of VPP-connected systems through establishing the said third party systems as tenants in a multi-tenant architecture. To define a general industry standards-based system architecture, both the software design methods and cloud computing architectural concepts are applied. An implementation is presented that is ready to integrate into systems in the field.

ABSTRACT. This work aims to describe an IT platform for the management of an energy community, where prosumagers (users capable of consuming, producing and storing electric energy) cooperate to maximize the self-consumption within the community, to increase efficiency of energy use and reduce energy costs.

ABSTRACT. The present trends in the area of smartgrids indicate that future transmission and distribution systems will heavily rely on digital and on communication technologies to operate. Indeed, the power systems are evolving progressively towards what is denoted as a cyber-physical system. This transition challenges the classical approaches for experimental testing and requires the development of testing platforms for cyber-physical systems able to capture the interactions between physical components, control and monitoring software and the communication infrastructure. This paper presents general considerations and requirements for a cyber-physical testing platform for power systems. The paper provides also an example of a testing platform specifying the characteristics of the major components and a summary of the experience matured in its setup and configuration. Finally, an example of an experiment on a notional smartgrid and the related results are reported.

ABSTRACT. The energy efficiency of the Wireless Sensor Network (WSN) deployed in a Smart Grid facility is a key criterion for the performance of a WSN integrated supporting system. Since small form factor sensors used in the Smart Grid have limited battery capacity, the energy saving for sensor nodes is a major design goal for WSN protocols. In the past, our strategy is to install a large number of core nodes with backhaul connections to ensure the reliable connectivity for low power sensor nodes. This strategy leads to a high overall cost for WSN deployment. In this paper, we propose a lightweight mesh networking solution to reduce the WSN core network deployment cost while keeping the same level of energy consumption for sensor nodes. We have developed a prototype system in our lab to demonstrate the feasibility of our proposed mesh solution using BLE chipsets. Comparing with the BLE-Mesh solution, our solution can save 40-70% energy for low power sensors and increase the channel capacity between 100-200% depending on the size of sensor data.

ABSTRACT. This paper proposes a new approach to determine a distribution network usage fee (DUF) for distributed generators. The methodology is based on distribution of costs associated with use of the distribution network, among all system users, including those with or without distributed generation (DG). To estimate the distribution network costs, impact of DG on the distribution network is valued. In order to achieve a fair and equitable electricity tariff, this methodology considers DUF calculation on the following DG variables: type (photovoltaic or wind), location, and size. Uncertainty associated with system demand and energy resources is probabilistically modeled through their cumulative probability functions. The proposed method is tested on a 33-node distribution system, under scenarios of variable location and capacity of the DG system. Furthermore, internal rate of return is calculated to evaluate DUF application on DG projects.

ABSTRACT. Residential electricity distribution grid capacity is based on the typical peak load of a house and the load simultaneity factor. Historically, these values have remained predictable, but this is expected to change due to increasing electric heating using heat pumps and rooftop solar panel electricity generation. It is currently unclear how this increase in electrification will impact household peak load and load simultaneity, and hence the required grid capacity of residential electricity distribution grids. To gain better insight, transformer and household load measurements were taken in an all-electric neighborhood over a period of three years. These measurements were analyzed to determine how heat pumps and solar panels will alter peak load and load simultaneity and hence grid capacity design parameters. Moreover, the potential for smart grids to reduce peak loads and load simultaneity, and hence reduce required grid capacities, was examined.

ABSTRACT. For sustainable development, potable water, green energy and fresh food are essential resources that are required by any society. Much devotion is given to energy and water infrastructures due to an interlinked networked system of global concerns under the energy-water-food (EWF) nexus paradigm. Generally, these infrastructure systems deliver power and water through separate and uncoupled systems. However, these are coupled infrastructures serving in their respective domains. Considering these interdependent networked systems, the power, water and food infrastructures required joint optimization. In this regard, a joint optimization program has been developed to optimally allocate the energy and water that includes power, water and co-generation facilities. Therefore, a mathematical optimization program for simultaneous co-dispatch of power and water from power generation, water production and cogeneration facilities is first developed. This optimization model runs considering production, demand, transmission and process limits. Furthermore, the inclusion of a co-generation facility helps in alleviating the binding constraints and leads to flat power generation and water production with reduced cost. Moreover, the proposed model is designed to follow a systematic approach in achieving the optimal results without violating the limits based on which the plants can easily achieve optimal control.

