Motivation
The change in traffic is a politically desired change from vehicles with conventional engines to electric vehicles. The number of electric vehicles and charging stations has risen during the last years. Especially in suburban residential areas, an increase in electric mobility and private charging infrastructure can be assumed in the next years. Battery capacities and charging powers will continue to enhance in the future. Charging powers will be possible up to 22 kW (AC). For this reason, approaches are necessary in order to be able to guarantee secure grid-operation even with an increasing penetration of electromobility
Challenges in Residentials Areas and
The grid was usually designed for electrical supply, like lighting or electric cooking. In recent years, the number of PV systems in houses in suburban areas has increased. Currently, such systems are still being installed, especially in connection with storages, or existing systems are being expanded with storages. At the same time, the number of electric vehicles will increase. As a result, households will become ProSumAgers with the corresponding behavior from a grid point of view.
Investigated Scenarios
In order to obtain the greatest possible impact of electromobility and the PV storage system, an extreme grid with 145 households is used for the investigation. The selected grid has a critical string of 69 households. Synthetic load profiles are used for households, PV-systems, and EV with 11 kW charging power. The simulations are carried out with the simulation program DigSILENT Powerfactor and its Quasi-Dynamic-Simulation (QDSL) tool.
Results/Conclusion
The investigated extreme grid offers the possibility of a 18 % penetration rate of electric vehicles without measures. This low penetration is due to the already low voltage during normal operation.
One approach for greater penetration is the use of PV-storage systems in conjunction with electric vehicles. However, operation according to the current directive does not bring any improvement because the storage system cannot be used completely or possibly not at all, especially on winter days. The provision of voltage-stabilizing measures in the form of reactive power is also not possible in this case due to the behavior required by the guidelines. For this reason, extended functions are necessary.
A battery inverter with STATCOM mode can provide a remedy. Depending on the voltage, this provides reactive power, even if the storage is neither charged nor discharged. This approach enabled the penetration of the investigated grid to be increased from 26 to 38 vehicles or 26 %. Another function is the multiple-use of the storage. With the help of this function, the penetration could be increased to 65 electric vehicle or 45 % in conjunction with STATCOM mode. Thus PV-storage systems with the extended functions can make a good contribution to the grid integration of electro mobility and grid expansion can be prevented.
In the course of the German Energiewende, the increasing number of decentralized power generation and the rising penetration of electromobility leads to several challenges in German distribution grids. Violations of the permissible voltage range or capacity utilization of electrical components are significant problems, which can be prevented by an Active Network Management System (ANM). For large scale distribution grids with many actuators, the complexity of finding suitable control solutions increases for the ANM. The question arises why the structure of an ANM cannot be decentralized like its actuators. As a possible solution, Multi-Agent-Based ANM can be mentioned. In the case of critical grid states, the ANM can perform various control solutions, whereby the development of the control is made possible by the decentralised coordination of several agents. Depending on the type of the problem, different mechanisms can be used to develop agent-based control solutions. The type of the grid state violation can be divided into global and local problems. Here, one of the local problems are voltage range violations, which can be solved by a single actuator agent without the coordination with other actuator agents. The actuator agent pursues a Droop Control strategy in order to solve the problem efficiently. If several actuator agents recognize the same problem, a coordination mechanism is started between these agents. In the case of a global violation of the grid state in the course of a cable overload, a hierarchical development of the control solution is necessary. Here, a superordinate agent identifies the actual grid state for the entire distribution grid and is able to detect cable overloads. For solving the detected cable overload, the superordinate agent informs the necessary actuator agents during an agent-based negotiation mechanism. The essential of the negotiation mechanism is the fact that the superordinate agent only knows the existence of the actuator agents, but he has no knowledge of the local objectives (loading objectives) and current system state (SOC) of the actuator agents. However, the actuator agents know the actual grid state and can provide the higher-level agent with offers for solving the problem. In general, the global problem is solved by the coordination of several agents. The different coordination mechanism are tested and critically examined with regard to their grid serving use for a Distribution Grid Operator.
The focus of this contribution are the decentralized organization and determination of control strategies based on multiple distributed agents with the following issues:
In recent years, the energy transition has massively changed the requirements on distribution grids in Germany. On the one hand, many renewable energy systems, e.g. photovoltaic systems, have been installed. On the other hand, the load also rises due to the progressive electrification of the heating and mobility sectors. Both developments take place particularly in the distribution grid. In combination with digitalization, which makes it possible to control new producers and consumers, new challenges arise for the electricity grids.
To determine the future challenges, it is necessary to create simplified models of the reality, which represent the questions to be investigated with sufficient accuracy. In the field of distribution grid simulation, it is possible to model either small grid areas in great detail or to model larger grid areas in aggregated form.
In the first case, a household, a PV system and an electric vehicle are modelled for each building. In addition, different charging strategies can be selected for the electric vehicle. The resulting loads for the low voltage grid are then calculated from these individual load profiles and the load on the transformer is finally determined.
In the second case, low voltage grid areas are modeled with a residual load placed at the node of the transformer. The common practice for this type of modelling includes the utilization of standard load profiles, which are used by the grid operators for billing and energy procurement. Therefore, each household customer is described by an identical profile, which is scaled with the corresponding energy quantity per annum. This method shows an acceptable accuracy for a high number of consumers. However, it is not possible to determine the impact of new components like e-mobility or its operating behavior using standard load profiles. For this reason, this paper presents a method to consistently model both detailed grids and aggregated grids with electric vehicles in one simulation. The described method is also transmissible to heat pumps or storage systems
First detailed simulations are carried out for different operating modes of vehicles with corresponding driving profiles. From these results standard load profiles are calculated and scaled down to fit the average energy consumption of one vehicle. In the last step a certain number of electric vehicles are assigned to each aggregated grid area based on the corresponding scenario and the profile is scaled with this number. Thus different charging strategies and penetration rates can be consistently calculated for detailed and aggregated grids.
This paper describes both the procedure for determining the profiles for selected scenarios and the application of the entire methodology to a real medium-voltage grid. Individual low voltage grids are examined in detail and others are aggregated using the methods described.
This paper presents an electric vehicle (EV) smart charging scheme at residential buildings based on installed photovoltaic (PV) output and household electricity consumption. The proposed EV charging scheme is designed to determine the optimal EV charging schedules for the purpose of minimizing the load-variance or flattening the load profile. When the net-load is taken into account in the smart charging scheme, not only the peak load can be reduced, but also the PV self-consumption in the building can be increased.
The charging scheduling problem is formulated and solved with a quadratic programming approach. The departure and arrival time and the distance covered by vehicle in each trip are specifically modeled based on available statistic data from Swedish travel survey. The scheme is applied on simulated typical Swedish detached houses without electric heating. The aggregation of distributed smart charging in multiple houses is conducted and compared to the smart charging in a single house. Numerical results are presented to show the effectiveness of the proposed smart charging scheme. Positive results on both the PV self-consumption and the peak load reduction are achieved.
With a catenary length of almost 100 km, Solingen has the largest operating trolleybus system in Germany. 50 electrically driven trolleybuses which are equipped with auxiliary combustion engines and 50 additional conventional diesel buses are serving the public transport system.
The aim of the project "BOB-Solingen" - the acronym BOB denotes the German words “Batterie-Oberleitungs-Bus”[1] – is to electrify the entire public transport sector by replacing the currently trolleybuses with the novel battery-trolleybuses, which combines proven trolleybus technology with the latest battery technology. The BOB creates the next generation of trolleybuses, which is able to drive on routes with no catenary as well by means of the included battery.
Moreover, within the project charging stations for electric vehicles (EV), decentralized renewable power generators such as photovoltaic (PV) systems as well as a stationary power storage system will be directly connected to the catenary. The stationary storage will consist of obsolete trolleybus batteries to increase their cost efficiency by establishing a second-life utilization concept.
To efficiently monitor and control the entire novel DC system, an intelligent control and management of the power flow in the overall system will be introduced. The Institute of Power System Engineering at the University of Wuppertal will develop and implement the essential automation system for the DC grid to use its existing catenary infrastructure as effective as possible within its physical limitations.
In order to realize an intelligent control of the grid, the load flow of the current grid (including the trolleybuses) as well as of the future grid (including BOB) has to be modelled and simulated. By means of the simulation, critical grid situations can be detected.
This paper intends to integrate the new battery-trolleybuses by replacing conventional diesel buses. Various simulated future scenarios will show the impact of the battery-trolleybuses to the dynamic grid. In addition, the exchange of trolleybuses by battery-trolleybuses is simulated. Then the scenarios are compared and evaluated.
By using battery trolleybuses, the grid load increases significantly, since the energy, which was previously obtained from the diesel fuel, has to be obtained from the overhead line in the future. The grid was not technically designed for this heavy load. Network monitoring techniques will be the key to point out the actual grid state in order to enable the essential intelligent grid control. The aim is to identify critical network situations at an early stage and to eliminate them preventively.
The presented work in this publication is based on research activities, supported by the Federal Ministry of Transport and Digital Infrastructure, the described topics are included in the project “BOB Solingen”.
[1] Eng: battery-trolleybus
In the course of the ongoing transformation of the energy supply towards a renewable and decentralized system, the share of volatile renewable infeeders, as well as high power consumers like plug-in electric vehicles (PEVs) in the German distribution grid is steadily increasing. Challenges in that context are likely to manifest mainly through asset overloads in urban distribution grids. Since charging processes of PEVs rarely require the maximum charging power over the entire parking time, they offer a high level of flexibility, which may be used to counteract grid overloads via controlled charging. Therefore, comparatively expensive grid enhancement could be avoided. Additionally, the integration of renewable energy sources could be supported by exploiting the PEVs’ charging flexibility (i.e. to match volatile infeed). A further increase in added value may be achieved through a suitable energy procurement strategy at the spot market for energy, which should also take into account forecast deviations.
In this paper a procurement optimization strategy for a car park with 40 PEV charging stations and decentralized infeeders is presented. In that context an automated temporal shift of the PEV charging power is carried out several times in order to adapt the charging schedule to the relevant energy prices and occurring forecast deviations. The corresponding grid-related power limit is constantly adhered to.