ABSTRACT. Photovoltaic (PV) power is an extensively used renewable energy resource, but its intermittent nature affects the power supply quality as it results in issues such as frequency aberrations and voltage variations. Battery Energy Storage Systems (BESS) are utilized to smooth out and resolve the fluctuation issues. However, a control method is required for BESS charging level regulation to prevent the need for larger storage systems and to extend its operational life through controlled charging/discharging. This paper proposes a novel solar power and battery state of charge (SoC) control technique through the incorporation of hydrogen electrolyzer (HE) fuel cell system. A machine learning based controller (MLC) is designed for dynamic control of the HE output to allow the dispatching of firmed PV power with controlled battery charging/discharging. The MLC takes the fluctuating power and various battery parameters as inputs and intelligently controls the HE output while obeying the imposed constraints. Results conclude that the MLC greatly reduces the PV power fluctuations and a comparison between the fuzzy logic control for SoC regulation shows that the MLC has better SoC management capability. The proposed methodology promotes the integration of hydrogen into the energy mix as a means for providing controlled solar power.

ABSTRACT. We examine the carbon mitigation potential of operational adjustments for a German university campus. To this end, we compare the modeled cost-optimal operation with CO2-minimizing dispatch plans where different limits for the additional, specific CO2 mitigation costs are set. At first, the operational mitigation potential of today’s combined heat and power (CHP) driven energy system is analyzed. Then we examine the possible effects of increased flexibility of this multi-energy system by adding heat pumps and heat storage. We include a detailed account of today’s operational cost structure including taxes and subsidies. To correctly represent the CO2 footprint of consumed electricity from the grid, we consider the CO2 intensity of Germany’s electricity mix as time-dependent. This is important to correctly honor the impact of the multi-energy system’s flexibility. We find that given the current regulatory environment, without considering investment costs, large CO2 reductions compared to the modeled cost-optimal operation can only be achieved for specific CO2 mitigation costs above 150\euro/t. Small reductions can be obtained at much lower cost when a heat pump operates in parallel with the CHP. However, for all scenarios the CO2 reductions can only be realized by exploiting periods with low CO2 intensity of the grid’s electricity.

ABSTRACT. In this paper, a grid expansion algorithm with the integration of new generation units in interlinked grids is developed. The method uses a modified Simulated Annealing approach and automatically performs various topological searches within a generated solution space containing options for adequate expansion measures. The solving process conducts towards generating a reinforced grid topology without congestions. A reduction of necessary grid expansion through flexibility usage is determined afterwards. A case study performed on a high voltage SimBench grid shows the feasibility of the developed expansion algorithm through the generated topologies and a reduction in expansion costs by considering flexibility.

ABSTRACT. For the design of future energy systems, simulation is a tool of utmost importance. Next to the simulation of the grid's behaviour itself, there is a need for the simulation to integrate various components in order to reach an optimized operation in terms of sustainability goals. The present paper therefore presents a simulation architecture that provides a base structure for analyzing the current simulation state, performing forecasts and optimizations during the simulation run. Application cases for control strategies in dependence to sustainability indicators are presented and discussed in terms of the benefit of using the proposed simulation architecture.

ABSTRACT. In this paper we discuss and compare a few methods combining mathematical optimization and heuristics to reduce the search space of the MIP problem when solving the Unit Commitment (UC) problem. Solving the UC problem is already today often slow and can be expected to become even more complex with the increasing number of small and scattered renewable generation units with associated energy storages. We show that by analyzing the problem before solving the full MIP we can remove (fix) a set of binary variables, leading to a speed-up between 2 and 5 times, empirically without compromising optimality.

ABSTRACT. As two effective ways to enhance power system reliability, self-healing control and microgrids have been implemented in advanced distribution systems. This paper presents a method to quantitatively evaluate the reliability improvement of a distribution system after deploying self-healing control and/or a microgrid. The proposed method is based on time-sequential Monte Carlo simulation method and further integrates the service restoration and three-phase power flow constraints to accommodate practical distribution system complexity. To reduce the computation burden in searching for optimal service restoration strategy, this paper proposes a heuristic algorithm to find a practical service restoration strategy based on the topology constraints and the three-phase power flow constraints. Case study on a four-feeder distribution system shows that the reliability of both the whole system and the critical load are greatly improved after deploying the self-healing control and microgrid.