Electric vehicles (EVs) and photovoltaics (PV) are swiftly being adopted to improve sustainability in both the transportation and the electricity sectors. Residential buildings might benefit from self-consuming the locally produced PV electricity to charge the EVs of the residents. However, the temporal mismatch between midday solar power production and late afternoon EV charging reduces the self-consumption (SC) potential. Here, we investigate the potential of battery storage in improving this SC. The batteries are intended to be used to store the PV energy, from midday, to charge the EVs during the late afternoon. Here we estimate the SC with various battery capacities. This work might be of value to grid operators interested in temporal load matching using battery storages. The results indicate that the houses benefit the most from a 5 kWh battery capacity in comparison with 10 kWh or larger. Using a 5 kWh battery, the SC and self-sufficiency (SS) of the median house without an EV improved by 40% and 14%, respectively. With EVs, the same scores improved by 38% and 11%, respectively. This indicates that the batteries were predominantly used to cover the load of the house and were rarely used to supply the load of the EVs.
The current discussions about the NOX emissions and the targets of CO2 reduction will push the transformation of the traffic sector towards e-mobility in the next years. Given the fact that one obstacle for electric vehicle (EV) acceptance still is the visibility of charging points, the expansion and distribution of the charging infrastructure at public spaces suiting the different use cases becomes an important part for the reliability related to the range. To dimension the charging infrastructure at destination charging locations (DCLs), such as supermarkets, shopping centers and cinemas, the consideration of regionalized impacts and factors depending on the particular location, such as the occupancy rate, is required. Additionally, a regionalization of an EV scenario for 2020 to 2050 in steps of five years provides the electrification rate on municipality level. These data are combined with charging profiles, which are based on real mobility statistics as well as data on charging probabilities, to determine the required EV parameters at the DCL. An optimization algorithm filters out the profiles matching best the occupancy of the DCL. By charging the EVs of the chosen profiles a total load profile for the location results, from which the number of required charging points is derived. In order to evaluate the efficiency of the simulated charging infrastructure, the distribution of the resulting states of charge is investigated. This paper carries out a case study by showing the results of the simulation for a supermarket for the year 2035 and a summary for the other years of the time scale.
In network analyses the simultaneity of the charging processes of plug-in electric vehicles (PEV) play a decisive role. It is often assumed that PEV are charged at a constant charging rate throughout the entire charging process. However, this reflects an empirically unrealistic simultaneity factor for PEV. In other studies, the simultaneity factor is based on current mobility behaviour or is limited to selected applications (e.g. PEV can only charge with a charging rate of 3.7 kW) [1], [2].
In general, network analyses should distinguish between two types of simultaneity. Since not every PEV is charged daily, there is the simultaneity factor which describes the percentage of PEV charged on the same day. On the other hand, the simultaneity of the charging processes taking place within one day must also be taken into account. With regard to the latter one, a tool is developed which will be offered as an open-source tool to download from a public website. In this paper the architecture and the required input data is described and exemplary results are presented.
The tool calculates the simultaneity of the charging processes within one day taking into account the number of EVs (divided into BEV and PHEV), arrival and departure time, vehicle class (small, medium, large), various charging rates (3.7 kW, 7.4 kW, 11 kW, 22 kW and 44 kW), battery capacity and State of Charge (SoC) at arrival and desired SoC at departure. The default data is based on different sources, see for instance [3], [4], or [5]. In addition, a special feature of this tool is the possibility to either use the default data or to change the input data individually. Thus it is possible to consider future trends in the tool. At the same time, the peak loads of EV and households can also be displayed, taking into account the EV and household specific simultaneities. Moreover, it is possible to consider the unbalanced distribution of EV due to one-, two- and three-phase charging processes. Hence, using our tool, different simultaneity factors can be easily considered when calculating the utilization of the network operation equipment (transformer and cable).
[1] Heinz, D. (2018), Erstellung und Auswertung repräsentativer Mobilitäts- und Ladeprofile für Elektrofahrzeuge in Deutschland. Working Paper Series in Production and Energy, No. 30, Oktober 2018, IIP, KIT.
[2] Rolink, J. (2013): Modellierung und Systemintegration von Elektrofahrzeugen aus Sicht der elektrischen Energieversorgung. Dissertation, Dortmund.
[3] Stöckl, G. (2014): Integration der Elektromobilität in das Energieversorgungsnetz. Dissertation, München.
[4] Probst, A. (2014): Auswirkungen von Elektromobilität auf Energieversorgungsnetze analysiert auf Basis probabilistischer Netzplanung. Dissertation, Stuttgart.
[5] Zimmer, W., M. Buchert, S. Dittrich, F. Hacker, R. Harthan, H. Hermann, W. Jenseit, P. Kasten, C. Loreck (2011): Optum: Optimierung der Umweltentlastungspotenziale von Elektrofahrzeugen, Schlussbericht.
Distribution utilities are already facing problems due to the integration of distributed generation (DG) and the foreseen large-scale deployment of electric vehicles (EVs) could aggravate the situation even further. Similarly, a large EVs share is a concern for transmission system operators as it may substantially increase the network peak power and power-flow fluctuations. The main aim of this paper is to provide a coherent probabilistic methodology for assessing the impact of EVs integration on distribution and transmission networks, including low-voltage (LV), medium-voltage (MV) and high-voltage (HV) network analysis. The simulations are carried out by means of sequential Monte-Carlo simulations, taking into account the variability of consumption and generation at distribution level and the probabilistic nature of EV charging, highly dependent on users’ habits and required comfort. This approach enables to address the high variability of power flows in power networks and can form the basis for network planning and for the development of measures to reduce system cost due to EVs integration.