ABSTRACT. This paper proposes a two-stage strategy to enhance the operation of active distribution networks by optimally controlling the active/reactive power of the dispatchable renewable energy-based distributed generation (RDG) to improve the power quality. The two stages of the proposed strategy are executed in two time frames: the operational planning time frame and the real-time operation time frame. The first stage is a centralized control implemented using optimal power flow via particle swarm optimization to dispatch the RDGs active/reactive power to minimize the real network loss. In the second stage, a decentralized architecture of intelligent agents is proposed using Multi-Agent System (MAS) developed in Java Agent Development Framework (JADE) to manage the power of the network to ensure that the difference between the power generated from RDGs and power consumed by loads is preserved in each zone until the optimal power flow (OPF) is implemented again in next timestep.

ABSTRACT. This paper proposes a resilience enhancement approach applied to medium- and low voltage (MV/LV) systems planning. In the mathematical model formulation, substations, distribution transformers, cables, switches, renewable energy sources (RES), and energy storage sources (ESS) are taken into account. In the resilience planning approach, RES and ESS are able to operate isolated from the network through microgrids formation. Numerical results, applied to an MV/LV distribution system, show that non-supplied energy costs are lower when techniques to increase network resilience are considered.

ABSTRACT. High penetration of renewable integration with the grid has invoked the necessity for battery energy storage applications. This has been further escalated by technology advances and descending costs. Utility-Scale Battery Energy Storage Systems (BESS) are being deployed worldwide in numerous projects due to their ability to provide grid ancillary services such as frequency regulation, flexible ramping, and black start services besides their energy storing and shifting capabilities. To ensure reliable and safe operation, identifying the bad battery cells in a Utility-Scale BESS is of profound importance. This paper proposes a comprehensible method of bad cell identification for Utility-Scale BESS by evaluating the necessary electrical and thermal properties of the cells through statistical approaches. This method only requires data that is easily available and accessible and can save a lot of time and. A detailed formulation of the analysis process was carried out and applied to data from a Utility-Scale BESS. Results demonstrated that the method can successfully identify bad cells in a BESS containing a large number of cells. This approach can offer prospective economic benefits by reducing the Operation and Maintenance (O&M) costs associated with Utility-Scale BESS.

ABSTRACT. This paper proposes a formal verification strategy of grid frequency control using reachability analysis. Reachability analysis calculates reachable sets, which are all possible evolution of system states, or output variables, given a bounded input uncertainty. Contrary to classical grid frequency control schemes that are generally tuned based on multiple simulations, reachability analysis provides a formal guarantee for the performance of the controller in one computation stage. The proposed method is applied on the IEEE 9-Bus system. The accuracy of reachable sets is validated by simulation results with randomized inputs. In addition, the paper analyses the effect of control parameters and input uncertainties on the reachable sets. Thus, the operator can verify if their tuned frequency controller could violate any mandated grid directives without performing large number of simulations

ABSTRACT. This paper presents an expansion planning study on unbalanced distribution systems with high residential electric vehicle (EV) charging and solar photovoltaic (PV) power generation. A large increase in residential EV charging may overload current distribution systems if they are not adequately prepared. PV systems, on the other hand, can assists in supplying some of the demand in the network. However, most residential PV generation and EV charging are single-phase, which could lead to an increase in voltage and power imbalance. In this paper, a data-driven method is developed for expanding distribution systems while considering the uncertainty of the electricity demand, EV charging demand, and PV power generation. A mixed-integer linear programming (MILP) model is designed for improving the distribution system by constructing distribution lines and upgrading or building substations. An algorithm that combines the Alternating Direction Method of Multipliers (ADMM) algorithm and Column-and-Constraint Generation is developed in order to solve the optimization problem efficiently. Numerical results showed the effectiveness of the proposed method in expanding distribution systems in order to handle the new demand while limiting the imbalance in the system.

ABSTRACT. Aggregators are expected to become an inevitable entity in future power system operation, playing a key role in unlocking flexibility at the edge of the grid. One of the main barriers to aggregators entering the market is the lack of appropriate models for the price elasticity of flexible demand, which can properly address time dependent uncertainty as well as non-linear and stochastic behavior of end-users in response to time varying prices. In this paper, we develop a probabilistic price elasticity model utilizing quantile regression and B-splines with penalties. The proposed model is tested using data from residential and industrial customers by assuming automation through energy management systems. Additionally, we show an application of the proposed method in quantifying the number of consumers needed to achieve a certain amount of flexibility through a set of simulation studies.

ABSTRACT. A major influence on the generated power of a photovoltaic (PV) system is the cloud cover over the system. For this purpose, this paper presents a forecast model that predicts the local cloud cover over a PV system. Based on satellite images, moving clouds are detected and their vectors are calculated using the optical flow method. The forecast model consists of an artificial neural network (ANN), which is trained with the motion vectors of clouds and can predict motion vectors for ten minutes. To determine the cloud cover, an artificial image is generated using the predicted motion vectors, which shows the future cloud positions. Over a cutout for a local forecast, the cloud cover is calculated by a thresholding method. The forecast model shows that cloud positions of known days can be predicted very accurately. The accuracy of the model decreases as expected for days that are unknown to the ANN. The predicted cloud cover is integrated into a PV forecast model. It is shown that this results in a lower deviation between the prediction and the measurement of the PV power under cloudy conditions.

ABSTRACT. We consider the problem of automated anomaly detection for building level heat load time series. An anomaly detection model must be applicable to a diverse group of buildings and provide robust results on heat load time series with low signal-to-noise ratios, several seasonalities, and significant exogenous effects. We propose to employ a probabilistic forecast combination approach based on an ensemble of deterministic forecasts in an anomaly detection scheme that classifies observed values based on their probability under a predictive distribution. We show empirically that forecast based anomaly detection provides improved accuracy when employing a forecast combination approach.

ABSTRACT. Load forecasting is becoming increasingly important for planning and operational studies of electricity networks, which feature much higher levels of interactions between supply and demand sides, resulting in much larger variations of power flows. This paper evaluates uncertainty and error in a stacked bidirectional variant of a long short-term memory (SB-LSTM) model, which is applied for a day-ahead load forecasting. First, the paper analyses importance of the correct setting of hyperparameters of SB-LSTM model. Then, four different SB-LSTM forecasting models with four different data/window lengths are used to assess the “model uncertainty”, i.e., variations in the forecasted demands due to multiple implementations of a specific SB-LSTM model on the same input data set. Afterwards, the four base SB-LSTM models are combined in a homogenous ensemble forecasting model, which dynamically integrates base learners and produces final predictions from their inputs. Finally, the fifth SB-LSTM model is built and trained using hindcasted errors of one base model to forecast its error on the test data. Input data for all SB-LSTM models are actual demands recorded in one Scottish MV substation, together with the corresponding meteorological and calendar data.

ABSTRACT. There is an increasing interest in obtaining reactive power services for the transmission system from distributed energy resources. This involves coordination between the transmission and distribution companies, and the distributed generators. This paper presents a methodology to quantify the extent of reactive power provided by distributed generators to the reactive power seen at the transmission and distribution interface. Two case studies using computer load flow simulations are presented based on a real-world network. Results showed that when the distributed generators connected to the distribution network absorb reactive power there will be a multiplier of around 110% in the reactive power drawn from the transmission system. Similarly, if the distributed generators export reactive power the multiplier is around 90%. This study shows that there is a potential for providing reactive power support from distribution networks to the transmission system at the expense of additional active power losses in the distribution system.

ABSTRACT. This paper proposes a graph neural network (GNN) based approach to locate fault spots within the power distribution systems. The GNN model is extended from graph convolutional networks (GCNs), and includes several graph processing layers and followed by several full connected layers. The graph processing layers incorporated with node and link attributes are used to map system topology, bus measurements and branch parameters into hidden node embeddings, and full connected layers are used to relevant fault locations to node embeddings. The node attributes of the graph include measured phase voltage and current measurements, and branch impedance, admittance and regulation parameters are integrated into link attributes of the graph. The fault locations are represented as output features of nodes for the graph, in which only terminal buses of faulted branch have non-zero values corresponding to faulted phases. The developed approach is applicable to both short-circuit faults and ground faults. Numerical examples are given to demonstrate the effectiveness of the proposed method.