In terms of distribution network modelling, unbalanced operation is assumed, allowing for the study of single-phase charging. Load and distributed generation (DG) probabilistic load-profiles are obtained from measured data and EV charging-load diagrams are constructed based on start-of-journey and travel-distance statistics. A reference MV and LV distribution network is used for simulations and the actual Slovenian transmission grid is used for HV-level simulations. The obtained results at distribution level are probability functions of transformer loading, feeder loading and network voltages. The results at transmission level are given as an increase in feeders’ peak-power and an increase of the loss-of-load expectation (LOLE).
The results show that first problems due to EV integration can be expected in LV networks, especially in rural LV networks, and will be associated with overloading of transformers as well as with high voltage drops. The results suggest that some actions in terms of controlled charging will be required in order to integrate a larger share of EVs and to avoid costly network reinforcements at the same time. These actions should result in more dispersed beginnings of charging and consequently lower power peaks. The proposed approach allows also for the analysis of advanced charging schemes and can be a useful tool in terms of network planning.
Zero emission mobility such as electromobility is a key to reduce the CO2 emissions of the transport sector. In order to increase the acceptance and demand of electromobility a reliable, publicly accessible charging infrastructure is needed. For planning issues as well as for investors the degree of utilization of charging points is important. Yet there is no consistent definition for the degree of utilization of a charging point in literature. This paper presents a set of indicators to measure the utilization of a charging point. Furthermore, these indicators are calculated with data from charging points that are founded by the Federal Program Charging Infrastructure. The analysis compares the indicators and discusses the significance and usability of these indicators. In conclusion, it is important to take several indicators into account as there is no single indicator that provides an overall impression of the utilization of a charging point.
The proliferation of low carbon technologies (LCTs) such as heat pumps and electric vehicles (EVs), which are critical to decarbonising domestic heat and transport infrastructure, will pose considerable challenges to the operation of distribution networks. As these technologies will be connected directly to the low voltage (LV) network, this is where these challenges will initially occur.
In Ireland, the pathways to achieving these objectives are now set out in Ireland's Climate Action Plan. While targets exist, the rate of adoption of these LCTs and other technologies such as domestic photovoltaic (PV) generation is uncertain. Furthermore, the impact of geographic clustering and the array of technology options for EVs and heat pumps further complicates the evaluation of the impacts on the LV distribution network.
This increased reliance of customers for electricity to support domestic heat, transport and the increased penetration levels of distributed generation on the LV network also increases the importance of this infrastructure. Domestic customers have expectations in terms of reliability and resilience of the electricity system. Therefore, it is critical that the planning approaches of Distribution System Operators (DSOs) cater for these anticipated changes are flexible and cost effective while ensuring the reliability of the system for domestic customers.
This paper presents a review of academic and industrial studies into the impacts of new domestic level LCTs on future LV distribution networks. We review of practices in a number of jurisdictions to cater for these challenges in the distribution network. Finally, conclusions are on the practices and methodologies in future LV design and development which will be required for future LV distribution networks in Ireland.
Electric vehicles (EVs), heat pumps (HPs) and the expansion of decentralized photovoltaic (PV) generation will lead to completely new load situations in power systems and may bring great challenges to electric utilities, especially at the distribution level. The methodology of modelling EVs, HPs and PVs is developed, which enables the generation of locally differentiated load scenarios for the power system simulation. Based on the dynamic load flow calculations, four scenarios are generated and simulated for a medium sized town in Germany. The results show the validation of the modelling methodology and the outlook of the distribution grid.
In the course of the German energy transition, the share of distributed energy resources is increasing significantly. This transformation poses major challenges for the planning and operation of power systems since new generation, load and flexibility patterns might induce additional grid congestions. Exploiting distributed flexibility for grid-oriented purposes has been widely discussed as an alternative to conventional grid expansion measures for congestion management purposes. Against this background the smart grid traffic light concept has been proposed as an unbundling compliant concept to procure and retrieve grid-orientated flexibility in a timely manner. This may provide a cost-efficient grid integration strategy for the assets, while also supporting their market integration. However, the existing conceptual approach provides a high-level framework definition, rather than a concrete implementation proposal. Therefore, this paper focuses first on the specification and then on the simulative assessment of the concept. As a result, the paper aims at substantiating ongoing discussions with regard to the traffic light concept with particular focus on the role of electric mobility for flexibility provision.
This work estimated the potential impact of utility initiatives as well as other variables outside of utility control to help predict adoption rates between 2020 and 2030. It differed from previous work in this area in that it focused specifically on utility incentives and was able to deliver results on a national as well as a regional basis.
Methodology
A discrete choice model of passenger vehicle demand was estimated using data collected through a survey of utility customers in the service territories of eight utilities. The estimated model of vehicle demand was then used in conjunction with the Market Acceptance of Advanced Automotive Technologies (MA3T) model developed by Oakridge National Laboratory to forecast adoption rates of electric vehicles in the service territories of the participating utilities. A total of 3,200 surveys were collected (400 in each utility service territory).
Results
Across utility service territories, the results of the study exhibit more similarities than differences. Where there were differences, they tended to be driven by demographic variation across each utility’s customer base.