ABSTRACT. This paper presents the initial implementation of cyberphysical testbed for Advanced Distribution Management System (ADMS). The testbed is used for cybersecurity assess-ment of smart power distribution network operation usecase ofTSO-DSO coordinated reactive power management. The test in our National Smart grid Laboratory (NSGL) demonstrates the impact of cyber security attacks on the operation of the physical power distribution network where intruders alter the voltage reference signals communicated from the control room to the On-load tap changer (OLTC) in the field. This paper highlights the use of laboratory ADMS testbeds for scientifically assessing operational benefits and cyber security vulnerabilities.

ABSTRACT. Converters connected at distribution level play an increasing role in the generation mix. Especially grid forming converters providing inertia to the system are vital to handle critical events in a converter dominated system. Thus, converters need to be properly represented in grid models for stability analysis. As opposed to detailed modeling, equivalent distribution networks allow comprehensive system stability studies without extensive knowledge of the distribution network and an impractical computational effort. Since converters are mostly connected at the distribution level, their dynamic response needs to be represented in the equivalent model of the active distribution network. Previous work elaborated that gray-box methods show satisfactory results when conventional grid following converters are considered. However, as soon as grid forming control is introduced, previous approaches fail to adequately capture the dynamic behavior. In this paper, a novel method based on voltage sensitivity clustering is proposed. Dynamic simulations are performed with a test system consisting of a transmission and distribution network with grid forming converters. The novel approach is capable of reproducing the detailed network’s dynamic behavior without neglecting the influence of grid forming converters. This qualifies the method to be used in stability studies of converter dominated systems.

ABSTRACT. With the ever-growing requirement of system stability enhancement, a comprehensive evaluation of overall system stability is coming into research focus to better understand and implement system stability enhancement solutions. This paper proposes a novel Composite Stability Index (CSI) for unified assessment of system frequency and angular stability and illustrates its application considering transient stability, small disturbance stability and frequency stability simultaneously, to assess the impacts of different load models and Demand Side Management (DSM) scenarios on either overall or individual aspects of system stability performance. The results show that the proposed CSI could clearly indicate the distance of operation points to the stability boundary, balance different and even opposite impacts on different stability aspects and quantify the impacts of different load models and DSM deployments on overall system stability performance. The results are illustrated on an equivalent model of four realistic interconnected transmission networks in a DigSilent/PowerFactory simulation environment.

ABSTRACT. This paper proposed a Mixed Binary Linear Programming (MBLP) approach to find the optimal sizing and energy management of Smart Building (SB). The considered SB is equipped by local resources such as Photovoltaic (PV) generation panels, Electric vehicles (EVs), and the Battery Energy Storage System (BESS) . Moreover, the whole SB has a single Contract Power (CP) that only is connected with the grid by an Energy Management System (EMS) such that EMS manages the power flow among external grid, local resources, apartments, and common services, for the goal of reducing the electricity bill, while satisfying SB demand. Hence, the wrong choice of CP and BESS capacity will impose unnecessary charges on the electricity bill. Therefore, finding the optimal decision of CP and BESS values has received a significant role from EMS in SB. The obtained results of this work show the efficiency of the model in which finding the optimal capacity of CP and BESS improves the electricity bill to a 34% reduction.

ABSTRACT. Nowadays, residential consumers can play a key role in providing flexibility services on grids. The present work proposes a multi-objective approach to assess the impacts of peak shaving and flexibility capacity control objectives on end user costs and equipment degradation. To identify and limit the effects of introduced services on the lifetime of equipment, aging indices are defined and included into the model as penalties. Four different sets of flexible resources are considered: (1) dishwasher (DW), (2) DW and photovoltaic (PV) panels coupled with Electric Energy Storage (ESS), (3) Heat pumps (HP) coupled with Thermal Energy Storage (TES), (4) a combination of all the previous resources. The resulting optimal control problems of each type are modeled as mixed-integer linear programming (MILP) problems. The proposed model is applied to 2000 test cases of buildings with these four types of flexible resources. Results show that the introduction of flexibility terms alongside the energy cost in the control objective increases the energy bill by less than 0.5\%, while reducing the peak demand by around 11\% and increasing the flexibility capacity by around 16.5\%. Applying proper equipment aging indices reduces the peak shaving capability, while it does not impact the flexibility capacity significantly.