An up-front purchase price or lease price discount is the most impactful driver of increased EV adoption. The efficacy of the incentive depends on the amount offered. For instance, on average we found that a $3,000 purchase price discount could increase EV sales by more than 45 percent. It is important to note that further analysis into the cost efficiency of incentives needs to be tackled.
This study’s assessment of the impacts of new incentives on EV adoption for various customer segments provides a useful basis for developing targeted EV marketing and customer outreach efforts. The impact that each incentive has varied by gender, political affiliation, prior EV experience, income, education, and location (i.e., city, town, suburb, or rural). Some of the summarized findings are listed below.
This work is an important first step in understanding how EV adoption differs amongst utility territories and how that might change given varying changes.
The transformation of the traffic sector initiated by the German federal government has an increasing influence on the structures and supply tasks in today's distribution grids. Due to the small number of electric vehicles, the effects are currently not widespread. According to the Federal Motor Transport Authority, a total of 47.1 million passenger cars were registered in Germany by 1 January 2019. Of these, 83,175 were electric vehicles - an increase of 54.4% compared to the previous year.
Today, local grid bottlenecks are already occurring, for example where a large number of charging points are bundled together, such as in car parks or modern residential areas. Intelligent charging management can ensure that as many charging points as possible are served and that the power supply to the connected apartments and commercial units is guaranteed at the same time. The solution comes from a single source and coordinated hardware and the software components are even compatible with each other because they were already coordinated in the planning phase. In such projects, the charging points in the parking areas are linked to a central control system via a network, as the distances are comparatively short. In addition, the hardware of a single manufacturer is usually used in such projects. Communication with the grid operator is not necessary as decentralized control is used.
In contrast, various components from different manufacturers are often already installed in public areas and the distribution grid is already in existence and may have to be enhanced. In addition, decentralized power generation plants can lead to internal overload situations which are not detected by the traditional network protection equipment installed in the local grid transformer (in contrast to residential areas). In addition, classical local grids often lacks spatial proximity and the present charging infrastructure is not homogenous. The preconditions for interconnecting the different charging infrastructures are standardized communication protocols and an interface to the distribution grid operator.
In this contribution, the University of Wuppertal, together with its partners DFKI, SPIE, STEAG, Voltaris and the VSE all partners of the SINTEG showcase project DESIGNETZ, shows how the combination of smart meters and decentralized distribution grid automation systems can increase visibility in the grid cost-effectively in order to enable an integrated charging infrastructure in the distribution grid. In this context, an intelligent Smart Meter Converter is presented, which provides the decentralized grid automation system with scenario-specific Smart Meter data in a usable form. An overriding objective of the paper is to show how smart meters can be applied to improve measuring and responses to the impacts of e-mobility.
Battery Electric Vehicles (BEV) are getting increasing intention when it comes to reaching the emissions targets within the transportation sector. In principle, BEV offer the flexibility to be charged in hours with low CO2 emission. The total amount of such flexibility for the overall energy system is determined in an agent-based simulation and applied within a fully sector-coupling optimization model (electricity, heat, transport). This allows to examine the effects from utilizing this flexibility on reducing the carbon footprint of charging. The flexible share of the BEV charging load is found to be correlated positively with the charging power and reveals that at least 77 % of the charging load can be shifted in time when the charging power peaks. It is shown, that charging processes during working hours can mostly be delayed for 4-9 hours whereas the major share of load while the vehicle is at home has possible delay times of more than 10 hours. While harnessing this flexibility, carbon-optimized charging can further reduce the carbon footprint of BEV by a maximum of 9.1 g CO2/km which equals 24 % in 2025.
In Japan, there is huge installed capacity of the distributed rooftop photovoltaic generations in addition to the large-scale solar power plants. It is important in normal condition of the power system that household consumption of the rooftop photovoltaic generations by use of HEMS (Home Energy Management System) and flexible energy devices including distributed energy storage because FIT(Feed-In-Tariff) price for the rooftop photovoltaics is to be reduced from 2019 in Japan. Grid independent operation would be required for preparing natural disaster and/or black-out. If all the houses equip the rooftop photovoltaics, voltage rise problem is issued. The frequency fluctuations are significant in case that regional share of the rooftop and large-scale photovoltaics is large.
Many electric vehicles will be interconnected with the household parking spot in the near future. There is huge potential of electric vehicle as the distributed energy storage because the vehicle is sleeping at the household in most of the day, and he/she have very large capacity battery, 24/40/62kWh, Nissan Leaf in the laboratory. The electric vehicles connected with the HEMS can contribute self consumption of rooftop photovoltaic generation, emergency power supply, and the grid-level and distribution-level control, and so on. The authors have proposed a smart inverter control for the electric vehicle, in which active and reactive power is managed as conditions of the power system frequency deviation and the voltage profiles in the distribution feeder. Performance of the proposed smart inverter control was evaluated by the campus HILS (Hardware-In-he-Loop Simulation) facility.
In this paper, an integrated HEMS strategy is proposed, and is implemented to the HILS facility including three electric vehicle fleet and two Vehicle-to-Grid capable chargers in Tokyo City University campus. The proposed HEMS strategy includes following control modes ;
1) Smart charging of the electric vehicles from the rooftop photovoltaic generation
2) Vehicle-to-Building to the university campus based on renewable charged energy
3) Emergency power supply and intentional microgrid from the electric vehicles to the laboratory
4) Flexible active and reactive power control as conditions of the power system and the distribution feeder
5) Auxiliary and supplemental control by remote control signal
i.e. grid ancillary service, neighborhood HEMS, electric vehicle aggregator
System structure in terms of electrical and communication among the HEMS, sensors, meters, photovoltaic generations, electric vehicles, and chargers is described in the paper. Demonstrating the HEMS strategies under some scenarios of vehicle usage and power system conditions, functionality and feasibility of the proposed schemes and implemented system is verified.
The quality and demand coverage of urban public charging networks depend on a high number of on-site specifics comprising e.g. mobility behaviour, socio-demographics as well as technology and economics related parameters. This study presents a model that develops integrated expansion strategies for public charging infrastructure in urban areas taking into account the mentioned factors. First, the model quantifies the demand of normal and fast charging infrastructure. Second, it optimizes the placement of the charging points in order to cover the charging demand as best as possible. Third, the model investigates to what extent the local power distribution grid is prepared for a considerable increase in the share of electric mobility. Fourth, it examines if the power grid restricts the identified installation sites and whether controlled charging is able to relieve critical grid situations.
The model is applied to the Pfaff area in the German city of Kaiserslautern. The exemplary district and its specifics are introduced. The resulting charging infrastructure demand and placement within the area are presented and discussed for an assumptive share of 30\% of electric mobility. It shows that charging the electric vehicles doesn't cause critical grid situations even without controlled charging.
However, because of on-site specifics and suitable charging concepts being highly individual, solutions need to be developed case by case. It can be concluded that the necessary charging infrastructure expansion leads to multiple challenges for cities. The presented method addresses these challenges by developing integrated charging infrastructure expansion strategies in order to support stakeholders such as municipalities, distribution network operators and charge point operators in their planning and decision making processes.
The integration of charging infrastructure into distribution grids is a major challenge of the mobility transition to electric vehicles (EVs), since the resulting increased power flow can lead to grid congestions. Grid integration studies investigating these effects require realistic and feasible methods for determining possible worst-case grid situations. Especially precise estimations of the maximal simultaneous power flow due to vehicle charging and of its spatial distribution in the grid, are crucial for analysing congestions and necessary reinforcement measures.
On last year's E-Mobility Integration Symposium we demonstrated a comprehensive approach for conducting such studies including probabilistic modelling of vehicle charging, an analysis of simultaneity factor approaches as well as automated grid reinforcement.
In this paper, we present further research, that puts a greater focus on modelling and identification of realistic worst-case grid situations caused by EV, charging in real low-voltage grid models. The proposed probabilistic method uses a pool of EV charging profiles and repeatedly places them randomly in a grid. The resulting distribution of different grid states contains all information that is necessary to derive realistic worst-case situations as well as the probabilities of their occurrence. Additionally we now include domestic load profiles in our probabilistic modelling approach in order to achieve an improved consideration of simultaneity effects of these load types. Furthermore, instead of assuming a predefined point in time, when the maximal simultaneous power flow occurs, we investigate a time window.
In last year's paper we already showed, that simultaneity factors are not well suited for modelling smaller numbers of charging vehicles - for example in low-voltage feeders. Now we extend this analysis to a comparison of simultaneity factors vs. load profiles for domestic loads when combined with EV charging profiles.
Finally, since probabilistic methods require a relatively high amount of computational resources, we present possible improvements for simultaneity factor approaches like feeder specific simultaneity factors or machine learning approaches. These methods aim at a good balance between accuracy in estimating worst-case grid situations and practical feasibility.
Analyzing realistic EV-grid integration (EVGI) with available simulation tools is cumbersome due to the software overhead associated with offline simulation. Alternatively, real-time hardware platforms are becoming convenient means for testing and evaluating systems before field implementation. This study presents a digital implementation of an EVGI model in real-time on a multi-core processor based simulation platform. Furthermore, an Interned-inspired EV charging control algorithm is proposed in a decentralized fashion to prevent congestion related problems in a residential distribution grid. The impact of the proposed EV charging control on the IEEE 37-node test system is evaluated through the real-time analysis. The developed controller results show promise for extension to any utility-interfaced power electronics system. Real-time simulation implementation requirements and challenges in the context of EVGI are also discussed.
This contribution presents preliminary results of a research project that tackles the issue of electrifying the heating and transport sectors in urban distribution power systems. The paper starts with presenting the concept of a load forecast demand model that considers the different trends and developments affecting cities, in particular the integration of electrified transportation sector (eMobility) and the heating sector (called as new loads). The paper then continues with an explanation of the work to be done within the scope of the project. Germany-wide scenarios for new loads are forecasted in the years 2030, 2040 and 2050 and afterwards regionalised up-to street level. In parallel, a selection of representative power networks is reached after performing a clustering analysis for the collected Low Voltage (LV) and Medium Voltage (MV) networks. After that within the scope of the project, several remedy measures are considered including conventional and innovative technologies. These remedy measures are then compared to each other on an economic basis. Finally, the project concludes new guidelines for the planning and operation of urban distribution power systems in order to prepare it for the future transformation of loads in cities.