ABSTRACT. With increasing the diffusion of Power Electronic Converters (PEC) in modern and smart grid systems, the academic and industry R&D researchers rely more and more on the accurate and efficient modelling and simulation of PECs for both hardware and controller design. This paper focuses on the PEC simulation model verification for (time consuming and costly) grid code compliance tests. The simulation model developed based on a full-power laboratory test bench and the waveforms of fault ride through (FRT) tests were compared. The results show that the modelling approach based on averaged inverter captures the dynamic behavior with a good accuracy. Hence, the dynamic behavior of the inverter in FRT tests can be assessed using the simulation approach saving the time and cost required by full power testing.

ABSTRACT. Industrial activities consume a large portion of the total energy demand worldwide, and thus, significantly contribute to greenhouse gas emissions. Hence, they face significant economic, social and environmental pressures to create energy efficient processes and systems of production and directly manage their energy consumption, looking at aspects beyond direct costs. One of the most effective ways to reduce energy consumption in the industrial sector is to implement an energy management system. Current research into Industrial Energy Management System (IEnMS) remains insufficient, and to the best of our knowledge, a holistic framework for an IEnMS using the IoT and big data does not exist. This paper provides a comprehensive systematic literature review of the existing academic publications on IEnMS. Further, the main requirements and components of an IEnMS are identified using literature. Based on these identified requirements and components, we have designed a theoretical framework for the IEnMS using IoT and big data analytics, forming a cyber-physical system. These results illustrate how the proposed framework provides an objective methodology that can be used to select the most suitable IEnMS for different industries based on their particular requirements.

ABSTRACT. The article is devoted to a methodology for environmental assessment of renewable energy sources (RES). The methodology allows to analyze and compare the impact of RES power installations on the environment at various stages of their life cycle, including their production, operation and disposal. The technique enables to carry out a qualitative and quantitative assessment of all possible impact factors. An example of using the developed methodology has been given. The environmental friendliness of wind, solar, biogas power plants and mini-hydroelectric power plants has been evaluated. It has been found that solar power plants, which practically do not pollute the environment during operation, turned out to be the most environmentally ineffective when assessing their life cycle. The most environmentally friendly were mini-hydroelectric power plants due to the materials from which they are made, their mass and higher efficiency.

ABSTRACT. In transition to the renewable-rich power grid, the involving system will face the challenges of loss of inertia and control paradigm change due to the retiring of synchronous machines and the fast integrating of inverter-based resources (IBRs). A fundamental and emergent question that needs to be addressed is how to achieve adequate frequency regulation by IBRs in a low inertia power system. This paper presents and compares several possible IBR control methods to provide frequency support including grid following control with frequency-real power droop, droop-based grid forming control and virtual synchronous machine (VSM)-based grid forming control. With each method, the frequency dynamic performance and transient stability of a test system are evaluated. The rate of change of frequency (RoCoF), frequency nadir and critical clearing time are selected as metrics to compare the performance and stability of the different methods. Preliminary results from the case study indicate that while the grid following control with frequency-real power droop can achieve similar frequency nadir as the grid forming methods, its RoCoF tends to be higher and transient stability margin tends to be smaller.

ABSTRACT. As part of a sustainable power system, a synergy between electric mobility and renewable energy sources (RESs) can play a crucial role on mitigating the nature of RESs and defer costly grid upgrades via smart-charging. This paper presents a distributed autonomous control architecture for electric vehicle (EV) chargers and a clustering method for charging coordination. The architecture framework is detailed depending on the number of chargers and specific location properties. Moreover, the framework unveils the communication, measurement and power flow. The aforementioned approach aims at simplifying the overall charging experience for the EV owners while coupling it with a healthy grid behavior. The proposed control architecture is simulated on a prosumer case with two EVs. The performance of the controller is considerably affected by observability capabilities of current smart-meters. Faster measurement cycles of smart meters can reduce the overshoot time span but not prevent it.

ABSTRACT. The increasing penetration of Electric Vehicles (EVs) and renewable energies into the grid necessitates tools to smooth the demand curve. To this end, this paper suggests an EV charging scheduling algorithm and a smart charging price. As EVs arrive at the charging station and leave at different times, the operator of the station applies at each EV arrival an online scheduling algorithm based on the concept of ``water filling''. The EV charging price is guaranteed at their arrival and defined as a function of the online algorithm's output, following the idea of locational marginal pricing. A numerical comparison with the offline version of the algorithm -- in which the operator knows in advance all future arrival and departure times -- shows the efficiency of the suggested online scheduling algorithm.