Electric mobility leads to an increasing challenge for power grid operators, particularly due to its high peak power demand in low voltage grids in the scenario of home charging. Power grid enhancements are considered either as cost-intensive or as environmentally unfriendly and, hence, more intelligent ICT-based solutions are needed for economic and ecological reasons. Therefore, our intention is to develop a practical approach of grid-friendly smart electric vehicle charging methods. The approach entails two methods, namely: (i) Proactive electric vehicle charging control via prediction of available charging capacity and a corresponding intelligent scheduling of charging processes; (ii) Reactive, decentralized charging process control as a response to critical grid situations. Proactive forecasting of available power capacity and energy from (distributed) renewable sources can lead to a better utilization of the power grid in place and extend the usage of renewable energy, which is required for a successful turnaround in energy policy. A reactive control of ongoing charging processes guarantees that the power grid infrastructure can run at its limits, while not overshooting power quality limits. This bipartite concept exploits the flexible potential of the power supply network and at the same time optimizes the ongoing charging processes to meet the requirements of the grid.
Australia is a global pathfinder in terms of rapid renewable energy deployment. It is on track to reach 50% renewable electricity in 2024 and 100% in 2030 if the current deployment rate is sustained. The electrification of transport and low temperature heating provides a significant potential for large emission reduction in Australia if this change is coupled with the decarbonisation of grid electricity. However, additional electric load will also affect the energy balance as well as the energy system costs, especially in a 100% renewable system in which energy generation relies largely on intermittent and non-dispatchable energy sources. In this paper, we model the electric load profiles from 100% electrified land transport fleet and electrified low temperature heating in Australia under various scenarios. The 77TWh additional electricity demand could be incorporated into a 100% renewable electricity system if the flexibility in the loads is managed wisely. The entire electric load could be supplied by solar photovoltaics and wind, and balanced by off-river pumped hydro energy storage, high voltage transmission and demand management at low costs.
The rapidly growing market for battery electric vehicles (xEVs) enters a new stage in 2019 when for the first-time cars with a charging power of over 300 kW will be made available. A comprehensive HPC charging infrastructure is necessary which poses challenges to the operators of charging parks in the form of high grid connection capacities which are connected to high investment and operational costs. To avoid that, battery storage systems (BSS) as well as decentralized generation plants such as photovoltaic (PV) plants can be used to relieve the grid connection by partially covering the load of the charging park. OEMs of charging hardware reacted to that demand by offering charging stations that are combined with integrated battery storage systems or with PV systems as a roofing to generate energy at low costs. To what extent those solutions can help to reduce the overall costs of a charging park needs to be examined.
Therefore, two mathematical models were developed with the objective to first determine the overall load of a charging park for any charging park configuration and subsequently cost-optimizing the dimensions and operation strategy of all charging park components (including alternative marketing options for the BSS and PV plant).
The load simulation is based on measured charging profiles of real xEVs.The probability of arriving vehicles is derived from automated vehicle counting on German highways and main roads. With the developed model any configuration of charging points and charging power can be simulated. The probabilistic model chronologically simulates the arriving vehicles and distributes them to free charging points if available.
The resulting load curve serves as an input for the optimization model, in which the dimensions and the operation of the charging park components are cost-optimized. The considered components include a grid connection, BSS and PV plant which are optimized regarding the installation and maintenance costs as well as operational costs (energy costs, grid connection fees etc.). To generate additional revenues, PV feed-in tariffs and the provision of reserve power to the grid are considered. The target function of the mixed-integer linear optimization model aims at providing the required charging power while minimizing all caused costs and maximizing possible revenues. A large number of constraints ensures that the technical limitations of the charging park’s components are not exceeded (charging/discharging of the BSS, usage of the PV generation etc.). The final result of the optimization model includes the dimensions of the components, as well as the overall capital value including the costs and revenues of the charging park during the observation period.
With the help of the model the economic and technical potential of current charging solutions can be analyzed as well as it helps charging park operators to optimize their charging park design to local conditions.
The growing penetration of electric vehicles has an impact on the development of charging infrastructures due to the resulting increase in load demand. The driving behaviour of the population, the visit characteristics for different types of locations, and of course the share of electric vehicles in an area basically determine the charging profile at site-specific nodes. This paper presents a bottom-up approach to calculate the temporal and local charging capacity of electric vehicles under consideration of a mobility model and visiting hours of different types of locations (e.g. home, work, shopping, leisure, etc.) within an electrical network.
The mobility model in this work is a stochastic model which represents parking space situations with a corresponding charging infrastructure of different locations. These locations are described by two set of parameters: First, each studied location is specified by the parking capacity, the opening hours, duration of stay of the vehicles (length of visit), weekday, and the electrical data of the installed charging infrastructure. Second, statistics of the mobility model are taken from public available studies which have analysed the driving distances to reach the different types of location. The required charging energy depends on the specified vehicle types, which defines the battery capacities and consumptions.
The subsequent allocation of different distances and durations of stay form a realistic representation of a parking space within a day. The driven distance and the duration of stay are estimated from a normal distribution for each location. Under the assumption that the driving behaviour with electric vehicles is the same driving behaviour as with combustion vehicles, the required energy and charging time is calculated for all parking electric vehicles based on the state of charge at arrival. By means of the generated load profiles of all locations, the simultaneity and accumulated power of the installed charging infrastructure can be specified and analyzed in a further step. In addition, the energy to be provided at peak load times will be taken from the daily energy balance and can later be used as a tool for system operators or investors for studies on the development of charging management systems.
The purchase period of the surplus electricity of residential PV systems under feed-in tariff law is ten years, and the number of residential PV systems those purchase period expire will increase gradually from November 2019 in Japan. Therefore, prosumers may start shifting to a self-consuming lifestyle and using storage batteries for effective use of PV.
On the other hands, electric vehicles (EVs) are spreading worldwide due to rising environmental awareness. Some residents may use EV battery as home power supply and thus EVs contribute to self-consumption of PV. Such use is called Vehicle-to-Home (V2H). The majority of power source in Japan is the thermal power, so EVs are not particularly environmentally friendly in terms of Well-to-Wheel in case of charged by the Japanese grid.
In this paper, we examined storage battery operation that aims to improve the self-consumption rate of PV energy, and consider the effect of V2H and aggregators (AGs). AGs bring multiple prosumers together and adjust their supply-demand balance. In case operating prosumers as a group, all prosumers can share PV energy. Moreover, difference in residents' lifestyles will create an opportunity for EVs to be charged by PV energy.
In our simulation, each house has residential PV system, demand load, stationary battery and EV. The prosumers first consume their own PV power by home appliance. They use preferentially the power generated by residential PV for the load, and then consume it by operating stationary battery or EV. Next, if a prosumer generates surplus power, the other consumers in the same group effectively consume their surplus power.
In addition, EVs are used according to the given driving pattern. Six drive patterns in this study are shown in the order of pattern name, driving time, driving distance, constraint time of battery charging.
Weekend pattern
-Holiday leisure long distance type: Sat.9am-Sun.9pm, 150km, Sat.12am-9am
-Holiday leisure short distance type: Sat.10am-Sun. 8pm, 50km, Sat.7am-10am
Four-day-a-week pattern
-Active driver type: 10am-5pm, 50km, 7am-10am
-Suburbs driver type: 1pm-5pm, 5km, 12pm-1pm
Weekday pattern
-Long distance commuting type: 7am-7pm, 50km, 4am-7am
-Short distance commuting type: 8am-6pm, 15km, 7am-8am
We randomly assigned these driving patterns to all consumers. In the time zone just before the driving start time, the EV battery compiles with the constraints to reach the required minimum charge rate. EVs are charged with as much PV power as possible to improve its environmental performance.
We evaluated the effect of the operation in the consumer group by the consumption rate of PV energy and the emission of CO2. As a result, the consumption rate of PV energy was improved by 1.15 times when multiple prosumers were operated as a group than when they were operated individually. Furthermore, CO2 emissions decreased by more than half.
This work reviews the 2010 to 2019 electric vehicle and related charging infrastructure market in the United States of America. It also provides a look at emerging trends with DC fast charging.
International CO2 reduction commitments are pushing increasing penetration levels of renewable energy sources and electrification of the transports sector. The expected growth for electric vehicles (EVs) will surely have a tremendous impact on the electricity distribution system. Despite the challenges ahead, EVs are able to make bidirectional energy transactions in what is designated as a Vehicle-to-grid (V2G) framework. Thus, EVs, when aggregated, have the potential to offer system management opportunities to the power grid, more specifically as providers of grid ancillary services. These services are considerably time critical and, therefore, is of utmost importance to understand the V2G characteristics and performance, particularly when remotely controlled by Aggregators. Currently, only DC charging through CHAdeMO and CCS/Combo standards enables V2G. This work presents an overview of V2G challenges and opportunities while introducing the relevance of a robust technical characterization. A set of tests focusing on V2G system accuracy, efficiency and response time were performed on remote operation and real environment. The presented results complement and support the current literature and contributes for the comprehension of V2G systems capabilities and suitability to provide flexibility grid services.
TBA
Electric Vehicles (EVs) are becoming increasingly common on UK roads. The growth in EV ownership could cause challenges for the UK electricity industry if the adoption of electrified transport is widespread, especially if groups of neighbours buy EVs creating localised clusters. These clusters could create issues on distribution networks – the networks that follow on from the National Grid transmission network and supply
homes and businesses with electricity.
Previous research by the My Electric Avenue project suggests that the impact of EV charging on LV networks may result in at least 30% of
these networks requiring upgrades by 2050. This would represent a present-day cost of billions of pounds and inevitably create disruption, affecting all of us.
The Electric Nation Smart Charging Trial recruited 673 EV owners, owning more than 40 different makes and models of plug-in vehicles, including plug-in hybrids and battery electric vehicles, to experience Smart Charging (demand management). Over two years, trial participants experienced periods of no management, and management without and then with apps to enable them to interact with the Smart Charging systems. The trial concluded with participants being financially incentivised to change their charging behaviour, producing clear indications that this could be a successful strategy for addressing distribution network congestion issues that could be created by EV charging at home.
Key conclusions are:
Within the mobility integration symposium EA Technology and GreenFlux will present the Electric Nation Project in a dedicated session with a special focus on:
More information, updates and the final report can be found at: https://www.westernpower.co.uk/innovation/projects/electric-nation