ABSTRACT. This paper compares different charging strategies for electric vehicles (EVs) and mechanisms to support local distribution grids. First, a general scheduling problem for EVs based on convex optimization and linearized power grid models is presented. Then, it is shown how it can be adapted to model different charging strategies. These include: i) uncoordinated charging, where EVs maximize a local utility function regardless of grid constraints; ii) Smart charging, where a charge schedule of all EVs is determined by maximizing their utility function subject to grid constraints. iii) Vehicle-to-grid (bidirectional power from the EVs is allowed); and iv) reactive power support (where chargers can provide reactive power). The performance of these strategies is investigated considering the CIGRE benchmark system for medium-voltage distribution grids. It shows that, in the proposed scenario, smart charging with reactive power support is conducive to the shortest global recharging time.

ABSTRACT. Multi-energy systems are a promising solution to solve some of the challenges that distribution systems face due to higher shares of intermittent generation and increasing demand. To accelerate research in this area, comprehensive and openly available modelling tools are required. This paper presents the extension of the open source tool Flexible Distribution Grid Demonstrator (FLEDGE) towards modelling conversion and cogeneration units in flexible thermal-electric distribution systems. The implementation of the formulated models is demonstrated in a case study where the optimal operation of a thermal-electric MES is analyzed – in unconstrained and constrained operation with the aim to quantify the benefits of coupling and coordinating thermal and electric distribution systems.

ABSTRACT. Rooftop PV systems are more likely confronted with weak grids since they are connected to low voltage networks that are originally constructed only to support the demand. It has been shown that PV systems that are connected to a high-impedance weak grid may face loss of synchronization due to instability in the phase-locked loop (PLL) unit of the PV's inverter. On this subject, whilst the PLL is designed under nominal grid conditions, its performance is highly affected when the grid condition deviates from the nominal one. One way to ensure the stable operation of the PLL is to characterize the network impedance, at the point of the connection of the PV, under different operating points of the grid, and subsequently devise a robust design for the PLL. In this paper, an approach is proposed to construct a stochastic representation of the network impedance seen at the PV's connection point. To do so, the variations of the resistance and inductance are simulated using Monte Carlo simulations and modeled via a Gaussian distribution. This distribution is subsequently employed to extract an ellipse that embraces the values of the aforementioned resistance and inductance with a given confidence level.

ABSTRACT. Forecasting cloud motion ad dynamics is crucial for many areas of study. Solar energy production depends on the cloud coverage over the area which impacts PV output by clouds limiting the incoming solar irradiance. In this paper we propose an adaptation of a nature-inspired technique called the Flocking Behavior Algorithm (or Boids Algorithm) for the problem of cloud motion forecasting for nowcasting and intra-day prediction windows. Without limiting the application of the algorithm we validate it on a sequence of satellite images and show its efficiency for short term forecasting.

ABSTRACT. Microgrids continue to gain traction as alternatives to traditional power systems in remote communities and smart cities. The green energy revolution has primarily driven this uptake amid climate change, coupled with the off-grid operability microgrids and flexibility in control. On the other hand, microgrids are inherently unpredictable, and energy security is continuously becoming critical to business continuity operation as disaster occurrences rise. Therefore, efficient energy management methods that maintain supply-demand balance have become essential to ensure the resilience of microgrids.  In this paper, the authors propose a useful energy management system (EMS) tool for microgrid operators that considers non-critical load shedding, referred to as non-critical load curtailment (NLC), and photovoltaic (PV) power curtailment (PVC). The new measures are additions to the power flow control method from the authors’ previous research based on the state of charge (SOC) of the battery energy storage system (BESS).  The measures can increase the robustness and uptime of a group of peer-to-peer prosumers. By varying the initial SOC of one prosumer, we evaluate the effect of the additional control measures on the performance of the rest of the prosumers in the network.

ABSTRACT. Power grids are evolving into smart grids in order to create a sustainable energy system. Decentralized, intermittent generation and demand management needs, among other things, make the grid more dynamic and demanding, thus posing a management challenge. Communication and information technologies are used extensively to address this challenge. Wider penetration of open, public communication networks and information technologies induces vulnerabilities into the system. Furthermore, critical infrastructure assets, such as power grids, are primary targets in cyber warfare and cyber criminality. The panelists will share experiences on reliability and security of power grids with perspectives of resilience, cybersecurity, operation and control, data-driven approaches, and mobile communication.